TRPM3 activation causes CGRP release in trigeminal neurons: Implications for migraine mechanisms.

Philip V. Reducha: Data curation; formal analysis; investigation; methodology; visualization; writing – original draft; writing – review and editing. Lukas K. S. Nielsen: Data curation; formal analysis; methodology; writing – review and editing. Mette N. Jensen: Data curation; writing – review and editing. Jacob C. A. Edvinsson: Data curation; formal analysis; methodology; writing – review and editing. Sofia L. Wæver: Data curation; writing – review and editing. Spyridoula Kazantzi: Data curation; formal analysis; writing – review and editing. Tanja Lylloff: Data curation; writing – review and editing. Connar S. J. Westgate: Data curation; investigation; methodology; writing – review and editing. Lars Edvinsson: Project administration; resources; writing – review and editing. Kristian A. Haanes: Conceptualization; data curation; formal analysis; funding acquisition; investigation; methodology; project administration; resources; supervision; visualization; writing – review and editing.

Kristian A. Haanes was supported by a Lundbeck Foundation Fellowship (R345‐2020‐1977). P.V. Reducha received a grant for the above‐mentioned work from Migrænefonden af 1988.

A total of 138 (male n  = 75, female n  = 63) Sprague–Dawley rats were used across experiments. A n  = 9 of female transgenic mice (Calca‐Cre + / − ; Salsa6f + / − , see below) were used for the calcium imaging study. Animals were anesthetized with a mixture of 70% CO₂ and 30% O₂ and euthanized via decapitation. For immunohistochemistry of TG and dura mater, rats were transcardially perfused with 4% paraformaldehyde in a 10% formalin buffer to ensure proper fixation. Tissue samples, including hemi‐skulls with intact dura mater, TG, MCA, and MMA, were collected. All experiments were approved by the Danish Animal inspectorate, license number 2020‐15‐0201‐00751 and 2023‐25‐0201‐01469. Human dura mater samples with MMAs were donated from three consenting patients undergoing neurosurgery at the University Hospital of Lund (Lund, Sweden). The male donors were 36, 66, and 79 years of age, whereas the female donors were 44, 60, and 75 years of age. The experiments were approved by the ethical committee at Lund University (LU‐818‐01). These tissues were fixed in 10% formalin buffer containing 4% paraformaldehyde for immunohistochemistry. All procedures involving human tissue were approved by the relevant ethics committees, with informed consent obtained prior to collection. Compounds and reagants were purchased from Sigma‐Aldrich (Germany), unless stated otherwise. TGs from a total of eight (male n  = 4; female n  = 4) Sprague–Dawley rats were fixed for 4 h in paraformaldehyde, and cryoprotected overnight in 10% and subsequently 25% sucrose solutions in Sorensen's phosphate buffer (0.1 M sodium dihydrogen phosphate (NaH₂PO₄) and 0.1 M disodium hydrogen phosphate (Na₂HPO₄)), embedded in a gelatin medium (30% egg albumin, 3% gelatin), and sectioned at 10 μm using a cryostat (Leica CM3050 S, Leica Biosystems, Germany). Sections were mounted on microscope slides and stored at −20°C. Tissue sections were thawed at room temperature, rehydrated, and encircled with a hydrophobic barrier pen before permeabilization in phosphate‐buffered saline with 0.25% Triton X‐100 (PBS‐T) for three 5‐min intervals. Blocking was performed for 20 min in PBS‐T containing 3% bovine serum albumin (BSA) and 0.3 M glycine. Primary antibodies targeting CGRP (ab81887, mouse, mono‐clonal, Abcam, UK) and TRPM3 (ACC‐050‐200UL, rabbit, polyclonal, Alomone Labs, Isreal) were diluted 1:200 in antibody diluent (PBS‐T containing 1% BSA) and incubated with the tissues overnight at +4°C. The antibodies used have been validated previously. 29 , 35 Slides were washed in antibody diluent for three 5‐min intervals to remove excess primary antibody, followed by incubation with Alexa Fluor 488 (Goat anti‐Rabbit/anti‐Mouse IgG, Abcam) and Alexa Fluor 568 (Goat anti‐Rabbit/anti‐Mouse, Abcam) secondary antibodies diluted 1:400, for 1 h in the dark. Excess secondary antibody was washed off with PBS‐T for three 5‐min intervals, and any remaining PBS crystals were removed by rinsing with ultrapure water for 1 min. Experiments were performed in triplicate for each human, or minimum n  = 4 for rats, to ensure reproducibility. Slides were mounted with Vectashield antifade medium (Vector Laboratories, USA) containing 4′,6‐diamidino‐2‐phenylindole for nuclear staining, and images were captured using a Nikon Ti2‐E microscope or Nikon 80i (Tokyo, Japan). A total of 72 (male n  = 42; female n  = 30) Sprague–Dawley rats, with a total of 104 hemi‐skulls (male n  = 58; female n  = 46) and 109 TG (male n  = 65; female n  = 44) were used throughout the CGRP release experiments. In a well‐validated CGRP release model, 20 , 36 rat TG and hemi‐skulls with intact dura mater were incubated in oxygenated synthetic interstitial fluid (SIF) buffer containing 108 mM NaCl, 3.5 mM KCl, 3.5 mM MgSO₄, 26 mM NaHCO₃, 11.7 mM NaH₂PO₄, 1.5 mM CaCl₂, 9.6 mM NaGluconate, 5.6 mM glucose, and 7.6 mM sucrose (pH ~7.3). The TG were incubated in 10 mL of SIF buffer, and hemi‐skulls were submerged in beakers containing the same buffer. Tissues were transferred to a humidified chamber at +37°C. TG were incubated for 30 min, whereas the SIF buffer in the hemi‐skulls was replaced with prewarmed (+37°C) oxygenated SIF buffer after 15 min, followed by an additional 15‐min incubation. All tissues underwent four sequential 10‐min washes with 300 μL of prewarmed SIF buffer. After washes, 300 μL of fresh buffer was added and baseline supernatants (200 μL) were collected after 15 min. Supernatants were transferred into Eppendorf tubes and mixed with 50 μL of enzyme immunoassay buffer, ensuring thorough mixing, and stored at −80°C until analysis. For stimulation, 300 μL of SIF buffer containing either 10 μM or 100 μM CIM0216 (MedChemExpress, USA) (left tissues) or 100 nM capsaicin (S (right tissues) was added. In TRPM3 specificity experiments, hemi‐skulls, and TG from a separate cohort of rats were pretreated with 10 μM isosakuranetin (MedChemExpress) (TRPM3 antagonist, left tissues) or vehicle (dimethyl sulfoxide [DMSO], right tissues) for 15 min before stimulation. After pretreatment, tissues were incubated with 300 μL of SIF buffer containing 100 μM CIM0216 with 10 μM isosakuranetin (left tissues) or 100 μM CIM0216 with vehicle (right tissues) for an additional 15 min. Poststimulation, supernatants (200 μL) were collected and processed as described above. For detailed overview of the timeline, see Figure  1 . Experimental timeline for CGRP release assay from rat hemi‐skull and TG preparations. Following euthanasia and tissue dissection, left and right hemi‐skulls and TGs were incubated for 30 min, followed by four sequential 10‐min washes. Baseline CGRP levels were then collected, followed by 15‐min incubations with one of the following agents: CIM0216 (10 μM or 100 μM), capsaicin (100 nM), isosakuranetin (10 μM), or vehicle. Tissues treated with vehicle or isosakuranetin underwent an additional 15‐min incubation with coapplication of 100 μM CIM0216. Each treatment condition was applied to distinct tissue samples in parallel. The figure was made using Biorender. CGRP, calcitonin gene‐related peptide; TG, trigeminal