Abstract

Introduction:

Chronic jaw muscle pain is a common clinical condition whose etiology remains ill-defined. Using acidic saline injections into the masseter muscle to mimic it, we examined the hypothesis that hyperexcitability of jaw closing muscles spindle afferents (MSA) that have previously been observed in this model result from neuron glia interactions in the trigeminal mesencephalic nucleus (NVmes) lead to activation of nociceptive pathways.

Methods:

This was assessed using whole-cell patch-clamp recordings from NVmes neurons combined to pharmacological and astrocytic optogenetic stimulations and immunohistochemistry against cFos in the ventrolateral pole of the subnucleus interpolaris/caudalis transition region (vl-Vi/Vc), and GFAP in vl-Vi/Vc and NVmes regions in rats and mice

Results:

Acidic saline injection into the masseter muscle led to increases in: (1) cFos expression in vl-Vi/Vc at 9 days after the injection, (2) reactivity of astrocytes in NVmes, and (3) Excitability of NVmes neurons that manifested spontaneously or in response to astrocytic stimulation. This increased activity is thought to result from the release of the astrocytic Ca2+-binding protein S100β, since it was not observed in S100β knock-out mice, which also did not show increased expression of cFos in vl-Vi/Vc, despite showing increased reactivity of NVmes astrocytes.

Discussion:

These findings suggest that acidic saline injection into the masseter muscles induced long-term activation of astrocytes in the NVmes and promoted ectopic firing of NVmes neurons via astrocyte-released S100β, and subsequent activation of nociceptive pathways.

1 Introduction

Chronic muscle pain, called myofascial pain syndrome, not only causes suffering to patients but is also recognized for its association with high unemployment rates among these patients (Landmark et al., 2013) and the substantial healthcare burden they impose (Gaskin and Richard, 2012). Consequently, it has been highlighted that addressing chronic muscle pain should be a high priority (Breivik et al., 2013), yet its causes and etiology are still poorly understood.

To mimic the myofascial pain and delayed-onset muscle soreness (DOMS) following eccentric or intensive exercise, Sluka et al. (2001) developed a model of muscle pain induced by injecting acidic saline into the gastrocnemius muscle of rats. This model is considered useful for understanding myofascial pain syndrome because pain persists even after the muscle tissue pH returned to normal and because this type of soreness closely resembles the reports of patients suffering from chronic jaw muscle pain, low back pain, and other muscle pain (Hood et al., 1988).

Several studies suggest that acid-sensing ion channels (ASICs) play an important role in DOMS (Fujii et al., 2008; Gazerani and Cairns, 2018; Lee et al., 2025). These are expressed in both small-diameter PA and MSA in muscles (Lin et al., 2016; Gazerani and Cairns, 2018; Lee et al., 2025), but interestingly, acid-induced hyperalgesia persisted in the conditional knockout of ASIC3 in small-diameter PAs mice, but not in the conditional knockout of ASIC3 in MSA (Lee et al., 2025), indicating the involvement of proprioceptors in acid-induced chronic myofascial pain rather than nociceptors. Other lines of evidence suggesting that these large-diameter PAs, whose activity is normally not related to pain, may also contribute to chronic pathological pain include the facts that: (1) mechanical, but not thermal, hypersensitivity develops normally after nerve injury in mice in which transmission from nociceptors is specifically prevented genetically (conditional VGluT2 knock-outs) (Scherrer et al., 2010), (2) specific activation or large diameter PAs with vibration or light mechanical stimuli produces pain after DOMS or peripheral tissue or nerve injury (Zhu and Henry, 2012; Colloca et al., 2017), and (3) In humans, as in several animal neuropathic pain models, pain onset coincides with appearance of ectopic firing in these afferents (Chul Han et al., 2000; Liu et al., 2000; Khan et al., 2002; Tal et al., 2006). However, besides the observation that specific stimulation of MSA with vibration at 80 Hz increased muscle pain after DOMS in the triceps surae muscle of humans (Weerakkody et al., 2001) it was unknown whether excitability changes leading to ectopic firing occur only after nerve lesion or could also appear after milder damage such as those producing muscle soreness and pain.

