The neurotrophic factor artemin and its receptor GFRα3 mediate migraine-like pain via the ion channel TRPM8

Abstract

Background

Migraine has a strong genetic foundation, including both monogenic and polygenic types. The former are rare, with most migraine considered polygenic, supported by genome-wide association studies (GWAS) identifying numerous genetic variants linked with migraine risk. Surprisingly, some of the most common mutations are associated with transient receptor potential melastatin 8 (TRPM8), a non-selective cation channel that is the primary sensor of cold temperatures in cutaneous primary afferents of the somatosensory system. However, it is unlikely that the temperature sensitivity of TRPM8 is relevant in migraine-related tissues, such as the meninges, suggesting other activation mechanisms underly its role in migraine pathogenesis. Thus, to define the basis of the channel's involvement, we reasoned that cellular processes that increase cold sensitivity in the skin, such as the neurotrophic factor artemin, via its receptor glial cell-line derived neurotrophic factor family receptor alpha-3 (GFRα3), also mediate TRPM8-associated migraine-like pain in the meninges.

Methods

To investigate the role of artemin and GFRα3 in preclinical rodent migraine models, we infused nitroglycerin acutely and chronically, and measured changes in periorbital and hind paw mechanical sensitivity in male and female mice lacking GFRα3, after neutralization of free artemin with specific monoclonal antibodies, or by systemic treatment with a TRPM8-specific antagonist. Further, in mice lacking GFRα3 we tested the effects of supradural infusions of a mix of inflammatory mediators, as well as tested if dura stimulation with artemin or a TRPM8-specific agonist induce migraine-related pain in mice.

Results

We find that mechanical allodynia induced by systemic nitroglycerin, or supradural infusion of inflammatory mediators, involves GFRα3. In addition, neutralization of circulating artemin reduces the nitroglycerin phenotype, demonstrating the importance of this neurotrophic pathway in headaches. Further, we show TRPM8 expression in the meninges, and that direct supradural infusion of either a TRPM8-specific agonist or artemin itself produces mechanical allodynia, with the latter dependent on TRPM8 and ameliorated by concurrent treatment with sumatriptan.

Conclusions

These results indicate that neuroinflammatory events in the meninges can produce migraine-like pain in mice via artemin and GFRα3, likely acting upstream of TRPM8, providing a novel pathway that may contribute to headaches or migraine pathogenesis.

Graphical abstract

This is a visual representation of the abstract.

Keywords

artemin GFRα3 headache migraine trigeminal TRPM

Introduction

Migraine is a strongly heritable neurovascular disorder, yet little is known of the underlying genetic basis for common migraine (1,2). To this end, multiple genome-wide association studies (GWAS) have searched for single-nucleotide polymorphisms (SNPs) linked to either protection from migraine or associated with increased migraine susceptibility (3 –7). Surprisingly, some of the most common mutations reside in or near the genetic locus of the menthol receptor, transient receptor potential melastatin 8 (trpm8) (3 –8), a cold-gated ion channel highly expressed in trigeminal ganglia (TG) (9,10). Although its role in cold somatosensation and nociception in the skin is well-studied (11), how TRPM8 channels might contribute to headaches or migraine pathogenesis remains unclear (12).

The SNPs identified in trpm8 reside in non-coding regions, suggesting they do not alter function but may influence channel expression (5,7,13 –20). For example, the reduced migraine risk allele rs10166942[C] is located 5′ of the start codon (3,5) and homozygous individuals express less TRPM8 and are less sensitive to cold than those with a T at this position (21). Intriguingly, this T allele, which is prevalent in individuals from higher latitudes and proposed to arise from positive selection for a thermoregulatory advantage in colder climates (22), is associated with increased migraine predisposition (23,24). Thus, increased TRPM8 expression may correlate with increased migraine risk. What is also intriguing is that cold triggers headaches in humans (headache-induced by cold stimuli), suggesting a link between headaches and TRPM8 activity (25).

Given its relative novelty as a migraine susceptibility gene, only a few preclinical studies have examined TRPM8 function related to migraine (26 –30). Previously, using both mouse genetics and TRPM8-specific pharmacology, we showed that TRPM8 is required for both chronic and acute migraine-like phenotypes induced by systemic application of either nitroglycerin (NTG) or calcitonin gene-related peptide (CGRP) (31). However, the molecular mechanisms whereby TRPM8 channels and cells mediate migraine-like pain are not fully understood. To investigate this, we reasoned that processes underlying TRPM8-mediated cold allodynia and hyperalgesia in the skin could also facilitate the channel's sensitization in migraine-relevant tissues. Specifically, the glial cell-line derived neurotrophic factor (GDNF) artemin and its receptor GDNF family receptor alpha-3 (GFRα3) are required for TRPM8-dependent cold allodynia and hyperalgesia in multiple pain models (32 –36). Further, artemin expression is present in meningeal vascular smooth muscle and GFRα3 expressing nerve fibers are found throughout the dura mater (37,38). Expression of both meningeal GFRα3 and vascular artemin in the rat increases after NTG treatment (38) and artemin is proposed to increase expression of the inducible form of nitric oxide synthase in trigeminal neurons (39). Thus, there is circumstantial evidence for a role of artemin/GFRα3 signaling in the meninges that may correlate with migraine pathogenesis.

Here, we report that, similar to TRPM8 (31), GFRα3 is required for NTG-induced acute and chronic tactile allodynia, a phenotype that is also attenuated by neutralization of artemin. Further, GFRα3 contributes to allodynia produced by a supradural infusion of a mixture of inflammatory mediators. Lastly, application of either a TRPM8 agonist or artemin directly onto the mouse dura matter leads to a migraine-like phenotype that is TRPM8-dependent, as well as sensitive to the abortive migraine agent sumatriptan. Our findings suggest a conserved mechanism underlies both TRPM8-dependent cold allodynia and migraine-like pain, highlighting a novel molecular pathway that may, in part, contribute to migraines and headaches.

Methods

Animals

All experiments were approved by University of Southern California Institutional Animal Care and Use Committee and in compliance with ARRIVE guidelines for Animal Research: Reporting of In Vivo Experiments (40). Mice aged from 8 weeks to 16 weeks were used in all experiments and were housed under a 12:12-hour light/dark photocycle in a temperature-controlled environment with ad libitum access to food and water. All experiments were performed during the light cycle from 8:00 am to 5:00 pm. Trpm8-null mice (Trpm8^−/−, Strain # 008198) (41) were purchased from Jackson Laboratories (Ben Harbor, ME, USA) and were in the C57Bl/6 background. Wild-type C57Bl/6 and Trpm8^GFP (10) mice were maintained in-house, as were Gfrα3-null (Gfrα3^−/−) mice (42) that were bred from Gfrα3^+/− animals to obtain Gfrα3-nulls and wild-type littermate controls, as described previously (34).

Supradural injections

Supradural injections were performed as previously described (43). Briefly, mice were anesthetized with 3–4% isoflurane and then maintained by 1.5–2% isoflurane through a nose cone. The various agents used were injected in a volume of 5 µl using an internal cannula (part #C313IS-5/SPC, Internal Canula, 28 gauge; Protech International, Inc., San Antonio, TX, USA) connected to plastic tubing (Cat#N0695476, 2-stop, inner diameter 0.19 mm; PerkinElmer, Waltham, MA, USA) and a 10 µl glass syringe (Cat# 80300; Hamilton, Reno, NV, USA). The inner projection of the cannula was adjusted to be 0.6 mm to inject through the soft tissue at the intersection of lambdoidal and sagittal sutures but not puncture the dura. Post-mortem analysis was carried out to confirm that the dura remained intact after each injection.

Retrograde labeling of trigeminal neurons innervating mouse dura

Trpm8^GFP mice were anesthetized and 5 μl of supradural injections of the retrograde tracer wheat germ agglutinin (1 mg/ml, WGA-Alexa-350; Cat#W11263, Thermo Fisher Scientific, Waltham, MA, USA) were performed as described previously (43). After recovery from anesthesia, mice were housed in their cages for 14 days to allow for the transport of the tracers to the dural afferent somata in the TG.