ganglia. [Color figure can be viewed at wileyonlinelibrary.com ] In order to investigate the potential involvement of the TRPA1 channel in the CIM0216 stimulation, a separate cohort of male Sprague–Dawley rats was used. TG and hemi‐skulls were pretreated with 3 μM A‐967079 (MedChemExpress) (TRPA1 antagonist, left tissues) or vehicle (DMSO, right tissues). After pretreatment, tissues were incubated with 300 μL of SIF buffer containing 100 μM CIM0216 with 3 μM A‐967079 (left tissues) or 100 μM CIM0216 with vehicle (right tissues) for an additional 15 min. Poststimulation, supernatants (200 μL) were collected and processed as described above. A separate cohort of male and female Sprague–Dawley rats was used to investigate the specificity of A‐967079 on the TRPA1 antagonism. TG and hemi‐skulls were pretreated with 3 μM A‐967079 (TRPA1 antagonist, left tissues) or vehicle (DMSO, right tissues). After pretreatment, tissues were incubated with 300 μL of SIF buffer containing 100 μM supercinnamaldehyde (MedChemExpress) (TRPA1 agonist) with 3 μM A‐967079 (left tissues) or 100 μM supercinnamaldehyde with vehicle (right tissues) for an additional 15 min. Poststimulation, supernatants (200 μL) were collected and processed as described above. CGRP levels were quantified using a verified Human CGRP ELISA kit (Bertin, France) 37 following the instructions of the manufacturer. Optical density was measured at 410 nm using a microplate photometer (Tecan, Infinite M200, Switzerland). Samples with baseline CGRP values exceeding 60 pg/mL were excluded from the experiments to minimize the risk of variability caused by overly stimulated tissue at baseline. Segments (~2 mm in length) of MCA from 20 (male n  = 10; female n  = 10) and MMA from six (male n  = 3; female n  = 3) Sprague–Dawley rats were mounted in a wire myograph system 38 (Danish Myo Technology, Denmark) in Krebs buffer containing 119 mM NaCl; 15 mM NaHCO₃; 4.6 mM KCl; 1.5 mM CaCl₂; 1.2 mM MgCl₂; 1.2 mM NaH₂PO₄; and 5.5 mM glucose, pH 7.4. Chambers were aerated with 95% O₂ and 5% CO₂ and maintained at +37°C. Arteries were equilibrated for 30 min and normalized to 90% of the internal circumference at a transmural pressure of 13.3 kilopascals (100 mmHg). After normalization, arteries were precontracted with a thromboxane A2 analogue (U46619, Cayman Chemical, USA) (10 −7  M). Cumulative concentration‐response curves were generated for CIM0216 (10 −9 to 10 −4  M), and responses were recorded. Vehicle control experiments used the same solution without CIM0216. A total of 32 (male n  = 16 male; female n  = 16) Sprague–Dawley rats were habituated to the electronic von Frey (EVF) test device on days −8, −6, and −4. Withdrawal thresholds were recorded for the periorbital and hindpaw regions, and this method has been validated previously. 39 , 40 The periorbital area was divided into three points (left, middle, and right), whereas the right hindpaw was divided into four points. Each point was tested once, except for the first point, which was tested twice (beginning and end of the session) to aid habituation. Testing was conducted individually to minimize stress, with auditory distractions reduced by placing remaining rats in a separate room during sessions. The maximum pressure threshold was set at 300 g to prevent harm to the animals. Rats were stratified by their baseline periorbital thresholds and assigned to two groups to receive subcutaneous injections of either vehicle or 3 mg/kg CIM0216. CIM0216 was prepared as a 1.5 mg/mL solution in vehicle (10% DMSO, 40% polyethylene glycol 300, 5% Tween‐80, and 45% saline). 3 mg/kg of CIM0216 with this vehicle solution were chosen because they represent the maximum concentration that we could achieve in the recommended vehicle mixture suggested by the supplier (MedChemExpress). EVF measurements were taken 1, 3, and 24 h postinjection. A crossover design was employed, with groups receiving the opposite treatment 1 week later. The investigators were blinded to the treatment. Female mice generated by crossing Calca‐Cre (B6.Cg‐Calca tm1.1(cre/EGFP)Rpa/J ) and Salsa6f reporter mice (B6(129S4)‐Gt(ROSA)26Sor tm1.1(CAG‐tdTomato/GCaMP6f)Mdcah/J ), both obtained from Jackson Laboratories (USA), were used to perform calcium imaging. Homozygous mice were crossed to produce heterozygous offspring (Calca‐Cre + / − ; Salsa6f + / − ) expressing tdTomato and GCaMP6f in Calca‐expressing cells. Female mice ( n  = 9) were used in the experiments, and estrous cycle was not monitored. TRPM3 expression was confirmed in the transgenic mice. Female mice were anesthetized with CO₂ and subsequently euthanized by decapitation. The TG were rapidly dissected, the dura mater carefully removed, and the tissues placed in room‐temperature oxygenated SIF buffer. TG tissues were washed three times for 1 min each in freshly oxygenated SIF. The ganglia were then mounted on a custom 3D‐printed holder to minimize movement during compound application. Samples were submerged in 2400 μL of freshly oxygenated SIF in a glass‐bottom dish (WillCo Wells, Netherlands) and positioned on a Nikon Eclipse Ti2 inverted fluorescence microscope (Nikon Instruments, Japan). Temperature was maintained at +37°C using a stage‐top incubator (UNO‐T‐H‐CO 2 , Okolab, Italy). For preincubation, 300 μL of SIF containing DMSO (vehicle control) or 10 μM isosakuranetin (TRPM3 antagonist) was preheated to +37°C and applied for 5 min. For agonist stimulation, 3 μM CIM0216 (TRPM3 agonist, preheated to +37°C in SIF) was added and calcium responses were recorded for 5 min. In antagonist experiments, 3 μM CIM0216 was coapplied with 10 μM isosakuranetin, and responses were recorded for the same duration. For vehicle treatment, DMSO was added and responses recorded for the same duration. A second stimulation with 30 μM CIM0216, 30 μM CIM0216 with 10 μM isosakuranetin, or DMSO was then applied and recorded for an additional 5 min. Calcium imaging was performed at a frame rate of 3 Hz. Image analysis was conducted using FIJI (ImageJ, USA), Motion correction and background adjustments were performed when necessary. Δ F / F min was calculated for each neuron. Here, F min represents the minimum fluorescence intensity, and Δ F represents the change from that baseline at each time point. Statistical analyses and graphs were generated using GraphPad Prism, version 10 (USA). Data are presented as mean ± standard error of the mean. Side‐to‐side comparisons in CGRP release experiments were analyzed using ratio‐paired Student's t ‐tests and paired Student t ‐tests for agonist values compared to antagonist values, with p  < 0.05 considered statistically significant. For behavioral and myograph tests, two‐way analysis of variance (ANOVA) was performed with appropriate factors for each experimental condition. We based the sample size on previous experience; no power calculation was used. Data distribution was assessed using the Shapiro–Wilk test. If the normality test was not passed, the data was analyzed using a nonparametric test (Mann–Whitney, highlighted in the legend), unless the corresponding quantile–quantile (Q–Q) plot indicated that the deviation was consistent with expected biological variability. The data for the calcium imaging analysis was performed using Fisher's exact test. ROUT tests were performed to identify potential outliers throughout experiments.