Chronic jaw muscle pain is one type of chronic myofascial pain syndrome. Acute jaw muscle pain, often described as inflammation or injury of jaw muscles, is experienced by 10% of the population (LeResche, 1997) and becomes chronic in some cases where it persists after extinction of inflammation. Lund et al. (2010) adapted Sluka’s model to the jaw closing muscles to further investigate mechanisms potentially underlying chronic jaw muscle pain and found that injection of acidic saline into the masseters (jaw closing muscles) of rats induced mechanical hypersensitivity for up to 5 weeks afterwards. Interestingly, this change in mechanical sensitivity was paralleled by changes in excitability of neurons in the trigeminal mesencephalic nucleus (NVmes) (Lund et al., 2010), which contains the somata of jaw closing MSA. They postulated that the ectopic firing resulting from this hyperexcitability could travel antidromically in the peripheral branches of the MSA and induce glutamate release in the spindle capsule. The released glutamate could then activate nociceptors’ free endings found in the vicinity of these release sites. They further provided anatomical evidence that nociceptive endings carrying glutamatergic metabotropic receptors (mGluR5) can be found closely opposed to peripheral endings of MSA containing the glutamate vesicular transporter VGlut1; thereby supporting the hypothesis.

Firing of NVmes neurons always emerges from subthreshold membrane oscillations (SMOs) (Verdier et al., 2004; Wu et al., 2005; Enomoto et al., 2007; Gaudel et al., 2025) which rely on a sodium persistent current (INaP) (Wu et al., 2001, 2005; Enomoto et al., 2007; Gaudel et al., 2025) whose amplitude increase with reduction of extracellular Ca2+ concentration ([Ca2+]e) (Gaudel et al., 2025). Astrocytes are known to release the calcium-binding protein S100β, which by decreasing [Ca2+]e enhances INaP-mediated changes of excitability of nearby neurons (Morquette et al., 2015; Ryczko et al., 2021). Gaudel et al. (2025) applied these findings to NVmes in mice and assessed whether astrocytes and their S100β can produce such a decrease in [Ca2+]e in NVmes (Gaudel et al., 2025). They showed that local applications of S100β or BAPTA (a Ca 2+ chelator that decreases [Ca2+]e) near the stem axon, where the channel mediating INaP is highly concentrated, caused ectopic firing. These effects were reproduced by optogenetic stimulation of astrocytes near the stem axon and were mediated by the release of S100β and INaP activation.

Since many studies reported activation of astrocytes (as detected by markers of reactivity) (Okada-Ogawa et al., 2009; Borghi et al., 2023; Yang et al., 2026) in conditions of pathological pain, we hypothesized that the hyperexcitability and ectopic firing of NVmes neurons associated with the acid-induced mechanical hypersensitivity result from astrocytic reactivity and release of S100β.

Here, using immunohistochemistry against the activity marker cFos (cellular oncogene fos) and astrocytic reactivity marker GFAP (glial fibrillary acidic protein) and electrophysiological recordings of NVmes neurons, we first validated that findings previously obtained in rats could be reproduced in mice to enable use of transgenic mice lines engineered to allow manipulation of astrocytes. Our results confirm findings previously reported by Lund et al. (2010) in rats and suggest that S100β may be involved in the observed effects (Lund et al., 2010).

2 Methods2.1 Animals

All experiments were conducted according to the Canadian Institutes of Health Research rules and were approved by the Animal Care and Use Committee of Université de Montréal.

A total of 12 rats and 60 mice were used, including 30 wild type (WT) mice (C57BL/6 J, Stock 000664, JAX), 21 mice expressing the channelrhodopsin 2 (ChR2) under the control of the GFAP promoter (GFAP-ChR2-EYFP mice) and 9 S100β null mice (S100β KO mice) (Table 1). GFAP-ChR2-EYFP mice were produced by crossing GFAP-Cre (B6. Cg-Tg(Gfap-cre)73.12Mvs/J, stock 12,886, JAX;105) and ChR2-lox mice [B6. Cg-Gt(ROSA)26Sortm32(CAG COP4∗H134R/EYFP)Hze/J, stock 24,109, JAX;106]. The S100β null mice were obtained by crossing heterozygous B6Brd; B6N-Tyrc-Brd S100btm1a(EUCOMM)Wtsi/WtsiCnbc mice (Wellcome Trust Sanger Institute) and selecting the homozygous offspring. Experiments were performed using both sexes, and sex was not factored into the analysis, because a clinical study suggests that there is no difference between sexes in chronicisation of temporomandibular disorders (Nguyen et al., 2019).