Tissue preparation and immunohistochemistry

TG were dissected from Trpm8^GFP mice injected with WGA tracer as described previously (10). Briefly, the mice were euthanized and perfused transcardially with phosphate-buffered saline (PBS) followed by fixative (cold 4% paraformaldehyde solution). Tissues were carefully removed and post-fixed in the same fixative for another two hours at 4°C. After cryoprotection in 30% sucrose solution for at least 24 h, the tissues were fast-frozen in OCT and sectioned with a cryostat at 16 µm. The tissue sections were mounted on Superfrost Plus slides (Thermo Fisher Scientific) and kept at −80°C. On the day of immunostaining, frozen slides were dried at 25°C for 30 minutes, washed with deionized water and PBS three times, and permeabilized in 0.1% Triton + PBS (PBST) for 30 minutes. The mounted sections were blocked in 0.1% PBST solution with 20% normal goat serum (NGS) for one hour. The slides were incubated at 4°C overnight with primary antibodies in 0.1% PBST solution with 0.5% bovine serum albumin (BSA) + 0.5% NGS (chicken anti-GFP antibody, 1:500, Cat# GFP-1020, Aves Lab, Davis, CA, USA; rabbit anti-CGRP antibody, dilution 1:1000, Cat# C8198, Thermo Fisher Scientific). The slides were washed in incubation buffer for 6 × 5 minutes and incubated in secondary antibodies for another two hours at room temperature (goat anti-chicken-IgY antibody, Alexa-488, dilution 1:2000, Cat# A11039, Thermo Fisher Scientific; goat anti-Rabbit-IgG antibody, CF594, dilution 1:2000, Cat# SAB4600234, Sigma-Aldrich). The slides were washed in incubation buffer for 6 × 5 minutes followed by 2 × 5 minutes in PBS and covered by media and coverslips. The slides were dried and maintained at 4°C overnight before imaging.

The meninges (including dura mater) were dissected from Trpm8^GFP mice using a protocol adapted from previous studies (44). Briefly, mice were euthanized and perfused transcardially with ice-cold PBS. Skulls with meninges attached were removed and fixed in 2% paraformaldehyde solution at 4°C overnight. The meninges were carefully detached from skulls in PBS solution the next day and transferred to a 24-well plate. The tissues were blocked for one hour in 0.1% PBST solution with 2% NGS + 1% BSA and then incubated overnight at 4°C in 0.1% PBST solution with 1% BSA and primary antibodies. After a 3 × 5 minutes wash with PBS, the tissues were incubated in 0.1% PBST solution with 1% BSA and secondary antibodies at room temperature for two hours. The tissues, after a 3 × 5 minutes wash with PBS, were flattened and dried in the dark on Superfrost slides for 10–15 min, covered with appropriate media and coverslips and imaged immediately.

Image acquisition and analysis

Digital images were acquired using an Axio Imager.M2 (Zeiss, Oberkochen, Germany) with an Apotome attachment and Zen Blue software. Quantification of overlap between expression of neuronal markers, retrograde tracers and GFP was obtained per field and expressed as percentage overlap with the mean ± SE of the mean between fields.

Pharmacological agents and migraine-like pain models

NTG was purchased from VWR (West Chester, PA, USA) (Cat# 101095-632) and supplied as 1% solution in propylene glycol (PG) and further diluted to the desired working concentration (1 mg/kg or 10 mg/kg) using a 0.9% saline solution. NTG or vehicle was administered intraperitoneally (i.p.) every other day for 9 days, as described previously (31,45), with mechanical sensitivity measured before and two hours after each injection. Icilin was purchased from Tocris Bioscience (Bristol, UK) (Cat# 1531), dissolved, and diluted to 1 nmol/5 µl in PG and injected supradurally as described above. 1-phenylethyl-4-(benzyloxy)-3-methoxybenzyl(2-aminoethyl)carbamate (PBMCs) were purchased from Focus Biomolecules (Plymouth Meeting, PA, USA) (Cat# 10-1413), dissolved, and diluted to working concentration (4 mg/kg) using 10% dimethylsulfoxide in a 0.9% saline solution and injected i.p. as described (32,46). Artemin was purchased from R&D Systems (Minneapolis, MN, USA) (Cat# 1085-AR), dissolved, and diluted to 1 pg/5 µl using Hank's balanced salt solution (HBSS: 140 mm NaCl, 5 mm KCl, 1 mm CaCl₂, 0.4 mm MgSO₄, 0.5 mm MgCl₂, 0.3 mm Na₂HPO₄, 0.4 mm KH₂PO₄, 6 mm d-glucose, 4 mm NaHCO₃) and injected supradurally. The inflammatory soup (IS) components were purchased from Sigma-Aldrich (St Louis, MO, USA), consisting of 1 mm bradykinin (Cat# 05-23-0500), 1 mm 5-HT (Cat# H9523), 1 mm histamine (Cat# 59964) and 200 µm PGE₂ (Cat# P5640) at pH 5.0 in Synthetic Interstitial Fluid (SIF: 10 mm HEPES, 5 mm KCl, 135 mm NaCl, 1 mm MgCl₂, 2 mm CaCl₂ and 10 mm glucose). Vehicle solutions were used as control injections in experiments accordingly. The IS or vehicle was injected once supradurally and mechanical sensitivity was measured from one to six hours post-injection, then again 24 hours later. The anti-artemin monoclonal and control IgG Isotype monoclonal antibodies were purchased from R&D Systems (Artemin: MAB1085, Control: MAB002) and reconstituted in sterile PBS solution. The antibodies were injected i.p. at a concentration of 10 mg/kg. Sumatriptan was purchased from Sigma-Aldrich (Cat# S1198), diluted using 0.9% saline solution and injected subcutaneously (s.c.) at a concentration of 0.6 mg/kg.

von Frey assay

The von Frey assay was used to assess the sensitivity to mechanical stimulation as previously described (31,43). Mechanical sensitivity (50% withdrawal threshold) was determined at the timepoints indicated with calibrated von Frey monofilaments (Semmes–Weinstein) using the up-and-down method. Briefly, the mice tested were conditioned by handling for 5 minutes and habituated to plexiglass chambers on metal mesh (hind paw) or horizontal paper cups (periorbital assay) for three consecutive days before experiments. The filaments (for hind paw assay: starting from 0.4 g filament; flexible force range 0.008–2 g; for periorbital assay: starting from 0.07 g filament, flexible force range 0.008–0.6 g) were applied to the plantar surface of the hind paw or the center of the face between eyes for up to five or six seconds until the filament bent or positive response was recorded. The next heaviest (if no response) or lightest (if a response) filament was used in the next measurement and four additional measurements were performed. The 50% withdrawal threshold was calculated using the SUDO method previously described (47). The mice were returned to their home cage after measurements and retested on the next treatment/test day.

Statistical analysis

The experimenters were blinded to either genotype of the mice receiving treatments or substances injected. Sample sizes were determined via power analysis (G*Power, version 3.1.9.7; http://www.gpower.hhu.de) and based on our previous experience, assuming power ≥80%, α = 0.05. Unless otherwise specified, mice in each group were equally balanced between male and female mice with no sex differences observed in the data. Further, the estrous phase of female mice tested was not measured as there are no reported differences in von Frey testing because of the estrous cycle (48). All data are presented as the mean ± SEM with the statistical assay used described in the figure legend or associated text, along with sample numbers. Time course measurements and comparisons between different injection conditions such as treatment or genotype were determined via a two-way analysis of variance (ANOVA) with either Bonferroni's, Dunnett's or Tukey's multiple comparison tests. Two-tailed paired t-tests were used to compare pre- versus post-injection conditions with the same injection conditions or genotype.

