Animals

The C57BL/6 J mice of either sex were used to establish the nitroglycerin-induced CM mouse model. For chemogenetic approaches, the SOM-Cre mice (stock no. 013044, The Jackson Laboratory) and PKC-δ-Cre mice (stock no. 011559, MMRRC) were used. Mice of 2–5 months old of either sex were used for all the experiments. All mice were bred in the C57BL/6 J genetic background. Mice were housed on a 12-h light-dark cycle and given food and water ad libitum. All behavioral tests were conducted after at least a 4-day habituation period. Animals were handled in accordance with the national and institutional guidelines. All behavioral procedures were conducted in accordance with the protocol approved by the Institutional Animal Care and Use Committee (IACUC) of the National Yang Ming Chiao Tung University.

Viral vectors for chemogenetic studies

To specifically express designer receptors exclusively activated by designer drugs (DREADDs) onto SOM and PKC-δ positive neurons, a recombinant adeno-associated virus serotype 5 (rAAV5) carrying hM4Di or hM3Dq conjugated to mCherry in a double-floxed inverted open reading frame (DIO), driven by human Synapsin I (hSyn) promoter (rAAV5-hSyn-DIO-hM4Di-mCherry or rAAV5-hSyn-DIO-hM3Dq-mCherry), was used. In addition, a virus carrying the red fluorescent protein (rAAV5-hSyn-DIO-mCherry) was used as the control. All viral vectors were purchased from the Vector Core at the University of North Carolina (Chapel Hill, NC, USA) or Addgene Vector Core (Watertown, MA, USA).

Stereotaxic virus injection

PKC-δ-Cre or SOM-Cre mice of 2–3 months old of either sex were injected with viruses in the CeA for chemogenetic studies. Mice were deeply anesthetized with isoflurane (4% induction, 1.5%–2% maintenance in oxygen, vol/vol; Halocarbon Laboratories, USA) and positioned in a stereotaxic injection frame (IVM-3000; Scientifica, Uckfield, UK). The coordinates of bilateral injection sites of the CeA were AP: − 1.31 mm, ML: ±2.87 mm, DV: − 4.82 mm. During all surgical procedures, mice were kept on a homeothermic pad (Physitemp Instrument, USA or TMP-5b, Supertech Instruments, Hungary) to maintain their body temperature at 34–37 °C. After securing the head with ear bars, the eyes of the mice were protected by the ophthalmic gel. For the virus injection, 0.35 μl/side of virus were bilaterally injected into the CeA using a 10 μl NanoFil syringe (World Precision Instruments, USA) and a 34-G beveled metal needle, controlled by the nano-pump controller (0.1 μl/min, KD Scientific, USA). After 10 min for virus distribution, the needle was withdrawn slowly. All animals were rendered to recover for at least 3 weeks before the behavioral tests for maximal viral expression.

Stereotactic or cannula microinfusion of calcitonin gene-related peptide receptor antagonist into CeA

To block the PBN CGRP neurotransmission to the CeA, we applied the CGRP receptor 1 antagonist, CGRP fragment 8–37 (HY-P0209, MedChemExpress), dissolved in normal saline at 1.8 μg/μl [19], into bilateral CeA with either acute stereotactic microinfusion or intermittent infusion via chronically implanted cannulas. The C57BL/6 J mice of 2–3 months old were deeply anesthetized with isoflurane (4% induction, 1.5%–2% maintenance in oxygen, vol/vol; Halocarbon Laboratories, USA) and positioned in a stereotaxic injection frame (IVM-3000; Scientifica, Uckfield, UK). The coordinates of bilateral stereotactic injection or cannulas implantation sites of the CeA were AP: − 1.31 mm, ML: ±2.87 mm, DV: − 4.82 mm. For acute stereotactic application, 0.5 μl/side (900 ng/side) of CGRP8–37 were bilaterally injected into the CeA using a 10 μl NanoFil syringe (World Precision Instruments, USA) and a 34-G beveled metal needle, controlled by the nano-pump controller (0.1 μl/min, KD Scientific, USA). After 10 min for CGRP8–37 distribution, the needle was withdrawn slowly. All animals were rendered to recover for at least 30 mins before the behavioral tests. For chronic intermittent CGRP8–37 application, we implanted cannulas to bilateral CeA for drug infusion. After aligning, additional screws were driven onto the cranium for stabilizing the cannulas. Then, the guide cannula (62,004, OD 0.41 mm × ID 0.25 mm, RWD Life Science, China) and dummy cannula (62,104, OD 0.2 mm, RWD Life Science, China) were inserted into the designated position and the components were fixed with dental cement (Super-bond C&B kit, Sun medical, Japan). During all surgical procedures, mice were kept on a homeothermic pad (Physitemp Instrument, USA or TMP-5b, Supertech Instruments, Hungary) to maintain their body temperature at 34–37 °C. During the surgical procedures, the eyes of the mice were protected by the ophthalmic gel. All animals were rendered to recover for at least 1 week before the behavioral tests. For each injection, 0.5 μl/side (900 ng/side) of CGRP8–37 were bilaterally injected into the CeA using a 50 μl syringe (80,975, Hamilton, USA), controlled by the dual-channel syringe infusion pump (0.1 μl/min, Fusion 100, CHEMYX, USA). After 2 min for CGRP8–37 distribution, the internal cannula was withdrawn slowly.