Animal groupTreatmentExperimentNeutral saline (CTL)Acidic saline (PAIN)ImmunohistochemistryElectrophysiologyRats (N = 12)66120WT mice (N = 30)1515723GFAP-ChR2-EYFP mice (N = 21)813021S100β KO mice (N = 9)4563

ChR2, channelrhodopsin 2; EYFP, enhanced yellow fluorescent protein; GFAP, glial fibrillary acidic protein; KO, knock-out; N, number of animals; WT, wild type.

Mice and rats were separated into 2 groups randomly (Table 1). One group (CTL, control group) was injected with 20 μL saline (pH 7.4) into both sides of the masseter muscle. The other group (PAIN, pain group) was injected with 20 μL acidic saline (pH 4). Following the protocol established by Lund et al. (2010), the injections were repeated 2 days later, that is, at postnatal day 8 and 10 in rats (12–22 g), postnatal days 7 and 9 in mice (around 4–5 g) for electrophysiological experiments, between postnatal 70 and 100 days in mice (17–25 g) for immunohistology experiments under isoflurane anesthesia. A blind person conducted injections throughout all experiments.

2.2 Immunohistochemistry

Rats were perfused 5 and 9 days after the second injection. Mice were subjected to the same procedure at 9 days post-injection. Stimulation of masseter muscles on both sides was conducted, 20 times using the 15 g Von-Frey filament by a blind experimenter under light urethane anesthesia, 90 min before the perfusion. The rats and mice were then deeply anesthetized again with urethane and perfused intracardially with saline, followed by 4% paraformaldehyde in phosphate-buffered saline (PBS) on the respective day. The brains were quickly removed and post-fixed overnight with 4% paraformaldehyde at 4 °C. For brain sectioning, the brains were first separated between the NVmes and the trigeminal spinal nucleus levels. For sectioning of the trigeminal spinal nucleus in rats at 9 days after the 2nd injection, the brains were submerged in 20% sucrose PBS overnight after post-fixation, and sectioned coronally at 50 μm using a cryostat (Leica CM3050 S, Leica Biosystems, Nussloch, Germany). For the other parts of the brains in rats and mice, the brains were embedded in 3% agarose just before sectioning and were sectioned coronally at 80 and100 μm at the NVmes level of mice and rats, respectively and 50 μm at the trigeminal spinal nucleus level of both rats and mice using a vibratome (Vibratome 1,000, Technical Products International Inc., St. Louis, MO, USA). For immunostaining, all steps were carried out at room temperature unless specified otherwise. Tissue sections were rinsed with PBS containing 0.5% Triton-X (Fisher Scientific, Waltham, MA, USA) and removed from embedded agarose or 20% sucrose-PBS, followed by a 120-min incubation in a blocking solution containing 5% normal donkey serum and 0.5% Triton X-100 in PBS. This solution was also utilized for all antibody dilutions. Sections were then incubated overnight at 4 °C in primary antibodies (see Table 2). Then, secondary antibodies were applied for 2 h (see Table 2). All sections were mounted on glass slides (Fisherbrand Superfrost Plus, Fisher Scientific, Waltham, MA, USA). In the negative controls, the primary antibodies were omitted, and in each case, there was no detectable labelling. Within a few days, the sections were observed and captured under the Colibri 7 solid-state LED light source (Carl Zeiss, Oberkochen, Germany).