Immunofluorescence staining of TRPM3 and CGRP localization were analyzed in the TG, dura mater, and MCA of male and female rats using immunohistochemistry (Figure  2 ). In the male TG (Figure  2A ) and female TG (Figure  2D ), TRPM3 and CGRP coexpression were observed in specific neuronal subpopulations. Neurons with widespread CGRP expression (white arrows, one fletching) were classified as likely C‐neurons, whereas those with CGRP localized in Golgi‐like structures (white arrows, two fletchings) were identified as likely Aδ‐neurons. 41 A clear membrane staining was observed in the fibers identified as likely Aδ‐neurons. TRPM3 and CGRP localization in male and female rat and human trigeminal system tissues. Immunofluorescence staining for TRPM3 (green), CGRP (red), and nuclei (blue) is shown in TG, dura mater, and MCA from male (A–C) and female (D–F) rats, as well as in human dura mater and MMA (G‐L). In rat TG (A, B), TRPM3 and CGRP coexpression is observed in distinct neuronal populations; neurons with diffuse CGRP labeling (white arrows, one fletching) likely represent C‐fiber neurons. And neurons with CGRP localized to Golgi‐like structures (white arrows, two fletchings) likely represents Aδ‐neurons. In the rat dura mater (B, E), CGRP‐positive fibers are observed alongside TRPM3‐expressing smooth muscle in the MMA (blue arrows) in single fibers (white arrow, one fletching) or bundle of fibers (white arrows, three fletchings). In the MCA (C, F), TRPM3 is localized to smooth muscle cells (blue arrows) and endothelial cells (pink arrows), whereas CGRP‐positive fibers (white arrows, one fletching), likely originating from the TG, are located near the adventitia. In human dura mater and MMA (G–L), TRPM3 is predominantly expressed in the smooth muscle layer (M) and the adjacent A, whereas CGRP‐positive fibers (white arrows) are observed primarily in the A and near vascular structures. Scale bars represent 50 μm in A–F and 25 μm in G–L. Images are representative of n  = 3–4 immunohistochemistry experiments for each condition. A, adventitia; CGRP, calcitonin gene‐related peptide; I, intima; M, media; MCA, middle cerebral artery; MMA, middle meningeal artery; TG, trigeminal ganglia; TRPM3, transient receptor potential melastatin‐3. [Color figure can be viewed at wileyonlinelibrary.com ] In the male dura mater (Figure  2B ) and female dura mater (Figure  2E ) of rats, CGRP‐positive fibers were observed running parallel to TRPM3 expression. Colocalization was observed both in singular fibers (white arrow, one fletching arrows) and in large bundles (white arrows, three fletchings). Colocalization was also observed close to blood vessels (blue arrows) of the dura mater. In the male and female MCA (Figure  2C,F ), TRPM3 was localized to smooth muscle cells (blue arrows) and endothelial cells (pink arrows), whereas CGRP‐positive fibers (white arrows, one fletching), originating from the TG, were located near the adventitia, albeit with weaker staining. TRPM3 expression was consistently observed in smooth muscle cells of the artery in both sexes. In the immunofluorescence staining of TRPM3 and CGRP in the layers of the human dural artery from male (Figure  2G–I ) and female (Figure  2J–L ) donors, we observe that TRPM3, indicated by green fluorescence, is distributed throughout the arterial wall, prominently in the adventitia, media, and intima layers. CGRP‐positive nerve fibers, visualized in red and highlighted by arrows, are concentrated in the adventitial layer, with some fibers extending into the media. Although some regions show yellow/orange fluorescence in the merged images, suggesting potential overlap, the primary observation is the close spatial proximity of CGRP fibers to TRPM3‐expressing structures. These observations suggest that TRPM3 activation may promote CGRP release throughout the TGVS, potentially influencing blood vessels both through fiber‐to‐artery interactions and via direct effects on the vascular tissue itself. There are no apparent differences in the distribution or density of CGRP fibers between sexes, emphasizing a consistent pattern of anatomical organization across the samples. In the CGRP release assays (Figure  3 ), we set out to investigate whether TRPM3 channel activation by CIM0216 (a TRPM3 agonist) induces CGRP release and whether sex‐specific differences modulate this response. Capsaicin was included as a positive control to confirm the capacity of the tissues to release CGRP upon stimulation. CGRP release from rat dura mater and TG in male and female animals in response to CIM0216, capsaicin, and isosakuranetin. Panels A–D show CGRP release from male and female dura mater (A, B) and TG (C, D) following stimulation with 10 μM CIM0216 and 100 nM capsaicin. CIM0216 increased CGRP release in male dura and in both male and female TG but not in female dura. Capsaicin increased CGRP release in all tissues. Panels E–H show CGRP release from the same tissues following stimulation with 100 μM CIM0216 and 100 nM capsaicin. At this higher concentration, CIM0216 induced CGRP release in all tissues from both sexes. Panels I–L show CGRP release after pretreatment with vehicle or 10 μM isosakuranetin, followed by coapplication with 100 μM CIM0216. In dura mater (I, J), isosakuranetin did not inhibit CIM0216‐induced CGRP release in either sex. In TG (K, L), isosakuranetin reduced CGRP release in both sexes from CIM0216 stimulation. Data are presented as mean ± SEM with individual data points shown. * p  < 0.05 versus baseline (ratio t ‐tests); # p  < 0.05 vs. vehicle + 100 μM CIM0216 (paired t ‐tests). CGRP, calcitonin gene‐related peptide; L, left; R, right; SEM, standard error of the mean; TG, trigeminal ganglia. [Color figure can be viewed at wileyonlinelibrary.com ] In the low‐concentration experiments, the effects of CIM0216 (10 μM) and capsaicin (100 nM) on CGRP release were evaluated in both dura mater and TG of male and female Sprague–Dawley rats (Figure  3A–D ). In the dura mater of male rats (Figure  3A ), CIM0216 (10 μM) induced CGRP release, with baseline levels of 8.6 ± 3.9 pg/mL and posttreatment values of 19.8 ± 3.6 pg/mL ( n  = 7; t  = 2.7, df = 6, p  = 0.038). Capsaicin induced CGRP release as expected, 36 with baseline levels of 14.7 ± 1.7 pg/mL and poststimulation values of 104.9 ± 14.0 pg/mL ( n  = 7; t  = 8.2, df = 6, p  < 0.001). In the dura mater of female rats (Figure  3B ), CIM0216 (10 μM) did not result in a change from baseline levels (18.6 ± 3.4 pg/mL vs. 16.1 ± 5.0 pg/mL, n  = 5; t  = 1.0, df = 4, p  = 0.366), whereas capsaicin induced CGRP release, with baseline levels of 17.5 ± 4.0 pg/mL and poststimulation values of 93.2 ± 28.3 pg/mL ( n  = 5; t  = 4.2, df = 4, p  = 0.014). Following TRPV1 activation, the CGRP response ratio (poststimulation/baseline) was 3.3 in males and 3.1 in females. In the TG of male rats (Figure  3C ), CIM0216 (10 μM) resulted in an increase in CGRP release, with baseline levels of 28.5 ± 3.4 pg/mL and poststimulation values of 52.5 ± 6.5 pg/mL ( n  = 11; t  = 2.9, df = 10, p  = 0.016). Capsaicin induced CGRP release, with baseline levels of 24.2 ± 3.6 pg/mL and poststimulation values of 78.8 ± 13.6 pg/mL ( n  = 13; t  = 4.2, df = 12, p  = 0.001). In female rats (Figure  3D ), CIM0216 (10 μM) induced CGRP release, with baseline levels of 23.6 ± 2.4 pg/mL and poststimulation values of 51.7 ± 9.1 pg/mL ( n  = 11; t  = 32.7, df = 10, p  = 0.022). Capsaicin induced CGRP release, with baseline levels of 21.1 ± 2.7 pg/mL and poststimulation values of 66.6 ± 11.8 pg/mL ( n  = 10; p  = 0.002). This experiment shows that 10 μM CIM0216 induces CGRP release in the TG of male and female rats, and in the dura mater of male rats but not female rats, whereas 100 nM capsaicin induces CGRP release in both males and females in all tissues. Following TRPV1 activation, the CGRP response ratio (poststimulation/baseline) was 7.4 in males and 5.4 in females. In the high concentration experiments, the effects of 100 μM CIM0216 and 100 nM capsaicin on CGRP release were evaluated in both dura mater and TG of male and female Sprague–Dawley rats. In the dura mater of male rats (Figure  3E ), 100 μM CIM0216 induced CGRP release, with baseline levels of 17.4 ± 3.7 pg/mL and poststimulation levels of 50.6 ± 9.2 pg/mL ( n  = 6; t  = 7.1, df = 5, p  < 0.001). Capsaicin also induced CGRP release, with baseline levels of 21.5 ± 6.2 pg/mL and poststimulation levels of 63.9 ± 12.8 pg/mL ( n  = 6; t  = 2.7, df = 5, p  = 0.042). In the dura mater of female rats (Figure  3F ), 100 μM CIM0216 induced CGRP release, with baseline levels of 14.7 ± 3.7 pg/mL and poststimulation levels of 74.6 ± 11.7 pg/mL ( n  = 6; p  = 0.031). Capsaicin also induced CGRP release, with baseline levels of 16.7 ± 2.7 pg/mL and poststimulation levels of 134.3 ± 21.5 pg/mL ( n  = 7; t  = 12.7, df = 6, p  < 0.001). Following TRPM3 activation, the CGRP response ratio (poststimulation/baseline) in the dura mater was 2.9 in males and 5.0 in females. Following TRPV1 activation, the CGRP response ratio (poststimulation/baseline) was 2.6 in males and 6.1 in females. In the TG of male rats (Figure  3G ), 100 μM CIM0216 induced CGRP release, with baseline levels of 23.1 ± 3.5 pg/mL and poststimulation levels of 67.5 ± 12.8 pg/mL ( n  = 7; t  = 2.9, df = 6, p  = 0.029). Capsaicin also induced CGRP release, with baseline levels of 41.1 ± 13.4 pg/mL and poststimulation levels of 134.5 ± 37.8 pg/mL ( n  = 6; t  = 4.9, df = 5, p  = 0.005). In the TG of female rats (Figure  3H ), 100 μM CIM0216 induced CGRP release, with baseline levels of 27.9 ± 4.8 pg/mL and poststimulation levels of 87.4 ± 12.7 pg/mL ( n  = 6; t  = 5.9, df = 5, p  = 0.002). Capsaicin also induced CGRP release, with baseline levels of 13.9 ± 2.8 pg/mL and poststimulation levels of 85.3 ± 29.2 pg/mL ( n  = 7; t  = 3.6, df = 6, p  = 0.011). Following TRPM3 activation, the CGRP response ratio (poststimulation/baseline) in the TG was 2.9 in males and 3.1 in females. Following TRPV1 activation, the CGRP response ratio (poststimulation/baseline) was 2.9 in males and 8.0 in females. This experiment shows that 100 µM CIM0216 induces CGRP release in both the dura mater and TG of male and female rats. Although the study was not designed or powered for sex comparisons, pairing left–right hemiskulls across sexes is not feasible, and samples were assayed on different ELISA plates (potential batch effects), the data suggest that female dura may exhibit greater CGRP release than male dura. Because of its more consistent ability to elicit release, the 100 µM concentration was used for specificity experiments, whereas 10 µM produced more variable responses. In the antagonistic experiments, in male ( n  = 4) and female ( n  = 6) rat dura mater (Figure  3I,J ), stimulation with 100 μM CIM0216 increased CGRP release compared to baseline (males: from 12.2 ± 5.2 to 33.3 ± 8.4 pg/mL, t  = 3.4, df = 3, p  = 0.043; females: from 9.1 ± 0.9 to 56.5 ± 9.0 pg/mL, t  = 6.4, df = 5, p  = 0.005). Coapplication of 10 μM isosakuranetin did not reduce CIM0216‐induced CGRP release in either males or females (males: 25 ± 4.7 pg/mL, t  = 0.8, df = 3, p  = 0.461; females: 48.6 ± 7.6 pg/mL, t  = 0.6, df = 5, p  = 0.561), showing potentially no antagonistic effect in the dura mater. The CGRP response ratio (poststimulation/baseline) was 1.8 in males and 3.7 in females. In the TG, CIM0216 stimulation increased CGRP release in both males ( n  = 10, from 33.4 ± 4.2 to 58.0 ± 6.2 pg/mL, t  = 2.4, df = 9, p  = 0.042) and females ( n  = 5, from 25.5 ± 4.3 to 61.1 ± 7.3 pg/mL, t  = 7.4, df = 4, p  = 0.003). Isosakuranetin reduced CGRP release following CIM0216 stimulation (males: 44.1 ± 4.0 pg/mL, t  = 3.1, df = 9, p  = 0.013; females: 45.4 ± 5.0 pg/mL, t  = 4.5, df = 4, p  = 0.011), indicating antagonism by isosakuranetin in TG tissues. The male data showed a borderline deviation from normality (Shapiro–Wilk p  = 0.042), but the Q–Q plot did not indicate substantial departure from a normal distribution (the nonparametric analysis gave p  = 0.0625). The CGRP response ratio was 1.5 in males and 1.7 in females. Addition of the TRPA1 antagonist A‐967079 only partially inhibited CGRP release induced by 100 μM CIM0216 in the male dura mater, with no effect observed in the female dura mater (Figure  S1 ). This suggests that the CIM0216‐induced response is not primarily mediated by TRPA1. A‐967079 demonstrated TRPA1 specificity because it effectively inhibited CGRP release induced by 100 μM supercinnamaldehyde in both the dura mater and TG of male rats (Figure  S2 ). The vasodilatory effects of the TRPM3 agonist CIM0216 were evaluated using myograph techniques in rats (Figure  4 ), comparing responses between male (Figure  4A ) and female groups (Figure  4B ) in the MCA. The potency of CIM0216 was expressed as half maximal effective concentration (EC 50 ) values in μM for the controls, with the negative logarithm of the EC50 (pEC 50 ) values and confidence intervals provided in brackets. There was no difference between male controls (12.8 μM, pEC 50  = 4.9 [5.3 to 4.6]) and female controls (15.0 μM, pEC 50  = 4.8 [5.1 to 4.5]), as indicated by overlapping 95% confidence intervals. Vasodilation induced by CIM0216 in male and female MCA in the presence of vehicle, isosakuranetin, or fremanezumab. Cumulative concentration‐response curves to CIM0216 are shown for male (A) and female (B) MCA precontracted with U46619 (10 −7  M). Responses were measured in the presence of vehicle, isosakuranetin (TRPM3 antagonist), or fremanezumab (CGRP receptor monoclonal antibody). No differences in vasodilatory responses were observed between conditions in either male or female arteries. Data are expressed as a percentage of precontraction (100%), with each point representing the mean ± SEM ( n  = 5). CGRP, calcitonin gene‐related peptide; MCA, middle cerebral artery; SEM, standard error of the mean; TRPM3, transient receptor potential melastatin‐3. [Color figure can be viewed at wileyonlinelibrary.com ] When the TRPM3 channel antagonist isosakuranetin was added, there was a difference observed between males (pEC 50  = 5.2 [5.4 to 4.9]) and females (pEC 50  = 4.6 [4.8 to 4.4]), but the response did not show a deviation from control values. This suggests that isosakuranetin did not potentiate or alter the vasodilatory effect of CIM0216, although a slight variation between sexes was noted. To explore whether CGRP mediated the vasodilatory response, fremanezumab (1 μM), an antibody that binds CGRP, was utilized. 