archives of oral biology 59 (2014) 251–257
Available online at www.sciencedirect.com
ScienceDirect journal homepage: http://www.elsevier.com/locate/aob
Peripheral GABAA receptor activation modulates rat tongue afferent mechanical sensitivity Sun Nee Tan a, Esther Song a, Xu-Dong Dong a,b, Rishi Kumar Somvanshi a, Brian E. Cairns a,c,* a
Faculty of Pharmaceutical Sciences, The University of British Columbia, Vancouver, BC, Canada College of Stomatology, Tianjin Medical University, Tianjin 300071, China c Center for Sensory-Motor Interaction, Faculty of Medicine, Aalborg University, Aalborg, Denmark b
article info
abstract
Article history:
Objectives: The expression of GABAA receptors and the effects of GABAA receptor agonists
Received 30 June 2013
on the response properties of tongue afferent fibres were investigated in female rats to
Received in revised form
determine if peripheral GABA receptors might be a target of topical benzodiazepines when
7 November 2013
used for pain relief in burning mouth syndrome patients.
Accepted 27 November 2013
Design: Nerve fibres in tongue sections from six female rats were identified using protein gene product 9.5, and the co-expression of the g subunit of GABAA receptor and substance P
Keywords:
assessed in the nerve fibres. In vivo extracellular recordings of trigeminal ganglion neurons
-Aminobutyric acid (GABA)
that innervate the tongue were undertaken in 27 anesthetised female rats and their
Burning mouth syndrome
responses to mechanical and thermal stimulation characterised before and after topical
Benzodiazepine
application of GABA, the GABAA receptor selective agonist muscimol or vehicle control.
Tongue
Results: The vast majority of tongue nerve fibres examined (95%) expressed the g subunit of
Mucosa
GABAA receptor. Bath application of muscimol, but not GABA, significantly increased the
Pain
mechanical thresholds of tongue afferent fibres compared to vehicle, but only after the
Nerve fibre
tongue had been heated with 60 8C water. Conclusions: GABAA receptors are present on tongue nerve fibres and their activation alters the mechanical sensitivity these fibres. These findings suggest that topical application of benzodiazepines to the oral mucosa may decrease pain in burning mouth syndrome through a local action on peripheral GABAA receptors. # 2013 Elsevier Ltd. All rights reserved.
1.
Introduction
Burning mouth syndrome is a chronic intraoral pain condition for which no medical or dental cause has yet been identified.1–3 The prevalence of this syndrome in the general population has been estimated to be around 1%.4–9 Burning mouth syndrome is characterised by symptoms of ‘‘burning-like’’ pain most commonly experienced on the anterior of the tongue, as well as the anterior hard palate and lips, although
other sites in the mucosa of the oral cavity and throat can also be involved.3,10–12 Burning mouth syndrome may go into remission after a period of years,13 and thus conservative medical treatment aimed at symptom resolution is the normal therapeutic avenue. Unfortunately, burning mouth syndrome symptoms have proven quite resistant to conventional pharmacotherapy. Recently, it was reported that a non-conventional approach that involved sucking a 1 mg tablet of the benzodiazepine clonazepam, without swallowing, three times a day for 14 days
* Corresponding author. Tel.: +1 604 822 7715; fax: +1 604 822 3035. E-mail address: [email protected] (B.E. Cairns). 0003–9969/$ – see front matter # 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.archoralbio.2013.11.015
252
archives of oral biology 59 (2014) 251–257
was associated with significant symptomatic improvement in pain scores.14 Benzodiazepines are anxiolytic drugs thought to mediate their effects in the central nervous system by increasing the ability of g-amino-butyric acid (GABA) to open the GABAA receptor; a ligand-gated chloride channel. The interaction between benzodiazepines and the GABAA receptor occurs at the interface between the a and g subunits.15–18 Activation of peripheral GABAA receptors in orofacial tissues is not excitatory and appears to exert anti-nociceptive actions in some animal models of acute craniofacial pain.19–21 These findings suggest that topical application of benzodiazepines to the oral mucosa may decrease pain in burning mouth syndrome through a local action on peripheral GABAA receptors to decrease the excitability of nociceptors.14 However, evidence for a direct effect of GABAA receptor activation on nociceptors in the oral cavity is lacking. In the present study, the expression of GABAA receptors in tongue nerve fibres and the effects of GABA and the GABAA receptor selective agonist, muscimol, on the response properties of tongue afferent fibres were investigated in female rats. It was hypothesised that GABAA receptors would be widely expressed by nerve fibres in the tongue and that activation of these receptors would alter the mechanical threshold of these nerve fibres.
2.
Materials and methods
2.1.