AntibodiesSourceIdentifierGuineaPig anti-ParvalbuminSynapticSystemsCat: #195308, RRID: 1–9
Cat: #195308, RRID: 1–10Chicken anti-GFAPAbcamCat: #AB4674, RRID: 1118849–3
Cat: #AB4674, RRID: 1012209–1Mouse anti-NeuNMilliporeCat: #MAB377, RRID: 3832727Rabbit anti-cFosAbcamCat: #AB190289, RRID: GR3393951-1Donkey anti-Chicken Alexa Fluor 488Jackson ImmunoResearchCat: #703–545-155, RRID: 170532Donkey anti-Rabbit Alexa Fluor 488Jackson ImmunoResearchCat: #711–545-152, RRID: 172897Donkey anti-Mouse Alexa Fluor 594Jackson ImmunoResearchCat: #715–585-151, RRID: 144883Donkey anti-GuineaPig Alexa Fluor 594Jackson ImmunoResearchCat: #706–585-148, RRID: 170762Donkey anti-GuineaPig Alexa Fluor 647Jackson ImmunoResearchCat: #706–605-148, RRID: 158548Donkey anti-chicken Alexa Fluor 594Jackson ImmunoResearchCat: #703–585-155, RRID: 1117942.3 GFAP image analysis

GFAP which is often used as a marker of astrocyte activation (Okada-Ogawa et al., 2009; Borghi et al., 2023; Fonseca-Rodrigues et al., 2026; Yang et al., 2026) was first measured in the caudal part of NVmes containing the highest densities of proprioceptors’ somata. These can be easily identified by parvalbumin immunostaining (Gaudel et al., 2025). Images of the NVmes were acquired with a 20x objective. Two sections were used in rats, while only one at a comparable level was used in mice (due to the smaller size of their NVmes). In each section, measurements were made within a Region of Interest (ROI; Figure 1A) of 150 μm x 400 μm in rats, and 60 μm x 250 μm in mice (again due to the smaller size of their NVmes), on each side of the section. The ROIs were positioned by a blind experimenter and with great care to avoid inclusion of the Locus coeruleus and Parabrachial nucleus adjacent to NVmes. Data from the left and right sides were analyzed as independent samples, and in the cases having two sections per animal (rats), the data were averaged for each side. The astrocytes that were running along blood vessels were excluded from analysis.

Histological analysis. (A) Left: Photomicrograph of a representative brainstem section at low magnification showing the NVmes nucleus (white square). The scale bar represents 500 μm. Middle: GFAP-IR astrocytes (green) in the NVmes. NVmes neurons are immunoreactive to parvalbumin (magenta). The scale bar represents 100 μm. Right: A skeletonized version of the middle image processed using the Fiji plugin. The yellow box indicates the ROI. The graph on the right shows the total length of GFAP-IR astrocyte processes within the ROI. (B) Left: Photomicrograph of a representative brainstem section at low magnification showing the vl-Vi/Vc (in the white square). The scale bar represents 500 μm. Middle: Examples of cFos-IR cells in the vl-Vi/Vc (yellow ellipse); (magenta: NeuN; green: cFos). The scale bar represents 100 μm. Right: Auto-detected cFos-IR cells processed using the Zen software. The yellow ellipse indicates the vl-Vi/Vc area (ROI). The graph on the right shows the total number of cFos-IR cells within the ROI. cFos-IR, cellular oncogene fos-immunoreactive; GFAP-IR, glial fibrillary acidic protein-immunoreactive; NeuN, neuronal nuclei; NVmes, trigeminal mesencephalic nucleus; ROI, region of interest; vl-Vi/Vc, ventrolateral pole of the subnucleus interpolaris/caudalis transition.

GFAP measurements were also made in the ventrolateral pole of the subnucleus interpolaris/caudalis transition region (vl-Vi/Vc) which is known to play an important role in deep orofacial pain (Wang et al., 2006; Nakatani et al., 2018), vl-Vi/Vc sections were taken at the level of the obex. Images were acquired at 40x magnification for rats and 64x for mice. As vl-Vi/Vc is a large region and all acquired images were considered to belong to vl-Vi/Vc, the entire acquired image (312 μm × 312 μm for rats, and 198 μm × 198 μm for mice) area was analyzed as the ROI.