Results

TRPM8-positive afferents are present in the mouse dura

To further investigate the role of TRPM8 in migraine-like pain, we confirmed that TRPM8-positive afferents are present in the mouse meninges, first using retrograde tracer back-labeling (43). Naïve adult Trpm8^GFP mice (10) received a supradural injection of the retrograde tracer wheat germ agglutinin (1%; WGA-Alexa-dye) and TG were processed for immunolabeling 14 days post-injection as described (10,34,35,49). We found that 15.7 ± 2.7% of WGA-labeled cells were GFP positive (n = 4 mice; 805 WGA-positive cells) (Figure 1a), consistent with the proportion of TG neurons per ganglia known to express TRPM8 (9,10,34,50,51). Next, we performed immunolabeling of dura dissected from adult Trpm8^GFP mice as described previously (10,34,52), observing GFP expression in Trpm8^GFP animals (Figure 1b), labeling that was absent in wild-type animals (Figure 1c). These results are consistent with prior studies showing TRPM8 expressed in this tissue (53). Of note, a proportion of GFP-positive axons also showed immunoreactivity for CGRP, consistent with soma co-expression of TRPM8 and CGRP in the trigeminal ganglia (9,10,34,50,51).

Figure 1.

TRPM8-positive afferents innervate mouse meninges. (a) Representative images of GFP (TRPM8) and WGA (Alex-350) co-labeling in trigeminal ganglia in adult Trpm8^GFP mice that received an intradural injection of WGA. Arrows highlight co-labeled cells. Immunoreactivity for GFP (TRPM8) and calcitonin gene-related peptide (CGRP) in the meninges of Trpm8^GFP mice (b) versus that in wild-type mice (c). Arrowheads denote axons that co-label for GFP and CGRP.

GFRα3 is required for NTG-induced tactile allodynia

Next, we asked how activation of TRPM8 channels in the meninges, which are likely not responding to temperature in this tissue, might lead to migraine-like pain. We reasoned that mechanisms known to induce TRPM8-dependent cold allodynia in the skin might also be relevant in the dura matter. The neurotrophic factor artemin, by binding to its cognate receptor GFRα3, can induce TRPM8-dependent cold hypersensitivity, and GFRα3 is also required for or involved in inflammatory and neuropathic pain of various etiologies, including migraine (32 –35). To test this premise, we measured acute periorbital mechanical allodynia in wild-type and GFRα3-null mice (Gfrα3^−/−) (34 –42) two hours after they were given an i.p. injection of 1 mg/kg NTG or vehicle, repeated every two days (31 –45) (Figure 2a). As expected, wild-type animals exhibited mechanical allodynia after each NTG injection that was significantly different than baseline (BL) at all days tested, as well as compared to vehicle injected mice by the second injection on day 3 (p < 0.01, two-way ANOVA with Tukey's multiple comparisons test). By contrast, Gfrα3^−/− mice were insensitive to NTG treatment because their mechanical sensitivity was not different than baseline or compared to vehicle injected animals (p > 0.05, two-way ANOVA with Tukey's multiple comparisons test) at all days tested. To ensure that the lack of a phenotype in Gfrα3^−/− mice was not the result of an NTG dosage effect, we also tested their acute response to 10 mg/kg NTG, again finding that the robust increase in mechanical allodynia in wild-type mice two hours after a single injection was absent in Gfrα3^−/− animals (Figure 2b). Lastly, to determine if GFRα3 was also required for extracephalic allodynia induced by NTG, we measured hind paw mechanical responses in wild-type and Gfrα3^−/− mice before and two hours after an i.p. injection of 10 mg/kg NTG, observing no differences in Gfrα3^−/− mice given NTG (Figure 2c).

Figure 2.

GFRα3 is required for nitroglycerin-induced acute cephalic and extracephalic tactile allodynia. (a) Left: Schematic of acute injections and mechanical testing timeline. Right: Wild-type mice exhibit periorbital mechanical allodynia measured two hours after each systemic nitroglycerin (NTG) injection (1 mg/kg) compared to baseline (BL) sensitivity (n = 8 for each condition, two-way ANOVA with Dunnett's multiple comparison test, *p < 0.05, ***p < 0.001, ****p < 0.0001). By contrast, GFRα3-null mice (Gfrα3^−/−) are insensitive to 1 mg/kg NTG injections on each day tested (n = 6 for each condition, two-way ANOVA with Dunnett's multiple comparison test compared to BL, p > 0.05). Wild-types receiving a 10 mg/kg dose of NTG show robust periorbital (b) and hindpaw (c) mechanical allodynia two hours post-injection compared to BL, a phenotype absent in Gfrα3^−/− mice (n = 6 and 8 for periorbital and hindpaw, respectively, paired two-tailed t-test, ^nsp > 0.05, **p < 0.01, ***p < 0.001).

Repeated NTG dosing leads to an increase in basal mechanical sensitivity that is considered a potential model of chronic migraine (31 –45). To test the role of GFRα3 in chronic migraine, we compared baseline behaviors at each injection day prior to treatment with 1 mg/kg NTG or vehicle, observing the expected increase in basal allodynia in wild-types, an effect that was absent in Gfrα3^−/− animals (Figure 3a). We again assessed 10 mg/kg NTG, also observing that basal mechanical allodynia was absent in Gfrα3^−/− mice two days after a single NTG injection (day 3) (Figure 3b). Similarly, hind paw mechanical allodynia did not develop in Gfrα3^−/− mice given 10 mg/kg NTG (Figure 3c). Thus, GFRα3 is required for both cephalic and extracephalic tactile allodynia in both acute and chronic models of rodent migraine-like pain, a phenotype similar to that observed in TRPM8-null mice (31).

Figure 3.

GFRα3 is required for chronic allodynia induced by repeated nitroglycerin (NTG) treatments. (a) Left: Schematic of chronic injections and mechanical testing timeline. Right: Wild-type and Gfrα3^−/− mice received 1 mg/kg NTG injections every two days, with only wild-types displaying enhanced basal periorbital mechanical allodynia compared to baseline (BL) at day 1 (n = 6–8 for each condition, two-way ANOVA with Dunnett's multiple comparison test, *p < 0.05, **p < 0.01, ***p < 0.001). Wild-types receiving a 10 mg/kg dose of NTG show robust basal periorbital (b) and hindpaw (c) mechanical allodynia two days post-injection compared BL, a phenotype is absent in Gfrα3^−/− mice (n = 6 each condition for periorbital and hindpaw, respectively, paired two-tailed t-test, ^nsp > 0.05, **p < 0.01, ****p < 0.0001).

Neutralizing artemin reduces NTG-induced tactile allodynia

Our results show that GFRα3 is required for NTG-induced tactile allodynia, suggesting that its specific ligand artemin might also play a role in this phenotype. To test this, we employed mAbs raised against artemin that we have shown can specifically ameliorate pathological cold allodynia (32 –34,54). Wild-types were tested for basal cephalic mechanical sensitivity then injected with either an anti-artemin mAb or an isotype control. The mice were re-tested one hour post-injection, demonstrating that mAb treatment did not alter basal mechanosensation (Figure 4a). Immediately after testing, mice treated with the two mAbs were injected with 1 mg/kg NTG and re-tested two hours post-injection. Mice given the isotype control showed robust mechanical allodynia, as expected, whereas mice injected with the anti-artemin mAb exhibited normal sensitivity (Figure 4a).

Figure 4.

Neutralizing artemin ameliorates nitroglycerin-induced tactile allodynia. (a) Top: Schematic of the experimental approach. Wild-type mice treated with either an anti-artemin or isotype control mAb exhibit normal periorbital mechanical sensitivity one hour post-injection compared to baseline (BL). A subsequent injection of 1 mg/kg nitroglycerin (NTG) induced allodynia in mice given the isotype control, but not the anti-artemin monoclonal antibody (mAb) (n = 6 for each condition, one-way ANOVA with Tukey's multiple comparisons tests for each time point and condition, ^nsp > 0.05, **p < 0.01). (b) Top: Schematic of the experimental approach. Wild-type mice treated with 1 mg/kg NTG to induce basal periorbital mechanical allodynia two days post-injection compared to BL, a phenotype that remained in mice subsequently treated with the isotype control mAb. However, allodynia was reduced in animals treated with the anti-artemin mAb (n = 10 for each condition, one-way ANOVA with Tukey's multiple comparisons test for different time point and condition, **p < 0.01, ^nsp > 0.05).