42 The results showed no difference between males (pEC 50  = 5.0 [5.3 to 4.7]) and females (pEC 50  = 4.7 [4.9 to 4.4]), suggesting that fremanezumab did not affect CIM0216‐induced vasodilation. In arteries treated with 100 μM CIM0216, the average vasodilation of the MMA was 86.5% ± 5.8% ( n  = 3) for females and 77.8% ± 11.3% ( n  = 3) for males. Collectively, the vascular responses to CIM0216 appear to act directly on the arteries and not via a CGRP dependent mechanism. There were no statistically significant differences between male and female responses under any condition, as indicated by overlapping 95% confidence intervals. Because no significant effect was observed with TRPM3 inhibition, a TRPA1 inhibitor (A‐967079) was applied. Neither A‐967079 alone nor in combination with isosakuranetin significantly altered the vasodilatory potency of CIM0216, indicating limited modulatory effect on TRPA1‐mediated responses to CIM0216 in the MCA (Figure  S3 ). The data presented in Figure  5 show the effects of subcutaneous injection of 3 mg/kg CIM0216 or vehicle on periorbital and hindpaw withdrawal thresholds in male and female Sprague–Dawley rats. None of the rats exhibited observable reactions to the injections. In male rats, two‐way ANOVA showed a significant effect of time ( F (3, 120) = 4.124, p  = 0.008) but no significant effect of treatment ( F (1, 120) = 0.0103, p  = 0.909) and no interaction between treatment and time ( F (3, 120) = 0.278, p  = 0.841). (Figure  5A ), Post hoc comparisons showed that withdrawal thresholds were similar at baseline for both the vehicle group (208.4 ± 7.1 g) and the 3 mg/kg CIM0216 group (200.2 ± 5.4 g, p  = 0.524). At 1‐h postinjection, thresholds for the vehicle group were 174.7 ± 9.8 g ( p  = 0.055) and 171.1 ± 11.4 g ( p  = 0.137) for the CIM0216 group. No differences were observed between groups at any time point, including at 3 h (vehicle: 182.8 ± 10.7 g, CIM0216: 190.3 ± 9.6 g; p  = 0.557) or 24 h (vehicle: 190.5 ± 6.8 g, CIM0216: 191.9 ± 9.8 g; p  = 0.916). Periorbital and hindpaw withdrawal thresholds following systemic administration of CIM0216. Withdrawal thresholds were assessed in male (A, C) and female (B, D) rats ( n  = 16 per group) in a crossover design at baseline and 1, 3, and 24 h following systemic administration of 3 mg/kg CIM0216 or vehicle. Periorbital thresholds (A, B) and hindpaw thresholds (C, D) are shown. In females, periorbital thresholds were reduced at 1‐ and 3‐h postinjection in the presence of CIM0216, whereas no changes were observed with vehicle. No effects were observed in males or in the hindpaw thresholds of either sex. Data represent mean ± SEM. A two‐way ANOVA was used for the statistical analysis. ANOVA, analysis of variance; SEM, standard error of the mean. [Color figure can be viewed at wileyonlinelibrary.com ] In female rats (Figure  5B ), two‐way ANOVA showed significant effects of both time ( F (2.602, 78.05) = 3.714, p  = 0.019) and interaction between time and treatment ( F (3, 90) = 2.842, p  = 0.042) but no main effect of treatment ( F (1, 30) = 0.323, p  = 0.574). Post hoc comparisons indicated that baseline thresholds did not differ between groups (vehicle: 239.7 ± 10.8 g vs 3 mg/kg CIM0216: 227.6 ± 9.2 g; p = 0.403). In the vehicle‐treated group, thresholds decreased at 1 h (202.6 ± 14.8 g; p = 0.009 vs baseline) and 24 h (197.9 ± 13.8 g; p = 0.021 vs baseline). In the CIM0216‐treated group, thresholds were unchanged relative to baseline at 1 h (228.8 ± 11.4 g; p = 0.999) and 24 h (221.7 ± 12.3 g; p = 0.987). At 3 h, there was no difference between groups (vehicle: 213.0 ± 15.0 g vs CIM0216: 210.5 ± 15.6 g; p = 0.909). The hindpaw withdrawal thresholds are shown in male (Figure  5C ) and female animals (Figure  5D ). There were no differences observed between the vehicle‐ and CIM0216‐treated groups at any time point for either sex. Baseline thresholds were consistent, and no changes were detected following injection (see Table  S1 for more details). Whereas CGRP release data suggested a potential involvement of TRPM3, the extent of its contribution remained insufficiently resolved. To further clarify this, we performed calcium imaging in TG from female mice, enabling selective identification of CGRP‐expressing neurons (Figure  6 ). A comparable distribution of TRPM3 and CGRP expression across neuronal subtypes was also observed in the TG of female transgenic mice (Figure  S4 ). We then applied the TRPM3 agonist CIM0216 and the antagonist isosakuranetin to assess the temporal dynamics of TRPM3 activity by analyzing intracellular calcium responses in this neuronal population. Three neuronal populations were analyzed for comparative purposes: TG treated with vehicle ( n  = 530 neurons, three biological replicates), CIM0216 alone ( n  = 291 neurons, three biological replicates), and CIM0216 coapplied with isosakuranetin ( n  = 462 neurons, three biological replicates). Due to limitations in our calcium imaging setup, 30 μM was the highest concentration achievable for a 1:10 dilution because it had to be prepared in an aqueous solution. Therefore, both 3 μM and 30 μM CIM0216 were used in our experiments. Calcium imaging of CGRP‐expressing neurons in female TG reveals TRPM3‐dependent activation by CIM0216 and inhibition by isosakuranetin. (A, B) Representative calcium imaging fields (top) and corresponding heatmaps (bottom) showing normalized fluorescence intensity (Δ F / F min ) over time in response to vehicle, CIM0216 (3 μM or 30 μM), isosakuranetin (10 μM), or their coapplication. Arrows indicate the time of compound application. (A) CIM0216 induced a concentration‐dependent increase in intracellular calcium, whereas vehicle had no effect. (B) Isosakuranetin alone did not induce calcium responses and reduced responses when coapplied with CIM0216. (C–F) Traces of individual neuron calcium responses (light lines) and the average response (black line) to 3 μM CIM0216 (C), 30 μM CIM0216 (D), 3 μM CIM0216 + 10 μM isosakuranetin (E), and 30 μM CIM0216 + 10 μM isosakuranetin (F). (G) Quantification of the proportion of CGRP‐expressing neurons responding to each treatment. Bars represent the percentage of responding (colored) versus nonresponding (white) neurons. Asterisks (*) indicate increase in responders compared to vehicle; dollar signs ($) indicate reduction with isosakuranetin cotreatment; number signs (#) indicate difference between 3 and 30 μM CIM0216 (Fisher's exact test, p  < 0.05, n  = 9). CGRP, calcitonin gene‐related peptide; TG, trigeminal ganglia; TRPM3, transient receptor potential melastatin‐3. [Color figure can be viewed at wileyonlinelibrary.com ] In the CIM0216‐treated group, background activity accounted for 0.69% of neuronal responses. Application of 3 μM CIM0216 increased activity (Figure  6G ) to 35.7% ( p  < 0.001 vs. background), exceeding the response observed with vehicle control (3%, p  < 0.001). This response was characterized by a transient rise and fall in cytosolic calcium levels, as visualized in the heatmaps and fluorescence images (Figure  6A,C ). Coapplication with isosakuranetin (Figure  6B,E,G ) reduced the response to 6.1% ( p  < 0.001 vs. 3 μM CIM0216 alone). Increasing the CIM0216 concentration to 30 μM (Figure  6A,D,G ) further elevated activity to 56.7% ( p  < 0.001 vs. 3 μM), which also exceeded vehicle control (1%, p  < 0.001). In the presence of isosakuranetin, the response at 30 μM CIM0216 (Figure  6G ) was reduced to 35.7% ( p  < 0.001 vs. 30 μM CIM0216 alone), accompanied by a delay in the calcium transients (Figure  6B,F ). These findings suggest that TRPM3 directly activates CGRP‐expressing neurons in a transient, concentration‐dependent manner. Whereas isosakuranetin is effective at suppressing responses at lower concentrations of CIM0216, it is less effective at higher concentrations but alters the temporal profile of calcium signaling.