CTB-594 retrograde tracing

For retrograde labeling the target circuits in the central nervous system, we bilaterally injected 0.35 μl CTB-594 (C22842, Thermo, Invitrogen) into CeA using a 10 μl NanoFil syringe (World Precision Instruments, USA) and a 34-G beveled metal needle, controlled by a nano-pump controller (0.1 μl/min, KD Scientific, USA). After 10 min for CTB-594 distribution, the needle was withdrawn slowly. All animals were allowed to recover for at least 1 week before the behavioral tests.

Nitroglycerin-induced chronic migraine model

Previously studies mostly employed nitroglycerin (NTG) with a dose of 10 mg/kg per injection to model acute or chronic migraine [20, 21]. However, this dose is 1000–10,000 times higher than that used clinically, which caused a drastic and prolonged of the blood pressure drop [22]. In contrast, a previous study has shown that a naturalistic dose of NTG, roughly 8 times higher than the reasonable pharmacological dose used clinically, is sufficient to induce allodynia in rats, without casting doubts that the animal behaviors could be confounded by the abnormal hemodynamic alterations [23, 24]. Thus, in our study, we tested multiple doses of NTG (10 mg/kg, 1 mg/kg and 0.1 kg/mg) to explore the minimal doses required to induce cranial and hindpaw allodynia in mice with least hemodynamic impact. After identifying the ideal dose (0.1 mg/kg) (see results), the mice were intraperitoneally (i.p.) injected with NTG once to mimic acute migraine or every second day for 9 days (i.e., a total of 5 NTG injections) to simulate chronic migraine. The periorbital and hindpaw mechanical thresholds of the mice were tested before and 2 hours after NTG administration on each test day. NTG solution was prepared from a stock solution of 2 mg NTG in 99% propylene glycol (#1466506, Sigma) and freshly diluted in 0.9% saline to a dose of 0.1, 1 and 10 mg/kg. The vehicle controls in these experiments contain 1% propylene glycol diluted in 0.9% saline, used as vehicle compares to that used for three different NTG doses. All injections of NTG were administered as a 0.1, 1 and 10 mg/kg volume via i.p. injection. To avoid confounding, the drugs or vehicles were prepared in a blinded fashion. To evaluate the predictive validity of the model, the migraine specific treatment sumatriptan (#S1198, Sigma), a 5-HT1B/1D agonist, was freshly dissolved in normal saline and i.p. injected at a concentration of 0.6 mg/kg, 5 mins after the NTG injection. The mechanical thresholds were only tested on the 1st, the 9th and the 10th day. These mice were similar to that in mice receiving repeated measurement in other groups [21].