Next, we asked if chronic allodynia induced by NTG was sensitive to artemin neutralization. Animals were tested for basal mechanical sensitivity (day 1) and given a single injection of 1 mg/kg NTG. Their basal sensitivity was remeasured two days later (day 3), with all animals exhibiting mechanical allodynia (Figure 4b). These mice were then given either an injection of the isotype control or anti-artemin mAbs and tested one hour later. As with acute sensitization, artemin neutralization decreased mechanical allodynia, bringing the 50% withdrawal threshold back to basal levels. These data indicate that both acute and the initial phase of chronic mechanical allodynia induced by NTG are dependent on free, circulating artemin, as well as suggesting that established allodynia can be alleviated by artemin neutralization.

Dura artemin treatment induces TRPM8-dependent tactile allodynia

Our results suggest a key role for artemin and GFRα3 signaling in the NTG model of migraine-like pain but do not directly determine whether GFRα3 is acting at sites involved in migraine, such as the meninges. To address this, we first considered whether periorbital mechanical allodynia induced by stimulation of the meninges via direct supradural injection of a cocktail of inflammatory mediators required GFRα3 (43). Wild-type and Gfrα3^−/− mice were tested for BL mechanical sensitivity as above, then given a 5-μl injection of either vehicle (SIF) or an IS mixture (in SIF at pH 5.0). Both wild-type and Gfrα3^−/− mice given IS showed robust mechanical allodynia compared to vehicle injected mice (wild-types: p < 0.05 at one hour, p < 0.0001 at two to four hours post-injection; Gfrα3^−/−: p < 0.01 at two hours, p < 0.05 at 3 h, two-way ANOVA with Tukey's multiple comparison tests) and basal sensitivity (wild-types: p < 0.01 at one hour, p < 0.0001 at two to four hours post-injection; Gfrα3^−/−: p < 0.01 at two hours, p < 0.001 at three hours, p < 0.05 at four hours post-injection; two-way ANOVA with Dunnett's multiple comparison test) for several hours post-injection. However, the level of sensitization after IS injection was significantly lower in Gfrα3^−/− mice compared to wild-types at two to four hours post-injection (Figure 5a), indicating that GFRα3 mediates, in part, migraine-like pain produced by meningeal stimulation with inflammatory mediators.

Figure 5.

Necessity of GFRα3 and artemin in tactile allodynia induced by direct stimulation of the meninges. (a) Direct supradural infusion of an inflammatory soup (IS) mixture induces periorbital tactile allodynia that is partially dependent on GFRα3. Gfrα3^−/− mice given IS exhibited a reduced level of mechanical allodynia compared to wild-types at two to four hours post-injection (n = 10 each genotype, two-way ANOVA with Tukey's multiple comparisons test, *p < 0.05, **p < 0.01). (b) Supradural artemin induces robust periorbital tactile allodynia in wild-type but not Trpm8^−/− mice (n = 12 wild-types and n = 6 for Trpm8^−/− mice for each condition, two-way ANOVA with a Tukey's multiple comparisons test for vehicle vs. artemin at each time point; **p < 0.01, ***p < 0.001, ****p < 0.0001). (c) Wild-type mice treated with artemin and sumatriptan exhibit a partial reduction in periorbital mechanical allodynia compared to those given artemin and vehicle (n = 8 for each condition, *p < 0.05, **p < 0.01, two-way ANOVA with a Tukey's multiple comparison test).

Next, we considered whether direct application of artemin to the meninges in wild-type mice could induce a similar migraine-like phenotype. Naïve wild-type mice were tested for baseline periorbital mechanical sensitivity, given a supradural injection of either vehicle (HBSS) or 1 pg artemin, then retested for mechanical allodynia over the next 96 h. Consistent with our Gfrα3^−/− and artemin neutralization results, we observed robust allodynia in mice receiving artemin that lasted over 24 hours post-injection (Figure 5b). No sex differences were observed for this artemin-induced allodynia (n = 6 for each sex, p > 0.05 males versus females at all time points tested, two-way ANOVA with Bonferroni's multiple comparison test). Thus, similar to other inflammatory agents, artemin directly induces tactile allodynia when applied to the dura. Next, because artemin induces TRPM8-dependent cold hypersensitivity in the skin (32,34,35), we reasoned that artemin and GFRα3 signaling underlies the TRPM8-dependent migraine-like phenotypes that we observed (31).To test this hypothesis, we repeated the dura artemin injections in TRPM8-null mice (Trpm8^−/−) (41), finding that this treatment did not induce mechanical allodynia in the absence of TRPM8 (Figure 5b). Thus, as with cold allodynia in the skin, artemin-induced sensitization in the meninges is TRPM8 dependent. Lastly, to further define this artemin phenotype, we concurrently treated wild-type mice with supradural artemin and 0.6 mg/kg (s.c.) sumatriptan, the latter a traditional migraine treatment in humans and an agent used in animal studies to determine if a phenotype is migraine-like pain (55 –57). Mice given artemin and the vehicle for sumatriptan (0.9% Saline) again showed robust mechanical allodynia starting at two hours post-injection (p < 0.5 at two and six hours, p < 0.0001 at three and four hours, p < 0.01 at five hours post-injection, n = 8, two-way ANOVA with Tukey's multiple comparisons test). By contrast, mice given artemin and sumatriptan exhibited reduced mechanical allodynia from two to four hours post-injection, compared to artemin/vehicle treated mice (Figure 5c), suggesting that the artemin phenotype is consistent with migraine-like pain.

Activation of meningeal TRPM8 induces cephalic tactile allodynia independent of GFRα3

Becuase we find that artemin neutralization alleviates both acute and chronic tactile allodynia induced by NTG, and artemin-induced mechanical allodynia is TRPM8-dependent, we posit that this effect of artemin is via TRPM8 sensitization. Previously, we showed that acute extracephalic allodynia induced by 10 mg/kg NTG could be prevented by pre-treatment with the TRPM8 antagonist PBMC, but that established allodynia induced by chronic NTG measured after four NTG treatments given every two days was insensitive to TRPM8 inhibition (31). Although inconsistent with data in Figure 4(b), these prior results were with a substantially higher dose of NTG (10 mg/kg versus 1 mg/kg) given repeatedly. Thus, to determine whether TRPM8 underlies the basal sensitization shown in Figure 4(b), wild-type mice were given a single 1 mg/kg NTG injection, then tested for basal sensitivity two days later (BL-D3) with all animals exhibiting sensitization (n = 13, paired two-tailed t-test, p < 0.0001). These mice were randomly separated into two groups with one receiving an injection of 4 mg/kg PBMC and the other vehicle, then tested for mechanical sensitivity every hour for five hours. Mice given vehicle remained allodynic over this time period as there were no significant differences in mechanical sensitivity compared to the day 3 baseline (D3-BL; p > 0.05, two-way ANOVA with Dunnett's multiple comparison test) and remained significantly different than the responses measured pretreatment (BL) at all time points tested (p < 0.01 two-way ANOVA with Dunnett's multiple comparison test). Conversely, mice given PBMC showed transient relief from this allodynia by two hours post-PBMC compared to the day 3 baseline and to vehicle injected animals (Figure 6a). Thus, inhibition of TRPM8 systemically relieved the initial NTG induced basal mechanical allodynia in this chronic migraine model. Further, to contrast with our prior results, mice were similarly given four injections of 1 mg/kg NTG at two-day intervals, then their baseline (BL-D9) was measured two days after the fourth injection (day 9), with all animals exhibiting robust allodynia (paired two-tailed t-test, p < 0.0001) (Figure 6a). By contrast to its effects on day 3, mice given 4 mg/kg PBMC on day 9 remained allodynic with periorbital mechanical allodynia that was similar to vehicle treated animals (Figure 6a). These results are consistent with previous findings and demonstrate that, although the initial basal allodynia induced with one dose of NTG is TRPM8 and artemin dependent, the robust phenotype developed by repeated NTG treatments is independent of TRPM8.