Skeletonization of the GFAP staining was performed using the Fiji distribution of ImageJ (NIH, USA) according to the method by Marques et al. (2023). Briefly, the images were converted to binary images by applying a threshold calculated from the histogram of the image fluorescence intensity using the MaxEntropy algorithm, after image processing by the method of Marques et al. (2023). Subsequently, the Skeletonize plugin of ImageJ was utilized, followed by the elimination of GFAP-immunoreactive (GFAP-IR) astrocyte segments shorter than 5 μm. Finally, the total length of GFAP-IR astrocytes residing within the ROI was calculated (Figure 1A).

2.4 cFos image analysis

Images of the vl-Vi/Vc were acquired with a 20x objective. cFos-immunoreactive (cFos-IR) cell counts in the vl-Vi/Vc region were automatically quantified using ZEN software (Carl Zeiss, Oberkochen, Germany) (Figure 1B). The automatically extracted cells were confirmed, and any label clearly not regarded as cFos-IR cells were excluded by a blind person.

2.5 Brainstem slice preparation

Using a VT1000S vibratome (Leica), coronal brainstem slices (350 μm) were prepared from mice aged 13–36 days. The control and pain groups were age-matched, more or less 2 days, and most often were from the same litter. Following anesthesia by isoflurane (Pharmaceutical Partners of Canada Inc., Richmond Hill, ON, Canada) inhalation, the mice were decapitated and their brain were rapidly extracted and sectioned in an ice-cold modified artificial cerebrospinal fluid (CSF, in mM: 3 KCl, 1.25 KH2PO4, 4 MgSO4, 26 NaHCO3, 10 Dextrose, 0.2 CaCl2, 219 Sucrose, pH 7.3–7.4, 300–320 mOsmol/kg) saturated with a mix of 95% O2 and 5% CO2. After sectioning, the slices were kept at room temperature in a holding chamber filled with artificial CSF (in mM: 124 NaCl, 3 KCl, 1.25 KH2PO4, 1.3 MgSO4, 26 NaHCO3, 10 Dextrose, and 1.6 CaCl2, pH 7.3–7.4, 294–300 mOsmol/kg) bubbled with 95% O2 and 5% CO2.

2.6 Electrophysiology and analysis

The experimenter was blinded to the experimental groups throughout all electrophysiological recordings. For recordings, one slice was transferred to a submerged chamber continually perfused with artificial CSF bubbled with 95% O2 and 5% CO2. Patch microelectrodes (resistance 6–10 MΩ) were pulled from borosilicate glass capillaries (1.5 mm outside diameter, 1.12 mm inside diameter, World Precision Instruments) using a P-97 puller (Sutter Instruments). For neuronal recordings, pipettes were filled with an internal solution containing (in mM): 140 K-gluconate, 5 NaCl, 2 MgCl2, 10 HEPES, 0.5 EGTA, 2 Tris ATP salt, 0.4 Tris GTP salt, pH 7.2–7.3, 280-300mOsmol/kg. Alexa Fluor 488 or 594 was added to the internal solution to visualize neuronal and axonal morphologies during the experiment. We used an Olympus Fluoview FV 1000 confocal microscope equipped with a 40x (N. A. 0.80) water immersion objective for imaging. All recordings were performed using a Multiclamp 700A amplifier, Digidata 1322A interface coupled to a computer equipped with pClamp 8 or 11 software (Molecular Devices, San Jose, CA). The pipette resistance and capacitance were compensated electronically. Neurons were discarded when action potentials did not overshoot 0 mV or when the resting membrane potential was depolarized (> − 45 mV). Standard scripts in Clampfit were used for analysis. The input resistance was determined as the slope of the linear part of the current–voltage (I-V) curve.

2.7 Optogenetic stimulation

Two lasers (440 and 488 nm) were used simultaneously in the SIM lightpath of an FV1000 microscope (Olympus) for optogenetic stimulation of the astrocytes in GFAP-ChR2-EYFP mice. The SIM scanner was used in normal scanning mode to photoactivate manually delineated small areas surrounding the recorded neuron. Optogenetic stimulations were applied using 30 s pulses (10–20% laser power/8.6–14.9 mW for laser 440 nm/8.7–15.9 mW for laser 488 nm).