Our findings provide further insight into the potential role of the TRPM3 channel in migraine pathophysiology. TRPM3 colocalizes with CGRP in neurons and fibers, and its activation leads to activation of CGRP neurons, which leads to CGRP release in TGVS tissues, supporting a functional interaction relevant to migraine mechanisms. TRPM3 activation also induces vasodilation in arteries innervated by the TGVS, where TRPM3 is expressed in smooth muscle cells, suggesting a direct vascular contribution. Although subcutaneous CIM0216 did not produce allodynia‐like effects in rats, the evidence of CGRP release and vasodilation supports TRPM3 involvement in migraine signaling. TRPM3 may offer a broader therapeutic target than CGRP alone, particularly for patients not responsive to CGRP‐directed therapies. Future studies should examine TRPM3 activation by endogenous ligands (e.g., PS and sex hormones) and assess its expression during different migraine and hormonal phases. Unlike TRPV1 antagonists, which have caused adverse effects such as hyperthermia, the TRPM3 antagonist BHV‐2100 has completed phase I trials without notable side effects, supporting its potential as a migraine therapeutic.

This study demonstrates that TRPM3 is expressed in migraine‐relevant sensory and vascular tissues and is functionally capable of inducing CGRP release in trigeminal structures. Activation of TRPM3 led to concentration‐dependent CGRP release in both TG and dura mater. Both sexes were investigated in ex vivo release and myograph assays, and female tissue was selected for calcium imaging to prioritize the clinical relevance of hormonally modulated migraine. TRPM3‐positive fibers were observed in proximity to vasculature in rat and human dura mater, and TRPM3 expression was also detected in vascular smooth muscle of MCA and MMA. Vasodilatory responses to TRPM3 activation were observed independently of CGRP neutralization. Systemic administration of CIM0216 did not elicit mechanical hypersensitivity in rats, although experimental limitations may have contributed. TRPM3‐mediated neuronal activation was confirmed by calcium imaging, demonstrating functional excitation of CGRP‐expressing neurons. These findings support a role for TRPM3 in modulating sensory and vascular components of the TGVS, with relevance to hormonally influenced migraine. To elucidate the potential interplay between the TRPM3 channel and CGRP, we visualized their spatial relationship using immunohistochemistry. In the TG of both male and female rats (Figure  2A,B ), TRPM3 was localized in large neuronal cell bodies. Additionally, small and medium‐sized neurons coexpressed both TRPM3 and CGRP. This observation aligns with previous studies demonstrating colocalization in the TG, 24 as well as in various cell types across different tissues. 29 , 43 These data imply that TRPM3, expressed in neuronal cell bodies, could contribute to the regulation of CGRP release. In both rat and human dura mater (Figure  2C,D,G–L ), CGRP‐positive and TRPM3‐positive fibers were observed to run in parallel. Although this spatial association was relatively sparse across the dura mater, it was more evident near arterial structures, suggesting increased potential for coactivity in perivascular regions. A similar spatial relationship between TRPV1 and CGRP has previously been reported. 18 We also observed TRPM3 immunoreactivity in smooth muscle cells of both the rat MCA and the human MMA (Figure  2E–L ), consistent with earlier studies. 29 These findings suggest that TRPM3 activation may promote CGRP release throughout the TGVS and may influence vascular targets through both fiber‐to‐artery signaling and direct effects on vascular smooth muscle cells. Although this study focused primarily on TRPM3‐induced CGRP release, the known localization of CGRP receptor complexes on cranial artery smooth muscle cells 44 raises the possibility of interaction or convergence between TRPM3‐driven signaling and CGRP receptor‐mediated responses. Specifically, calcium‐dependent intracellular signaling triggered by TRPM3 activation may overlap with or modulate the pathways downstream of CGRP receptor stimulation in smooth muscle cells. Both the TG and dura mater contain neurons and nerve fibers capable of releasing CGRP. In the TG, CGRP‐producing C‐fibers project centrally and peripherally, 11 whereas in the dura mater, C‐fibers terminate near the MMA and pia mater vessels, 44 forming a site critical for migraine‐associated pain signaling. 45 , 46 To evaluate whether TRPM3 can trigger CGRP release in the TGVS, we stimulated ex vivo tissues with CIM0216. This agonist is more selective and potent for TRPM3 than the endogenous neurosteroid PS 28 and has previously been shown to induce CGRP release in other peripheral systems. 28 , 29 Capsaicin was included as a reference TRPV1 agonist for comparison. 20 , 36 , 47 Both male and female Sprague–Dawley rats were used to explore potential sex‐specific differences, although estrous cycle staging was not performed. Pairing of male and female hemi‐skull samples was not feasible within this experimental design, and tissue samples from each sex were processed using separate ELISA plates. These limitations restrict the ability to interpret biological sex effects with confidence. At 10 μM CIM0216 (Figure  3A–D ), we observed CGRP release from dura mater in male rats, whereas no response was detected in females. In TG, CGRP release occurred in both sexes. These findings were consistent with prior reports showing weak CGRP release from dermal tissue at 10 μM and stronger release at 100 μM CIM0216. 28 Accordingly, 100 μM CIM0216 was tested and elicited strong CGRP release from all tissues, comparable to that induced by 100 nM capsaicin. A trend toward greater CGRP release in female tissues was noted, but due to the aforementioned methodological constraints, it is not possible to determine whether this reflects a true biological difference. Previous studies have shown that PS, the endogenous TRPM3 agonist, can induce CGRP release in noncranial tissue, 28 but its relevance in migraine is unclear. The effects of progesterone on CGRP modulation have also been documented, but the literature presents conflicting data: One study suggested that progesterone can facilitate CGRP release in TGVS tissue, 48 whereas our previous study found no stimulatory effect. 