Figure 6.

Activation of TRPM8 afferents in the mouse dura induces tactile allodynia. (a) Wild-type mice treated with 1 mg/kg nitroglycerin to induce basal periorbital mechanical allodynia at two days post-injection (day 3), as well as after four injections repeated every two days, measured nine days after the initial injections, compared to BL, a phenotype that remained in mice subsequently treated with vehicle (n = 8, two-way ANOVA with Dunnett's multiple comparisons test, compared to baseline (BL), p < 0.01 on day 3, p < 0.001 on day 9). However, basal allodynia on day 3, but not day 9, was reduced in animals treated with the peripheral blood mononuclear cells (PBMC) at two hours post-injection compared to vehicle treated mice (n = 8, two-way ANOVA with Dunnett's multiple comparisons test, ***p < 0.001). (b) Periorbital mechanical sensitivity measured in wild-type mice that received an intradural injection of either icilin or vehicle. Mechanical allodynia was evident by two hours post-injection and lasted for up to five hours (n = 12 for each condition, two-way ANOVA with a Bonferroni's multiple comparisons test for vehicle vs. icilin at each time point; *p < 0.05, ***p < 0.001, ****p < 0.0001). (c) TRPM8-null (Trpm8^−/−) mice displayed normal mechanical sensitivity in their hindpaws three hours after an intradural injection of icilin, in contrast to wild-type animals (n = 10 for each condition, one-way ANOVA with a Tukey's multiple comparisons test for vehicle vs. icilin in each condition; ^nsp > 0.05, **p < 0.01, ****p < 0.0001). (d) Intradural icilin induces periorbital tactile allodynia in Gfrα3^−/− mice (three hours post-injection, n = 8, paired two-tailed t-test, **p < 0.01).

These data, and evidence that mouse meningeal afferents are TRPM8-positive (Figure 1), suggest that sensitization of TRPM8 channels by artemin and GFRα3 signaling in this tissue leads to migraine-like phenotypes in the mouse, but it is unclear whether these are functional channels. Thus, we next asked if direct application of the potent and selective TRPM8 agonist icilin (9,58) to the mouse dura could induce behaviors consistent with mouse migraine-like phenotypes. Previously, Burgos-Vega et al. (26) demonstrated that supradural application of icilin in male rats induced both cephalic and extracephalic mechanical allodynia that was inhibited by co-application of sumatriptan and l-NAME. Although this effect was also sensitive to TRPM8 antagonism, the genetic or sex dependence of this icilin phenotype in mice has not been established. Thus, we directly applied icilin (1 nmol in 5 μl) to the mouse dura and measured changes in periorbital mechanical sensitivity pre- and post-injection, observing robust allodynia that was significantly different than vehicle injected mice by two hours, lasting up to five hours before returning to baseline (BL) by six hours post-injection (Figure 6b). This time course is consistent with previous findings in the rat (26). As with supradural artemin, no sex differences were observed for icilin-induced allodynia (n = 6 for each sex, p > 0.05 males versus females at all time points tested, two-way ANOVA with Bonferroni's multiple comparison test). We also tested extracephalic sensitivity in the mouse hind paw, observing mechanical allodynia at three hours post-injection, sensitization that was absent in Trpm8^−/− mice (Figure 6c). Lastly, in a final test of this hypothetical pathway, we predict that artemin and GFRα3 act upstream of TRPM8 and thus the dura icilin induced allodynia should occur independent of GFRα3, a hypothesis supported by clear periorbital tactile allodynia in Gfrα3^−/− mice treated with supradural icilin (Figure 6d). Thus, TRPM8 expressed in the mouse meninges is functional and activation of these channels leads to both cephalic and extracephalic mechanical allodynia that is consistent with preclinical rodent migraine models.

Discussion

TRPM8 is the principal detector of environmental cold in the skin but, similar to the heat-sensitive channel TRPV1, is also present in afferents innervating tissues not involved in thermosensation (59,60). For example, TRPM8-positive afferents extensively innervate the rodent bladder and contain nociceptor markers such as CGRP at a higher percentage than the overall TRPM8-positive somas per ganglion, consistent with the channel's role in bladder dysfunction (61 –63). Similarly, TRPV1 is expressed in visceral and muscle afferents in larger proportions compared to those innervating the skin (64). Thus, although present and functional, the thermosensitivity of these receptors is likely not relevant in these tissues, indicating that endogenous agonists or modulators sensitize or activate these channels to assert their physiological roles in regions other than the skin (65,66). For example, agonists such as menthol and capsaicin sensitize the temperature sensitivity of thermoTRPs such as TRPM8 (9) and TRPV1 (67), respectively, such that the channels gate at physiological temperatures (68). Similarly, several signal transduction cascades induced by inflammation sensitize TRPV1 such that the channel is active at skin temperatures (69,70). Thus, their sensitivity to endogenous and exogenous stimuli likely underlies their potential contribution to migraines (65). For TRPM8, artemin is the only known endogenous agent that leads directly to TRPM8-dependent hypersensitivity, a pathway that has only been studied in relation to cutaneous cold sensation (11). Here, we provide compelling evidence that this pathway also underlies a mechanism whereby TRPM8 sensitization may lead to headaches or migraine-like pain.

Artemin is a ligand of the GDNF family and plays crucial roles in nervous system development, nociception, and the pathogenesis of neurological disorders (71). At the transcriptional level, artemin mRNA is expressed in developing nerve roots, Schwann cells, vascular and visceral smooth muscle cells, endothelial cells, skin fibroblasts, and immune cells, indicating its potential release from these sites (72,73). GFRα3 is a high affinity receptor for artemin and shows a similar elevated expression pattern in sensory ganglia, smooth muscle cells, skin fibroblasts and endothelial cells (73). Notably, both artemin and GFRα3 are expressed in dura afferents, with elevated levels in rodent models of migraine, suggesting a significant role in migraine pathogenesis (37,38). Intriguingly, our previous work has demonstrated that artemin, through GFRα3, induces nociception, including cold hypersensitivity, in vivo (32 –35). This effect is TRPM8-dependent, with GFRα3 expressed in approximately half of TRPM8 neurons in the TG, and both inflammatory and neuropathic cold pain are absent in GFRα3-null mice (32,34). Peripheral ARTN neutralization alleviates established inflammatory and neuropathic cold pain, highlighting a potential therapeutic intervention (32,34,54). These findings suggest similarities in ARTN/GFRα3 signaling in both cold and trigeminovascular pain.

NTG, a nitric oxide donor, can induce headaches in both migraineurs and non-migraineurs by sensitizing the trigeminovascular system, causing vasodilation of meningeal blood vessels, and promoting the release of proinflammatory factors (74 –76). In this study, we administered a moderate dose of NTG (1 mg/kg, one-tenth of the commonly used dosage) (45) to induce a significant tactile allodynic phenotype in the cephalic region of mice, minimizing the risk of overwhelming sensitization. We found that the absence of GFRα3 and the neutralization of artemin via a monoclonal antibody prevented acute migraine-like pain triggered by NTG, suggesting that systemic NTG administration could induce artemin release and interaction with GFRα3 at the meninges. Additionally, a chronic migraine-like phenotype established by repeated NTG treatment was also attenuated by artemin neutralization or the absence of GFRα3, indicating sustained increased artemin levels facilitated baseline hypersensitivity in the rodent model of chronic migraine (45). These data compellingly demonstrate that the NTG-induced migraine-like phenotype is highly dependent on artemin release and signaling through GFRα3. This significant correlation is underscored by the observed elevation of artemin and GFRα3 levels in the dura mater of NTG-induced rodent migraine models and trigeminal ganglion cultures (39). Artemin, which is abundantly expressed in vascular smooth muscle cells, promotes vasodilation by relaxing these cells, and its level positively correlates with inducible nitric oxide synthase levels (39). Consequently, neutralizing artemin could mitigate NTG-induced vasodilation, a primary contributor to nociception in this model. Furthermore, the excessive production of nitric oxide in the NTG model can trigger an inflammatory response and subsequent neurogenic inflammation (77). The precise molecular mechanisms linking NTG effects and artemin release remain to be elucidated, and uncovering these relationships is crucial for developing future treatments.