2.8 Drug application

1,2-bis(o-aminophenoxy)ethane-N, N, N0, N0-tetraacetic acid tetrasodium salt [BAPTA, 5 mM; Sigma-Aldrich (Oakville, Ontario, Canada)] diluted in artificial CSF was locally applied with glass micropipettes (tip diameter around 2 μm) with 2–20 psi pressure pulses of variable duration (1–30 s, Picospritzer III, Parker Instrumentation, Fairfield NJ USA).

2.9 Statistics

Statistical analyses were performed using the commercial software SPSS (IBM, NY, USA). Normality was assessed using the Shapiro–Wilk test with a significance level of α = 0.05. For non-normally distributed data, the Mann–Whitney U test was performed. When normality was confirmed, statistical significance was determined using either Student’s t-test (for equal variances) or Welch’s t-test (for unequal variances), depending on the results of Levene’s test for homogeneity of variance (α = 0.05). For categorical data, Fisher’s exact test was used when expected frequencies were <5. Otherwise, the χ2 test was applied. In all immunohistochemical experiments, two or three animals were analyzed per group for each parameter, and data were analyzed separately for the left and right sides to minimize animal use. In all electrophysiological experiments, to minimize animal use as well, we confirmed that no marked inter-animal differences were observed and therefore recorded multiple NVmes neurons (n: number of cells) from a single mouse (N: number of animals), finally pooling them as samples. All data are presented as mean ± standard error to the mean (SEM) and as proportions (%) in the figures and results sections. The significance of differences was accepted at a p < 0.05.

3 Results3.1 Acidic saline injections into the jaw muscles induce cFos activity in Vc in rats and mice

Our previous work suggested that the long-lasting hyperexcitability of NVmes neurons that paralleled the acid-induced mechanical hypersensitivity could lead to activation of nociceptors free endings within the capsule in the muscle (Lund et al., 2010). Here, to further validate the model, we first tested, using immunohistochemistry against the activity marker cFos, whether acidic-saline injections led to an increase of activity of neurons in the area of the trigeminal spinal nucleus known to be associated to orofacial muscle pain (vl-Vi/Vc) (Wang et al., 2006; Nakatani et al., 2018). This area, located ventrolaterally at the junction of the subnucleus interpolaris and the subnucleus caudalis, is known to receive inputs from nociceptors but not from jaw closing MSA. We assessed cFos expression in rats on days 5 and 9 after the 2nd injection. At 5 days post-injection, the number of cFos-IR cells in the PAIN group is slightly higher than that of the CTL group, although there is no statistical difference (CTL group: 6.5 ± 7.8, PAIN group: 13.7 ± 10.7; Mann–Whitney U test, p = 0.18; Figure 2A left). The trend is maintained, and the difference becomes significant at 9 days post-injection (CTL group: 4.0 ± 1.5, PAIN group: 25.3 ± 13.8; Welch’s t-test, p = 0.02; Figure 2A right). Similar findings were obtained in mice 9 days post-injection, although the difference did not reach significance (CTL group: 16.0 ± 13.9, PAIN group: 34.7 ± 24.5; Mann–Whitney U test, p = 0.18; Figure 2B).

Acidic saline injections increase the number of c-Fos-IR cells in the vl-Vi/Vc. (A) Bar chart comparing the number of c-Fos-IR cells in the vl-Vi/Vc of rats, 5 (left panel) and 9 (right panel) days after the 2nd injection in their masseter muscles, in the CTL (black) and PAIN (red) groups. The PAIN group showed more cFos -IR cells than the CTL group at 9 days after the 2nd injection. (B) Bar chart comparing the number of c-Fos-IR cells in the vl-Vi/Vc of wild-type mice, 9 days after the 2nd injection in their masseter muscles, in the CTL (black) and PAIN (red) groups. Data are represented as mean ± SEM. Diamonds indicate individual data. *p < 0.05 for comparisons between CTL and PAIN groups, Welch’s t-test. cFos-IR, cellular oncogene fos-immunoreactive; vl-Vi/Vc, ventrolateral pole of the subnucleus interpolaris/caudalis transition.