Immunohistochemistry and immunofluorescence

The mice were deeply anesthetized with isoflurane and perfused through the left ventricle to the body circulation with phosphate buffered saline (PBS, 0.9% NaCl in 0.01 M phosphate buffer, pH 7.4) followed by 4% paraformaldehyde (PFA) in the PBS. The mice brain and TG were rapidly removed and post-fixed in the 4% PFA overnight at 4 °C following previously established protocols [25,26,27]. Further, the solution was replaced with 15% sucrose overnight at 4 °C, followed by 30% sucrose overnight at 4 °C. The brain was then embedded in the O.C.T. solution at frozen state. Coronal brain and sagittal TG sectioning with the thickness of 50 μm [25] and 20 μm [28, 29] respectively were sliced by cryostat microtome (LEICA, CM1900, Germany) throughout the whole brain. Sections were washed with 0.1% Tween 20 in Tris-buffered saline (TBS) 5 min each free-floatingly for 3 times and then treated with 3% H2O2 in TBS for 10 min and washed by TBS again. Sections were blocked with 2% bovine serum albumin (BSA, Sigma) and 2% normal goat serum (NGS, Vector Laboratories) in TBS for 1 hour at room temperature, followed by overnight incubation with the anti-pERK1/2 antibody (1:500, #4370, Cell Signaling Technology), anti-CGRP (1:500, sc-57,053, Santa Cruz) at 4 °C and then replaced by biotinylated goat anti-rabbit antibody (1:500, Invitrogen) for 1 hour at room temperature. Further, the sections were incubated in the avidin-biotin complex reagent (ABC, Vector Laboratories). The DAB kit (Vector Laboratories) was the final step for counterstaining of the sections. For the immunofluorescence studies, the sample were overnight incubated with the anti-pERK1/2 antibody (1:500, #4370, Cell Signaling Technology), anti-CGRP (1:500, sc-57,053, Santa Cruz), anti-CALCRL (1:100, HPA008070, Sigma), anti-SOM (1:100, sc-74,556, Santa Cruz), and anti-PKC-δ (1:100, 610,398, BD Biosciences) at 4 °C and then incubated with secondary antibody Goat anti-rabbit Alexa 488 and Goat anti-mouse Alexa 594 (1:500, Invitrogen) for 2 hours at room temperature. After washing three times with TBS, slices were mounted onto slides using Vectashield mounting medium containing 4′,6-diamidino-2- phenylindole (DAPI, H-1500, Vector Laboratories,).

Immunoblotting

The CeA isolated from brain was homogenized on ice in the RIPA buffer (Sigma) supplemented with cocktail inhibitors protease. Five microgram of protein was submitted to SDS-polyacrylamide gels 10% and transferred onto a PVDF membrane (Bio Rad). After blocking with 5% BSA, the membrane was incubated overnight at 4 C° with primary anti-pERK1/2 antibody (1:1000, #4370, Cell Signaling Technology), and anti-ERK1/2 antibody (1:1000, #9102, Cell Signaling Technology). Blots in the membrane were probed with a horseradish peroxidase coupled secondary antibody anti-mouse IgG (1:2000, #7076, Cell Signaling Technology) or anti-rabbit IgG (1:2000, #7074, Cell Signaling Technology). The enhanced chemiluminescence substrate (ECL, Pierce™ ECL Western Blotting Substrate, Invitrogen) was applied for visualization and the image was captured by luminescence imaging system (LAS-4000, Fujifilm).

Behavioral tests

All behavioral tests were conducted in the light period of the cycle, and both male and female mice were tested in this study. Mice were transferred to the behavior room with dim light at least 30 min for habituation. Animals were randomized to experimental or control groups. For the most behavioral and biochemical tests, the experimenters were blinded and randomized to the treatment information. However, due to the specific consideration for breeding the GENSAT BAC transgenic PKC-δ-Cre mice, which should be utilized in the hemizygous state. Thus, some of the behavioral and biochemical tests of the PKC-δ-Cre mice were not randomized. In chemogenetic experiments, after a 3-week recovery from surgery, mice expressing DREADD receptors were i.p. injected with clozapine N-oxide (CNO, 5 mg/kg, Sigma). The CNO was dissolved in 0.9% NaCl with 10% dimethyl sulfoxide (DMSO). The vehicle control also contained 10% DMSO in 0.9% NaCl solution. The CNO was freshly dissolved in normal saline and i.p. injected at a concentration of 5 mg/kg for both hM4Di and hM3Dq groups. Behavioral tests were performed 1 hour (for hM4Di) [30] or 2 hours (for hM3Dq) after the drug administration. The detailed procedures of each behavioral test were as follows:

von Frey filament test

To determine the periorbital and hind paw mechanical pain of the mice, a von Frey filament test was used. All of the von Frey tested animals were under at least a 4-day habituation period. For hind paw mechanical pain threshold measurement, mice were habituated for at least 30 min before the test. A series of von Frey filaments (0.04–1 g, Touch-Test, USA) were applied to the wire mesh onto the plantar surface of both hind paws in an up-down testing paradigm. A withdrawal response was considered valid once the hind paw removed completely from the platform. For each paw, a von Frey filament was applied five times at 5-sec intervals. The threshold was determined when paw withdrawal was observed in more than three of five applications [31]. Animals were tested for basal responses immediately before i.p. injection with NTG and vehicle control. After 75 min of the NTG and vehicle control application, animals were habituated for another 45 min, and the 2-hour post-treatment responses for mechanical sensitivity were tested. For chronic experiments, test was conducted every second day over 9 days after the drug administration (5 test days total). The test was continued for one or two further weeks without any drug administration to evaluate the sustained effect of chronic NTG treatment. As for periorbital mechanical pain measurement, the mice were habituated in the 13 × 28 × 13 cm cage 30 min before the test. A von Frey filament of 0.4 g force was applied to the periorbital area rostral to the eyes and near the midline 12 times at approximately 90° angle. The responses were recorded and scored as follows: uni- or bilateral forepaw swipes across the face (1 point), aggression/biting of the filament following stimulus (0.25 points) or clear withdrawal of the head from the stimulus (0.25 points). The points (accumulated in the 12 trials) were summed for each animal separately for each testing time to give the overall response score [32, 33].

Marble burying test

The testing apparatus contains 6 cm depth of beddings in the 13 × 28 × 13 cm cage. Twenty-four glass marbles (about 1.5 cm of diameter) were evenly distributed on the bedding spaced 3 cm between each marble. Mice were placed individually into the testing apparatus for 30 min. Buried marbles was defined as at least two-third of their surface embedded into the bedding.

Light/dark box (L/D box) test

After injection of NTG or vehicle drugs, mice were individually tested in a L/D box, which consists of bright and dark compartments within the same apparatus. Mice were placed individually in the middle of the light (house light, ~ 100 lx) compartment of the box and allowed to access freely to the entire apparatus for 10 min. All behaviors of mice were monitored by Tru-scan 2.0 system (Coulbourn instruments, USA). The transition between two chambers, total locomotor activity and total time spent in the two chambers were analyzed.

Elevated Plus Maze (EPM) test

The EPM apparatus consists of two opposite open arms (30 × 5 cm) without walls surrounded by a 0.5 cm-high edge and two arms of the same dimensions enclosed by 25 cm high walls that were elevated to a height 50 cm from the floor. Animals were placed onto the center platform of the maze facing an open arm and allowed to search the maze for 10 min then return to their cages. The behavioral parameters including the percentage of time spent in both arms, the percentage of time spent in the center and total traveling distance were measured with the video tracking software EthoVision XT 13 (Noldus Information Technology, USA).

Corticosteroid measurement

Following the behavior tests, the submandibular blood (cheek punch) was collected by a disposable lancet (5 mm, Goldenrold Animal Lancet, MEDIpoint, USA). In total, 0.1–0.5 ml of blood were quickly drawn without anesthesia [34]. The blood sample was centrifuged at 1900×g at 4 °C for 10 min, and the separated serum was stored at − 20 °C until further analysis. The concentration of serum corticosterone was quantified using an enzyme-linked immunoassay based commercial kit (Enzo Life Sciences, ADI-900-097, USA).

Non-invasive blood pressure measurement

The non-invasive blood pressure sphygmomanometer for mice (NIBP System, ADINSTRUMENTS, USA), in conjunction with a data acquisition device (PowerLab System, ADINSTRUMENTS, USA) was used. This device includes a specialized tail cuff and pulse transducer, used for intermittent mouse blood pressure measurement based on the periodic occlusion of tail blood flow in unanesthetized mice. To restrain mice without anesthesia, mice were fixed to warmed stages with rodent restrainers. The baseline blood pressure was measured before and 2 hrs after the vehicle and 0.1, 1 and 10 mg/kg NTG i.p. injections.

Statistics

The immunohistochemistry and immunofluorescence data were quantified by Image J (NIH, USA). The immunoblotting data were quantified by ImageQuant TL (GE healthcare, USA). The behavioral data were analyzed by Prism 7.0 (GraphPad Software, USA). Normality test was performed before analysis. The data were analyzed by two-way repeated measures ANOVA followed by Bonferroni post hoc test and independent t-test if they passed normality test. Otherwise, the data were analyzed with non-parametric tests such as the Friedman tests with Dunn’s post hoc test and Mann-Whitney-U test. Data were presented as mean ± SEM. Significance levels set at p < 0.05 *, p < 0.01 **, and p < 0.001 ***.