A limitation of the NTG model is the systemic nature of its effects, making it difficult to determine whether the phenotype is primarily produced in the meninges. However, we found that direct application of artemin to the mouse dura mater triggers cephalic tactile allodynia, a TRPM8-dependent phenotype. Artemin induced allodynia was also inhibited by concurrent treatment with sumatriptan, further supporting this phenotype to be migraine-like. As shown here and in other reports, TRPM8-positive afferents are functionally expressed in mouse meninges, and activation of this channel in this tissue leads to both cephalic and extracephalic tactile allodynia (26,27,53,78). This phenotype is reminiscent of our prior work showing that peripheral nociceptor activation by endogenous proalgesics leads to artemin-dependent cold allodynia (32) and that subcutaneous artemin application leads to TRPM8-dependent cold pain (35). It is also essential to recognize that artemin might not account for all migraine episodes. For example, in the inflammatory soup migraine model, deleting the GFRα3 receptor only partially alleviates cephalic tactile allodynia induced by inflammatory factors. Hence, the development of migraine-like pain likely involves numerous molecular pathways, necessitating a therapeutic approach that targets multiple pathogenic routes for effective treatment.

Our results suggest that activating TRPM8 leads to migraine-like pain, but cooling is also reported to have a short-term effect alleviating headaches, and menthol modestly relieves migraines when applied either intranasally or to the forehead (28 –30,79 –81). Further, one report suggests TRPM8 has a protective effect and contributes to a faster recovery in the chronic migraine model in males but not females (82). TRPM8 is the predominant cold sensor in mammals, mediating innocuous cool and noxious cold sensation both acutely and in pathological settings (9,33,41,46,49,83 –85). Conversely, modest thermal or chemical cooling of the skin inhibits both inflammatory and neuropathic pain (49,86 –88). These same stimuli also alleviate acute and chronic itch, airway irritation and aversion to oral nicotine (89 –92). These processes are TRPM8-dependent and likely occur via TRPM8 afferent activation of spinal cord dorsal horn inhibitory interneurons, suggesting a similar mechanism in the trigeminocervical complex may provide headache relief (90,93). Because cephalic and extracephalic cutaneous allodynia with migraines is proposed to result from spinal and supraspinal central sensitization, respectively (94 –97), stimulation of cephalic cutaneous TRPM8 afferents, which converge with meningeal afferents in the trigeminocervical complex, may provide an inhibitory input that reduces sensitization produced in the meninges. Therefore, in the context of migraine-like pain, there may be a distinction between the activity of TRPM8-expressing afferents in the facial skin and those in the meninges, an intriguing posit that requires further investigation.

How can TRPM8 activation lead to migraine-like pain when its only known role in nociception is related to cold sensation? The prevailing model for the sequence of events leading to migraine suggests that a variety of factors alter cortical excitability that eventually leads to activation of meningeal trigeminal nociceptors which induce sensitization of second order neurons in the brainstem (98,99). Our results suggest that artemin/GFRα3 sensitization of TRPM8 in the meninges is one such pathway that can result in migraine and headaches. Further, when considered in the context of the TRPM8 variants identified in GWAS on migraine (21,23), individuals with elevated TRPM8 expression would be more susceptible to an increase in artemin or other endogenous agents that sensitize TRPM8 channels, perhaps underlying their predisposition to migraines (21). This hypothesis requires further study, but our results now implicate artemin/GFRα3/TRPM8 as novel players in the molecular mechanisms of migraine.

Article highlights

The neurotrophic factor artemin, and its receptor GFRα3, are required for NTG-induced tactile allodynia in mice.

Neutralization of free artemin relieves NTG-induced tactile allodynia in mice

Direct application of artemin to the mouse dura induces tactile allodynia that is dependent on the menthol receptor TRPM8.

Direct activation of TRPM8 in the meninges leads to tactile allodynia that is independent of GFRα3

Footnotes

Acknowledgements

We thank current and former members of the McKemy Laboratory for their assistance in the completion of this work.

Declaration of conflicting interests

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

Ethical statement

All experiments were approved by University of Southern California Institutional Animal Care and Use Committee and in compliance with ARRIVE guidelines for Animal Research.

Funding

The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The present study was funded by the National Institutes of Health –7 National Institute of Neurological Disorders and Stroke Awards #R01NS106888 and #R21NS118852 (grant number R56NS106888).

ORCID iDs

Chenyu Yang

David D. McKemy

References

Anttila

Wessman

Kallela

, et al. Genetics of migraine. Handb Clin Neurol 2018; 148: 493–503.

Peck

Johnson

Smitherman

. Migraine. Handb Clin Neurol 2016; 138: 283–293.

Chasman

Schurks

Anttila

, et al. Genome-wide association study reveals three susceptibility loci for common migraine in the general population. Nat Genet 2011; 43: 695–698.

Gormley

Anttila

Winsvold

, et al. Meta-analysis of 375,000 individuals identifies 38 susceptibility loci for migraine. Nat Genet 2016; 48: 856–866.

Freilinger

Anttila

de Vries

, et al. Genome-wide association analysis identifies susceptibility loci for migraine without aura. Nat Genet 2012; 44: 777–782.

Chasman

Anttila

Buring

, et al. Selectivity in genetic association with sub-classified migraine in women. PLoS Genet 2014; 10: e1004366.

Esserlind

Christensen

, et al. Replication and meta-analysis of common variants identifies a genome-wide significant locus in migraine. Eur J Neurol 2013; 20: 765–772.

Key

Abdul-Aziz

Mundry

, et al. Human local adaptation of the TRPM8 cold receptor along a latitudinal cline. PLoS Genet 2018; 14: e1007298.

McKemy

Neuhausser

Julius

. Identification of a cold receptor reveals a general role for TRP channels in thermosensation. Nature 2002; 416: 52–58.

10.

Takashima

Daniels

Knowlton

, et al. Diversity in the neural circuitry of cold sensing revealed by genetic axonal labeling of transient receptor potential melastatin 8 neurons. J Neurosci 2007; 27: 14147–14157.

11.

McKemy

. Molecular basis of peripheral innocuous cold sensitivity. Handb Clin Neurol 2018; 156: 57–67.

12.

Dussor

Cao

. TRPM8 and Migraine. Headache 2016; 56: 1406–1417.

13.

Kaur

Ali

Ahmad

, et al. Rs2651899 variant is associated with risk for migraine without aura from North Indian population. Mol Biol Rep 2019; 46: 1247–1255.

14.

Yang

, et al. Association between PRDM16, MEF2D, TRPM8, LRP1 gene polymorphisms and migraine susceptibility in the She ethnic population in China. Clin Invest Med 2019; 42: E21–E30.

15.

Chen

Fuh

Chung

, et al. Genome-wide association study identifies novel susceptibility loci for migraine in Han Chinese resided in Taiwan. Cephalalgia 2018; 38: 466–475.

16.

Sintas

Fernandez-Morales

Vila-Pueyo

, et al. Replication study of previous migraine genome-wide association study findings in a Spanish sample of migraine with aura. Cephalalgia 2015; 35: 776–782.