3.2 Acidic saline injections within the jaw muscles cause astrocytes to become reactive in NVmes but not in Vc in rats and mice

The origin and mechanisms underlying the increased excitability and ectopic firing observed in large diameter primary afferent neurons in association to pathological pain are still unidentified (Sas et al., 2023). Since in many pain models, astrocytes have been shown to become reactive (Okada-Ogawa et al., 2009; Borghi et al., 2023; Ahmed et al., 2025) and having demonstrated that optogenetic stimulation of neighboring astrocytes causes firing in NVmes neurons in GFAP-ChR2-EYFP mice (Gaudel et al., 2025), we then sought to determine if astrocytes also become reactive in our acidic saline injection-induced chronic jaw muscle pain model. Total length of GFAP-IR astrocytes in NVmes and vl-Vi/Vc was evaluated at post-injection 5 and 9 days in rats, and post-injection day 9 in mice. The total length of GFAP-IR astrocytes of the PAIN group is significantly larger than that of the CTL group at 5 (CTL group: 464.9 ± 203.1 μm, PAIN group: 1164.8 ± 243.5 μm; Student’s t-test, p = 0.003; Figure 3A left) and 9 days (CTL group: 727.8 ± 108.0 μm, PAIN group: 986.5 ± 200.1 μm; Mann–Whitney U test, p = 0.04; Figure 3A right) in rats and at 9 days post-injection in mice (CTL group: 64.1 ± 48.3 μm, PAIN group: 137.4 ± 50.8 μm; Student’s t-test, p = 0.04; Figure 3B). In contrast to NVmes, total length of GFAP-IR astrocytes in vl-Vi/Vc did not differ between CTL and PAIN groups in both rats (Day5: CTL group: 3220.9 ± 524.1 μm, PAIN group: 2885.1 ± 408.3 μm; Student’s t-test, p = 0.29; Day9: CTL group: 3477.2 ± 333.9 μm, PAIN group: 3394.9 ± 234.2 μm; Student’s t-test, p = 0.66; Figure 4A left and right, respectively) and mice (CTL group: 1071.6 ± 312.7 μm, PAIN group: 1058.8 ± 207.0 μm; Student’s t-test, p = 0.94; Figure 4B).

NVmes astrocytes become reactive following injections of acidic saline in the masseter of rats and mice. (A) Bar chart comparing the total length of GFAP-IR astrocytes within the ROI in the NVmes of rats, 5 (left panel) and 9 (right panel) days after the 2nd injection in their masseter muscles, in the CTL (black) and PAIN (red) groups. ROI was set as a 150 μm x 400 μm rectangle (see Figure 1). (B) Bar chart comparing the total length of GFAP-IR astrocytes within the ROI in the NVmes of Wild-type mice, 9 days after the 2nd injection in their masseter muscles, in the CTL (black) and PAIN (red) groups. ROI was set as a 60 μm x 250 μm rectangle. Data are represented as mean ± SEM. Diamonds indicate individual data. *1P < 0.05 for comparisons between CTL and PAIN groups by Student’s t-test. **p < 0.01 for comparisons between CTL and PAIN groups by Student’s t-test. *2P < 0.05 for comparisons between CTL and PAIN groups by Mann–Whitney U test. GFAP-IR, glial fibrillary acidic protein-immunoreactive; NVmes, trigeminal mesencephalic nucleus; ROI, region of interest.

vl-ViVc astrocytes do not become reactive following injections of acidic saline in the masseters of rats and mice. (A) Bar chart comparing the total length of GFAP-IR astrocytes within the ROI in the vl-ViVc of rats, 5 (left panel) and 9 (right panel) days after the 2nd injection in their masseter muscles, in the CTL (black) and PAIN (red) groups. ROI was set as a 312 μm × 312 μm square (see Figure 1). (B) Bar chart comparing the total length of GFAP-IR astrocytes within the ROI in the vl-ViVc of wild-type mice, 9 days after the 2nd injection in their masseter muscles, in the CTL (black) and PAIN (red) groups. ROI was set as a 198 μm × 198 μm square. Data are represented as mean ± SEM. Diamonds indicate individual data. There were no significant differences between the control and pain groups across all conditions. GFAP-IR, glial fibrillary acidic protein-immunoreactive; ROI, region of interest; vl-Vi/Vc, ventrolateral pole of the subnucleus interpolaris/caudalis transition.