49 These discrepancies may be due to species, hormonal state, or timing of administration, and remain an area of uncertainty. To further characterize CIM0216‐induced CGRP release, we performed additional experiments at 100 μM (Figure  3E–H ). In these experiments, an added vehicle group resulted in longer tissue incubation time. CGRP levels were generally lower, possibly due to time‐dependent degradation or depletion of releasable stores. These results suggest that timing is a critical parameter in CGRP release assays. We next tested whether CIM0216‐induced CGRP release was specific to TRPM3. The flavanone isosakuranetin was selected as a TRPM3 antagonist based on previous studies showing inhibition of TRPM3 activation at 10 μM without affecting cell viability. 28 , 34 At this concentration, isosakuranetin does not activate TRPA1 and is unlikely to inhibit other TRP channels. In TG tissue, isosakuranetin reduced CIM0216‐evoked CGRP release, but it had no effect in dura mater from either sex (Figure  3I–L ). Isoform diversity in TRPM3 may explain these tissue‐specific pharmacological differences. TRPM3α2, expressed in trigeminal neurons, is sensitive to isosakuranetin, whereas TRPM3α1 for example is not. 50 These differences may account for the failure of isosakuranetin to block CIM0216‐induced responses in dura mater and arteries, despite its effect in neuronal tissues. TRPM3 exhibits considerable splice variant diversity, 51 and the α2 isoform, which is sensitive to isosakuranetin, is expressed in TG, 52 whereas other isoforms such as TRPM3α1 or other nonsensitive variants may predominate in dura mater. It remains unclear whether isosakuranetin acts through competitive or noncompetitive inhibition. Previous reports indicate that the compound shows strong inhibition at lower CIM0216 concentrations but reduced efficacy at higher doses. 28 Other antagonists such as ononetin may offer broader isoform coverage. 34 To assess the contribution of TRPA1 to CIM0216‐induced CGRP release, we coapplied the selective TRPA1 antagonist A‐967079 with 100 μM CIM0216 (Figure  S1 ). In TG and female dura mater, A‐967079 had no significant effect. However, in male dura mater, a minor but statistically significant reduction in CGRP release was observed, indicating a partial involvement of TRPA1 in this response in males. The specificity of A‐967079 was validated by its ability to inhibit CGRP release induced by the TRPA1 agonist supercinnamaldehyde (Figure  S2 ). 53 These findings suggest that while TRPM3 is the primary mediator of CIM0216‐induced CGRP release, TRPA1 makes only a minor contribution in the male dura mater, with no detectable contribution in TG or in female dura. To examine the vascular effects of TRPM3 activation, we tested the ability of CIM0216 to induce vasodilation in ex vivo segments of the MCA. Vasodilation occurred in a concentration‐dependent manner in tissues from both sexes, with no observed difference between male and female arteries. Vasodilation was evident at 10 μM CIM0216, a concentration that had minimal effect on CGRP release in the dura mater. These results suggest that CIM0216‐induced vasodilation is not solely mediated by CGRP. To test whether CGRP contributed to the response, we used fremanezumab, a CGRP‐sequestering antibody. Fremanezumab had no effect on CIM0216‐induced dilation, even though it has been shown to inhibit TRPV1‐ and TRPA1‐induced vasodilation in similar systems. 54 Isosakuranetin also failed to reverse CIM0216‐induced vasodilation (Figure  4 ). This may be explained by the potential expression of TRPM3α1 in vascular smooth muscle, which isosakuranetin does not effectively inhibit. 50 A‐967079 was also tested but did not reduce CIM0216‐induced vasodilation in MCA, either alone or in combination with isosakuranetin (Figure  S2 ). These results indicate that CIM0216‐induced vasodilation likely involves a mechanism independent of CGRP and TRPA1. One plausible explanation involves TRPM3‐mediated calcium influx triggering local calcium sparks in vascular smooth muscle cells, which could activate large‐conductance potassium channels. Similar pathways have been demonstrated for TRPV1 55 and may account for vasodilation at subthreshold concentrations for CGRP release. To assess potential behavioral effects, we tested whether systemic administration of 3 mg/kg CIM0216 induced mechanical hypersensitivity. This dose was the highest deliverable concentration based on the vehicle formulation. No effect on periorbital or hindpaw sensitivity was observed at any time point in either sex (Figure  5 ). In female rats, a modest increase in sensitivity was noted at 1‐h post‐vehicle injection (Figure  5B ), which may reflect natural variability or handling‐related effects. 56 The lack of behavioral response may be due to limited systemic bioavailability because no pharmacokinetic data are currently available for CIM0216. Systemic administration of CIM0216 was used to assess rodent behavior, expanding on previous research that primarily focused on organ‐specific effects or local injections, which have been shown to induce nocifensive behaviors in rats, albeit locally and within a shorter time frame. 57 , 58 Furthermore, many preclinical migraine models use direct dural application to induce migraine‐relevant hypersensitivity, 39 , 40 , 59 suggesting that systemic delivery may be suboptimal. The absence of allodynia in this model does not exclude TRPM3 involvement in migraine because not all patients exhibit allodynia. Importantly, established migraine treatments such as fremanezumab or onabotulinumtoxinA do not consistently reverse mechanical hypersensitivity in all animal models. 39 , 40 To confirm that TRPM3 activation leads to functional neuronal responses, we performed calcium imaging in CGRP‐reporter mice. Application of CIM0216 at 3 μM and 30 μM caused increases in cytosolic calcium in CGRP‐expressing TG neurons. The response was dose‐dependent, with earlier and broader activation at the higher concentration (Figure  6A,C,D,G ). This TRPM3‐induced calcium influx is presumed to promote CGRP release through calcium‐dependent exocytosis, consistent with known mechanisms of neuropeptide secretion. Coapplication of 10 μM isosakuranetin strongly attenuated the 3 μM response and partially inhibited the 30 μM response (Figure  6B,E–G ), supporting the involvement of TRPM3 under both conditions. The calcium transients observed were rapid and reversible, with clear temporal kinetics corresponding to agonist application (Figure  6A,C,D ), indicating that TRPM3 activation directly triggers excitatory neuronal responses in CGRP‐expressing trigeminal cells. The transient nature of the calcium signal suggests rapid channel gating, but the duration and downstream signaling remain to be clarified in future studies.