17.

Fan

Wang

Fan

, et al. Replication of migraine GWAS susceptibility loci in Chinese Han population. Headache 2014; 54: 709–715.

18.

Ghosh

Pradhan

Mittal

. Genome-wide-associated variants in migraine susceptibility: a replication study from North India. Headache 2013; 53: 1583–1594.

19.

Lin

, et al. PRDM16 Rs2651899 variant is a risk factor for Chinese common migraine patients. Headache 2013; 53: 1595–1601.

20.

Schurks

. Genetics of migraine in the age of genome-wide association studies. J Headache Pain 2012; 13: 1–9.

21.

Gavva

Sandrock

Arnold

, et al. Reduced TRPM8 expression underpins reduced migraine risk and attenuated cold pain sensation in humans. Sci Rep 2019; 9: 19655.

22.

Edvinsson

. Role of CGRP in migraine. Handb Exp Pharmacol 2019; 255: 121–130.

23.

Ling

Chen

Fann

, et al. TRPM8 Genetic variant is associated with chronic migraine and allodynia. J Headache Pain 2019; 20: 15.

24.

Vigano

Manica

Di Piero

, et al. Did going north give US migraine? An evolutionary approach on understanding latitudinal differences in migraine epidemiology. Headache 2019; 59: 632–634.

25.

Chebini

Dilli

. Cold stimulus headache. Curr Neurol Neurosci Rep 2019; 19: 46.

26.

Burgos-Vega

Ahn

Bischoff

, et al. Meningeal transient receptor potential channel M8 activation causes cutaneous facial and hindpaw allodynia in a preclinical rodent model of headache. Cephalalgia 2016; 36: 185–193.

27.

Ren

Dhaka

Cao

. Function and postnatal changes of dural afferent fibers expressing TRPM8 channels. Mol Pain 2015; 11: 37.

28.

Lopresti

Smith

Drummond

. Herbal treatments for migraine: a systematic review of randomised-controlled studies. Phytother Res 2020; 34: 2493–2517.

29.

Borhani Haghighi

Motazedian

Rezaii

, et al. Cutaneous application of menthol 10% solution as an abortive treatment of migraine without aura: a randomised, double-blind, placebo-controlled, crossed-over study. Int J Clin Pract 2010; 64: 451–456.

30.

Rafieian-Kopaei

Hasanpour-Dehkordi

Lorigooini

, et al. Comparing the effect of intranasal lidocaine 4% with peppermint essential oil drop 1.5% on migraine attacks: a double-blind clinical trial. Int J Prev Med 2019; 10: 21.

31.

Wei

Kim

McKemy

. Transient receptor potential melastatin 8 is required for nitroglycerin- and calcitonin gene-related peptide-induced migraine-like pain behaviors in mice. Pain 2022; 163: 2380–2389.

32.

Yang

Yamaki

Jung

, et al. Endogenous inflammatory mediators produced by injury activate TRPV1 and TRPA1 nociceptors to induce sexually dimorphic cold pain that is dependent on TRPM8 and GFRalpha3. J Neurosci 2023; 43: 2803–2814.

33.

Yamaki

Chau

Gonzales

, et al. Nociceptive afferent phenotyping reveals that transient receptor potential ankyrin 1 promotes cold pain through neurogenic inflammation upstream of the neurotrophic factor receptor GFRalpha3 and the menthol receptor transient receptor potential melastatin 8. Pain 2021; 162: 609–618.

34.

Lippoldt

Ongun

Kusaka

, et al. Inflammatory and neuropathic cold allodynia are selectively mediated by the neurotrophic factor receptor GFRalpha3. Proc Natl Acad Sci U S A 2016; 113: 4506–4511.

35.

Lippoldt

Elmes

McCoy

, et al. Artemin, a glial cell line-derived neurotrophic factor family member, induces TRPM8-dependent cold pain. J Neurosci 2013; 33: 12543–12552.

36.

Shu

Song

, et al. Downregulations of TRPM8 expression and membrane trafficking in dorsal root ganglion mediate the attenuation of cold hyperalgesia in CCI rats induced by GFRalpha3 knockdown. Brain Res Bull 2017; 135: 8–24.

37.

McIlvried

Albers

Gold

. Distribution of artemin and GFRalpha3 labeled nerve fibers in the dura mater of rat: artemin and GFRalpha3 in the dura. Headache 2010; 50: 442–450.

38.

Shang

Wang

Mao

, et al. Expression of artemin and GFRalpha3 in an animal model of migraine: possible role in the pathogenesis of this disorder. J Headache Pain 2016; 17: 81.

39.

Shang

Wang

Chao

, et al. Artemin transiently increases iNOS expression in primary cultured trigeminal ganglion neurons. Neurosci Lett 2017; 660: 34–38.

40.

Percie du Sert

Ahluwalia

Alam

, et al. Reporting animal research: explanation and elaboration for the ARRIVE guidelines 2.0. PLoS Biol 2020; 18: e3000411.

41.

Bautista

Siemens

Glazer

, et al. The menthol receptor TRPM8 is the principal detector of environmental cold. Nature 2007; 448: 204–208.

42.

Honma

Araki

Gianino

, et al. Artemin is a vascular-derived neurotropic factor for developing sympathetic neurons. Neuron 2002; 35: 267–282.

43.

Burgos-Vega

Quigley

Trevisan Dos Santos

, et al. Non-invasive dural stimulation in mice: a novel preclinical model of migraine. Cephalalgia 2019; 39: 123–134.

44.

Louveau

Filiano

Kipnis

. Meningeal whole mount preparation and characterization of neural cells by flow cytometry. Curr Protoc Immunol 2018; 121: 50.

45.

Pradhan

Smith

McGuire

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

46.

Knowlton

Daniels

Palkar

, et al. Pharmacological blockade of TRPM8 ion channels alters cold and cold pain responses in mice. PLoS One 2011; 6: e25894.

47.

Bonin

Bories

De Koninck

. A simplified up-down method (SUDO) for measuring mechanical nociception in rodents using von frey filaments. Mol Pain 2014; 10: 26.

48.

Zhao

, et al. Behaviors related to psychiatric disorders and pain perception in C57BL/6J mice during different phases of estrous cycle. Front Neurosci 2021; 15: 650793.

49.

Knowlton

Palkar

Lippoldt

, et al. A sensory-labeled line for cold: TRPM8-expressing sensory neurons define the cellular basis for cold, cold pain, and cooling-mediated analgesia. J Neurosci 2013; 33: 2837–2848.

50.

Dhaka

Earley

Watson

, et al. Visualizing cold spots: TRPM8-expressing sensory neurons and their projections. J Neurosci 2008; 28: 566–575.

51.

Peier

Moqrich

Hergarden

, et al. A TRP channel that senses cold stimuli and menthol. Cell 2002; 108: 705–715.

52.

Avona

Mason

Burgos-Vega

, et al. Meningeal CGRP-prolactin interaction evokes female-specific migraine behavior. Ann Neurol 2021; 89: 1129–1144.

53.

Ren

Chang

Zhang

, et al. Quantitative analysis of mouse dural afferent neurons expressing TRPM8, VGLUT3, and NF200. Headache 2018; 58: 88–101.

54.

Jeong

Zhang

Yamaki

, et al. Immunoglobulin somatic hypermutation in a defined biochemical system recapitulates affinity maturation and permits antibody optimization. Nucleic Acids Res 2022; 50: 11738–11754.

55.

Edelmayer

Yan

, et al. Activation of TRPA1 on dural afferents: a potential mechanism of headache pain. Pain 2012; 153: 1949–1958.

56.

Edelmayer

Vanderah

Majuta

, et al. Medullary pain facilitating neurons mediate allodynia in headache-related pain. Ann Neurol 2009; 65: 184–193.

57.

De Felice

Ossipov

Wang

, et al. Triptan-induced latent sensitization: a possible basis for medication overuse headache. Ann Neurol 2010; 67: 325–337.