3.3 Acidic saline injections within the jaw muscles increase the excitability of NVmes neurons in mice

Using whole cell patch-clamp recordings, we then test whether NVmes neurons in mice display increased excitability, as in rats, following the same bilateral injection protocol of either normal (CTL group) or acidic (PAIN group) saline within their masseters (Table 1). As in Lund et al. (2010), the animals were sacrificed at various times (1–26 days) following the second injection. One hundred and four neurons recorded in the NVmes nucleus of 23 WT mice and 21 GFAP-ChR2-EYFP mice fulfilled our inclusion criteria. Those neurons were equally distributed between the control and the experimental groups (n = 52 each). They were filled with Alexa Fluor (488 or 594), and all showed the typical pseudo-unipolar morphology of primary sensory afferents. The recorded neurons also showed the typical electrophysiological signature of NVmes neurons consisting of a prominent sag caused by a strong inward rectification upon membrane hyperpolarization (Figure 5A), and their responses to membrane depolarization could still be classified in three distinct firing patterns: spike-adaptative, burst and tonic-firing (Figure 5B, top, middle and bottom traces, respectively). However, there was a significant between-group difference in the distribution of firing patterns (Exact Fisher test, p < 0.001) where a significantly greater percentage of neurons from the PAIN group were of the bursting type as shown in Figure 5C. Additionally, all but one neuron from the PAIN group (98%), fired a rebound action potential at the offset of the hyperpolarizing pulses (Figure 5A, arrowhead) while only 43 (83%) of the control neurons displayed this behavior. The basic electrophysiological characteristics of the recorded neurons are summarized in Table 3. There were no statistical differences in the resting membrane potential (RMP), the firing threshold, and the input resistance between neurons from the CTL and PAIN groups (Figure 5D and Table 3). Only the threshold of the voltage-dependent SMOs was found to be significantly more hyperpolarized in the PAIN relative to the CTL group (CTL group: −44 ± 1.4 mV, PAIN group: −48 ± 0.7 mV; Student t-test, p = 0.01; Figure 5E left and Figure 5F). There were also significantly more neurons in the PAIN group that exhibited SMOs (CTL group: 25%, PAIN group: 48%; X2 = 5.971, df = 1, p = 0.01; Figure 5E right).

Acidic saline injections augment NVmes neurons excitability by decreasing the SMOs threshold and by increasing their incidence. (A) Membrane response (top trace) of an NVmes neuron to injection of a hyperpolarizing current pulse (bottom trace). The arrow and arrowhead point the sag and the rebound action potential, respectively. (B) Membrane responses of a spike-adaptative (top), a burst-firing (middle), and a tonic-firing (bottom) NVmes neuron. (C) Bar chart of the relative distribution of the firing profiles of the NVmes neurons in the control (black) and pain (red) groups. (D) Bar chart comparing the RMP, the firing voltage threshold (left) and the input resistance (right) of the NVmes neurons from the control (black) and pain (red) groups. (E) Left: bar chart comparing the SMOs threshold of the NVmes neurons from the control (black) and pain (red) groups. Right: bar chart of the relative distribution of the oscillating and non-oscillating neurons in the control (black) and pain (red) groups. (F) Membrane responses of an NVmes neuron from the control group (left) and an NVmes neuron from the pain group (right) showing the emergence of SMOs at a more hyperpolarized potential in the neuron from the pain group. The membrane potential and the amount of injected current are given on the left of each trace. Data in (D,E) are represented as mean ± SEM. *P < 0.05, Student t-test. NVmes, trigeminal mesencephalic nucleus; RMP, resting membrane potential; SMOs, subthreshold membrane oscillations.

Electrophysiological characteristicsWT + GFAP-ChR2-EYFP mice
n = 104S100β KO mice
n = 12CTL (n = 52)PAIN (n = 52)CTL (n = 6)PAIN (n = 6)RMP (mV)−54 ± 0.4a−54 ± 0.4a