During the preparation of this work, the authors used ChatGPT to improve readability and language. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Migraine is a multifactorial neurological disorder affecting approximately 15% of the global population, with a higher prevalence observed in women compared to men at a ratio of 3:1. 1 It is associated with a substantial reduction in quality of life and characterized by recurrent episodes that may include unilateral pulsating headache, nausea, vomiting, photophobia, phonophobia, and prolonged attack duration of up to 72 h. 2 Whereas the etiology of migraine remains to be fully elucidated, advancements in research have provided valuable insights into its pathophysiology. The sensory neuropeptide calcitonin gene‐related peptide (CGRP) plays a key role in migraine pathophysiology by inducing vasodilation, neurogenic inflammation, and the modulation of pain mechanisms upon binding to its receptor. 3 , 4 Medications targeting CGRP or its receptor have demonstrated efficacy in alleviating migraine symptoms in approximately 50% of patients. 4 , 5 The pathology of migraine is believed to involve both the central nervous system (CNS) and the peripheral nervous system, with key tissues implicated being the brainstem and the trigeminovascular system (TGVS). 6 , 7 , 8 Migraine treatments are thought to alleviate pain by targeting peripheral components of the TGVS, particularly meningeal blood vessels such as the middle meningeal artery (MMA) and middle cerebral arteries (MCA). 9 These vessels are innervated by fibers originating from the trigeminal ganglion (TG), which contains a dense population of neurons transmitting signals between the periphery and the CNS. 10 , 11 Transient receptor potential (TRP) channels, a diverse family of cation channels, have been implicated in migraine, particularly the TRP vanilloid 1 (TRPV1), TRP ankyrin 1 (TRPA1), and TRP melastatin 8, due to their presence in the TGVS and in the brainstem. Activation of these channels results in CGRP release, potentially contributing to periorbital pain in rodent models upon injection into the dura mater. 12 TRP channels have been implicated in various sensory functions, including heat and cold sensation, itch, and pain transmission. 13 Among these channels, TRPV1, which detects heat and acidic conditions, has been extensively studied in the context of pain and migraine. TRPV1 is abundantly expressed in TGVS tissue, where it is often colocalized with CGRP in neuronal cell bodies and C‐fibers. 14 , 15 , 16 , 17 , 18 , 19 When activated using capsaicin, TRPV1 channels can trigger CGRP release from peripheral nerve endings. 20 , 21 However, targeting TRPV1 with the antagonist SB‐705498 for migraine treatment has proven ineffective in a phase II clinical trial, 22 suggesting that TRPV1 alone is insufficient as a therapeutic target or that its role in migraine pathophysiology is minimal. Recent attention has shifted toward the transient receptor potential melastatin 3 (TRPM3) channel, which shares the nociceptive capacity of other TRP channels and is localized in migraine‐relevant tissues. 23 , 24 , 25 Of particular interest is the modulation of the TRPM3 channel by hormones, specifically progesterone and estrogen. Its most potent endogenous ligand is pregnenolone sulfate (PS), 26 which may play a role in menstrual migraine. 27 Activation of the TRPM3 channel has been shown to cause CGRP release in nonmigraine relevant tissue 24 , 28 and to be colocalized 24 with the neuropeptide in TG. Activating the TRPM3 channel has been observed to cause vasodilation of human dermal arteries; however, olcegepant—a CGRP receptor antagonist—was unable to inhibit this vasodilation. 29 Targeting the TRPM3 channel could be more effective in treating migraine in women because higher spiking activity has been observed in the trigeminal nerve of female mice compared to male mice. 23 Although progesterone and estradiol may modulate TRPM3 channel activity, their effects (stimulatory or inhibitory) are still debated. At physiological conditions, these hormones appear to have a partial agonistic effect on the channel. 26 Interestingly, TRPM3 gene expression has been shown to increase during the proestrus phase in mice, 30 whereas in patients with endometriosis, TRPM3 expression tends to be elevated during the follicular and luteal phases. 31 In both cases, the expression levels are increased when hormone levels are high and fluctuate. Interestingly, the TRPM3 channel antagonist BHV‐2100 has passed a phase I clinical trial without showing any notable adverse events 32 and has advanced to phase II clinical trials for migraine treatment, further highlighting the increased interest for this channel in migraine research (ClinicalTrials.gov ID NCT06603623 ). Further investigation into the role of the TRPM3 channel in migraine pathophysiology may yield additional mechanistic insight and support its evaluation as a potential therapeutic target. Clarifying the interaction between TRPM3 activation and CGRP signaling could contribute to a more detailed understanding of migraine biology and inform the development of targeted pharmacological interventions. Ex vivo, in the TGVS we investigated the localization of the TRPM3 channel and its association with CGRP, the effect on CGRP release, its vasoactive properties, and its ability to increase cytosolic calcium in CGRP neurons. In vivo, we set out to inject TRPM3 agonist subcutaneously in rats to explore if this administration could indicate a putative involvement of TRPM3 on allodynia. In our study, we selected CIM0216 as the TRPM3 agonist due to its greater potency and selectivity compared to PS, 28 which also acts on N‐methyl‐D‐aspartate receptors and may introduce confounding effects. 33 Similarly, isosakuranetin was chosen as the TRPM3 antagonist because it has been shown to be the most potent and subtype‐specific inhibitor in trigeminal neurons. 34 Given the higher prevalence of migraine in women compared to men, and evidence suggesting greater TRPM3 channel reactivity in females, both male and female rats were included in this study to account for potential sex‐specific differences.

Philip V. Reducha , Lukas K. S. Nielsen , Mette N. Jensen , Jacob C. A. Edvinsson , Spyridoula Kazantzi , Sofia L. Wæver , Tanja Lylloff , Connar S. J. Westgate , Lars Edvinsson , and Kristian A. Haanes declare no conflicts of interest pertaining to the submitted work.

Figure S1. Figure S2. Figure S3. Figure S4. Table S1.