58.

Wei

Seid

. AG-3-5: a chemical producing sensations of cold. J Pharm Pharmacol 1983; 35: 110–112.

59.

Stein

Santos

Nagatomi

, et al. Cool (TRPM8) and hot (TRPV1) receptors in the bladder and male genital tract. J Urol 2004; 172: 1175–1178.

60.

Harrington

Hughes

Martin

, et al. A novel role for TRPM8 in visceral afferent function. Pain 2011; 152: 1459–1468.

61.

Hayashi

Kondo

Ishimatsu

, et al. Expression of the TRPM8-immunoreactivity in dorsal root ganglion neurons innervating the rat urinary bladder. Neurosci Res 2009; 65: 245–251.

62.

Lashinger

Steiginga

Hieble

, et al. AMTB, a TRPM8 channel blocker: evidence in rats for activity in overactive bladder and painful bladder syndrome. Am J Physiol Renal Physiol 2008; 295: F803–F810.

63.

Tsukimi

Mizuyachi

Yamasaki

, et al. Cold response of the bladder in Guinea pig: involvement of transient receptor potential channel, TRPM8. Urology 2005; 65: 406–410.

64.

Christianson

McIlwrath

Koerber

, et al. Transient receptor potential vanilloid 1-immunopositive neurons in the mouse are more prevalent within colon afferents compared to skin and muscle afferents. Neuroscience 2006; 140: 247–257.

65.

Benemei

Dussor

. TRP Channels and migraine: recent developments and new therapeutic opportunities. Pharmaceuticals (Basel) 2019; 12: 54.

66.

Spekker

Kortesi

Vecsei

. TRP channels: recent development in translational research and potential therapeutic targets in migraine. Int J Mol Sci 2022; 24: 700.

67.

Caterina

Schumacher

Tominaga

, et al. The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature 1997; 389: 816–824.

68.

Julius

. TRP Channels and pain. Annu Rev Cell Dev Biol 2013; 29: 355–384.

69.

Chuang

Prescott

Kong

, et al. Bradykinin and nerve growth factor release the capsaicin receptor from PtdIns(4,5)P2-mediated inhibition. Nature 2001; 411: 957–962.

70.

Premkumar

Ahern

. Induction of vanilloid receptor channel activity by protein kinase C. Nature 2000; 408: 985–990.

71.

Forrest

Osborne

Keast

. Characterization of axons expressing the artemin receptor in the female rat urinary bladder: a comparison with other major neuronal populations. J Comp Neurol 2014; 522: 3900–3927.

72.

Zhu

Bennett

, et al. The role of glial cell line-derived neurotrophic factor family member artemin in neurological disorders and cancers. Cell Prolif 2020; 53: e12860.

73.

Uhlen

Fagerberg

Hallstrom

, et al. Proteomics. Tissue-based map of the human proteome. Science 2015; 347: 1260419.

74.

Akerman

Romero-Reyes

Karsan

, et al. Therapeutic targeting of nitroglycerin-mediated trigeminovascular neuronal hypersensitivity predicts clinical outcomes of migraine abortives. Pain 2021; 162: 1567–1577.

75.

Christiansen

Thomsen

Daugaard

, et al. Glyceryl trinitrate induces attacks of migraine without aura in sufferers of migraine with aura. Cephalalgia 1999; 19: 660–667. discussion 626.

76.

Demartini

Greco

Zanaboni

, et al. Nitroglycerin as a comparative experimental model of migraine pain: from animal to human and back. Prog Neurobiol 2019; 177: 15–32.

77.

Chen

Hsu

Wang

, et al. Genome-wide analyses identify novel risk loci for cluster headache in Han Chinese residing in Taiwan. J Headache Pain 2022; 23: 47.

78.

Huang

Dhaka

, et al. Expression of the transient receptor potential channels TRPV1, TRPA1 and TRPM8 in mouse trigeminal primary afferent neurons innervating the dura. Mol Pain 2012; 8: 66.

79.

Hsu

Chen

, et al. Cold intervention for relieving migraine symptoms: A systematic review and meta-analysis. J Clin Nurs 2022; 32: 2455–2465.

80.

Ugurlu

Enc

. The effect of local cold compresses for nitroglycerin-induced headache: An observational pretest-posttest study. Nurs Crit Care 2023; 28: 1097–1105.

81.

Cyr A

Chen

Bradley

, et al. Efficacy and tolerability of STOPAIN for a migraine attack. Front Neurol 2015; 6: 11.

82.

Alarcon-Alarcon

Cabanero

de Andres-Lopez

, et al. TRPM8 Contributes to sex dimorphism by promoting recovery of normal sensitivity in a mouse model of chronic migraine. Nat Commun 2022; 13: 6304.

83.

Knowlton

Bifolck-Fisher

Bautista

, et al. TRPM8, But not TRPA1, is required for neural and behavioral responses to acute noxious cold temperatures and cold-mimetics in vivo. Pain 2010; 150: 340–350.

84.

Pogorzala

Mishra

Hoon

. The cellular code for mammalian thermosensation. J Neurosci 2013; 33: 5533–5541.

85.

Ongun

Sarkisian

McKemy

. Selective cold pain inhibition by targeted block of TRPM8-expressing neurons with quaternary lidocaine derivative QX-314. Commun Biol 2018; 1: 53.

86.

Dhaka

Murray

Mathur

, et al. TRPM8 Is required for cold sensation in mice. Neuron 2007; 54: 371–378.

87.

Proudfoot

Garry

Cottrell

, et al. Analgesia mediated by the TRPM8 cold receptor in chronic neuropathic pain. Curr Biol 2006; 16: 1591–1605.

88.

Liu

Fan

Balakrishna

, et al. TRPM8 Is the principal mediator of menthol-induced analgesia of acute and inflammatory pain. Pain 2013; 154: 2169–2177.

89.

Palkar

Ongun

Catich

, et al. Cooling relief of acute and chronic itch requires TRPM8 channels and neurons. J Invest Dermatol 2018; 138: 1391–1399.

90.

Kardon

Polgar

Hachisuka

, et al. Dynorphin acts as a neuromodulator to inhibit itch in the dorsal horn of the spinal cord. Neuron 2014; 82: 573–586.

91.

Fan

Balakrishna

Jabba

, et al. Menthol decreases oral nicotine aversion in C57BL/6 mice through a TRPM8-dependent mechanism. Tob Control 2016; 25: ii50–ii54.

92.

Smith

Cichocki

, et al. Menthol attenuates respiratory irritation and elevates blood cotinine in cigarette smoke exposed mice. PLoS One 2015; 10: e0117128.

93.

Pergolizzi

Jr Taylor

Jr LeQuang

, et al. The role and mechanism of action of menthol in topical analgesic products. J Clin Pharm Ther 2018; 43: 313–319.

94.

Noseda

Burstein

. Migraine pathophysiology: anatomy of the trigeminovascular pathway and associated neurological symptoms, cortical spreading depression, sensitization, and modulation of pain. Pain 2013; 154: S44–S53.

95.

Burstein

Yamamura

Malick

, et al. Chemical stimulation of the intracranial dura induces enhanced responses to facial stimulation in brain stem trigeminal neurons. J Neurophysiol 1998; 79: 964–982.

96.

Burstein

Yarnitsky

Goor-Aryeh

, et al. An association between migraine and cutaneous allodynia. Ann Neurol 2000; 47: 614–624.

97.

Burstein

Cutrer

Yarnitsky

. The development of cutaneous allodynia during a migraine attack clinical evidence for the sequential recruitment of spinal and supraspinal nociceptive neurons in migraine. Brain 2000; 123: 1703–1709.

98.

Ashina

Hansen

, et al. Migraine and the trigeminovascular system-40 years and counting. Lancet Neurol 2019; 18: 795–804.

99.

Charles

Brennan

. The neurobiology of migraine. Handb Clin Neurol 2010; 97: 99–108.