Immunohistochemistry
Tongues from six female Sprague–Dawley rats were removed and frozen with liquid nitrogen. The first 5 mm of frozen tongue tissue, measured from the tip of the tongue was cut coronally and cut into 10 mm-thick sections with a cryotome. A total of four sections from each tongue were chosen for analysis, with each section separated from the previous section by a distance of at least 50 mm. The sections were mounted on glass slides and stored at 20 8C. Sections were incubated in 0.2% Triton-X100 and 10% normal goat serum (NGS) in phosphate-buffered saline (PBS) for 1 h at room temperature. The slides were then washed with PBS three times and incubated overnight at 4 8C with primary antibodies. Axonal fibres were identified using rabbit polyclonal antibody against rat protein gene product (PGP) 9.5 (1:2000; Abcam Inc., Cambridge, MA, USA). PGP 9.5 is an axonal marker and served to identify nerve fibres. The g2 subunit of GABAA receptors were labelled with the use of goat polyclonal antibody (1:50, Santa Cruz Biotechnology, Inc., CA, USA). Substance P expression, which identifies neuropeptide containing afferent fibres, was assessed through the use of a guinea pig polyclonal antibody (1:700; Abcam Inc., Cambridge, MA). Alexa Fluor 555 donkey anti-rabbit IgG antibody (1:700; Invitrogen, NY, USA) was used to visualise the PGP 9.5immunoreactivity (IR); Alexa Fluor 488 donkey anti-goat IgG antibody (1:700; Invitrogen, NY, USA) was used to visualise GABAA g2-IR and Alexa Fluor 633 goat anti-guinea pig IgG antibody (1:700; Invitrogen, NY, USA) was used to visualise the substance P-IR. The entire mucosal area of each section was imaged using a Leica TCS SPE confocal microscope. Some tissue sections were
incubated without primary antibodies to assess the degree of non-specific labelling. In addition, hemaglutinin-tagged cDNA for g2 subunit of GABAA receptor was transfected into HEK-293 cells and Western blots run on the protein extract from these cells. The blots revealed that GABAA g2 antibody labelled a single band at 65 kDa corresponding to molecular weight of the GABAA g2 A nerve fibre was defined as any fibril-shaped fluorescence of at least 4.0 mm in length which stained positive for PGP9.5 antibody. Nerve fibres were considered positively-stained for any of the antibodies used when the intensity of the fluorescence signals were more than 2 standard deviations (SDs) above the mean background intensity. The number of neuronal fibres on the mucosae, with or without colocalisation with SP and GABAA Rg2 were counted using Image J software programme (National Institutes of Health Image). Images of the entire tissue sections were captured using a digital camera (Olympus) and areas of the tissue sections were calculated using Image J software programme (National Institutes of Health Image).
2.2.
In vivo electrophysiology
Female Sprague–Dawley rats (n = 27; 260–340 g; Charles River Inc., Wilmington, MA, USA) were used. Rats were prepared to allow recording of trigeminal ganglion neurons under surgical anaesthesia (O2: 0.3–0.4 l/min; isoflurane 2.5–3.0%;).22 A tracheal cannula was inserted to initiate artificial ventilation and the carotid artery was catheterised to measure blood pressure. The rat’s head was placed in a stereotaxic frame, the skin over the dorsal surface of the skull was reflected and a trephination was made on the right side of the skull. A parylene-coated tungsten-recording electrode was lowered into the trigeminal ganglion via the trephination to record action potentials from single trigeminal ganglion neurons.23 A suture was placed in the middle of the tongue to stretch and hold it in place (Fig. 1). In some experiments, a thermal probe was inserted into the tongue to measure changes in tongue temperature. Throughout the entire experiment, heart rate, mean blood pressure and core body temperature were continuously monitored and kept within physiological ranges of 300–400 beats/min, 60–80 mm Hg and 36.8–37.1 8C, respectively. Upon completion of all surgical procedures, the isoflurane level was reduced to 1.5–2.5% to achieve a depth of anaesthesia characterised by continued absence of a toe pinch reflex. All procedures were performed in adherence with the principles of the Canadian Council on Animal Care and were approved by the University of British Columbia Animal Care Committee. All efforts were made to minimise the number of animals used and their suffering. A blunt probe was used to mechanically stimulate the tongue to identify mechanoreceptive afferent fibres (Fig. 1). Upon identifying one or more afferent fibres, a stimulating electrode was placed in contact with the tongue and moved until increasing electrical current (1–10 mA, 0.5 ms duration at 1 Hz frequency) evoked single orthodromic action potentials of fixed latency. Then, an electronic von Frey hair (model 160IC, IITC) was used to determine the baseline mechanical threshold of the tongue afferent fibres. The von Frey hair was applied at one minute intervals for 10 min. The mechanical
archives of oral biology 59 (2014) 251–257
253
Fig. 1 – An electronic Von Frey hair was used to determine the MTs of afferent fibres before and after topical drug applications. A rubber dam sealed the water bath to prevent leaking of drug solutions so that tongue could be immersed in the bath solution for 10 min each time. A suture was used to fix tongue in a tensed position. A. Method to record and calculate mechanical threshold (MT) from a tongue afferent fibre. At one minute intervals throughout the recording, mechanical pressure was applied using the electronic Von Frey instrument to the afferent receptive field until an action potential (AP) (top trace, near B) was fired. The lowest magnitude of mechanical pressure required generate an AP from this sensory fibre was taken as its MT (from bottom bar; value ii minus value i). B. An example of an afferent fibre with no spontaneous discharge. A template was created by the Spike2 software to identify all action potentials (APs) fired by this particular afferent fibre (upper bar). The baseline tongue temperature for this sensory fibre was 23.3 8C. As temperature gradually rose, the lowest temperature when this fibre started firing APs was taken as its thermal threshold i.e. 42.0 8C.
thresholds determined from these 10 consecutive mechanical stimuli were averaged to calculate the baseline mechanical threshold. Nine females were used to assess baseline properties only, the remaining 18 female rats were randomly assigned to receive a 10-min topical bath application of either vehicle control (2.0–4.0 ml, 1% PBS; n = 6), aqueous GABA solution (2.0– 4.0 ml, 0.5 M; n = 6) or muscimol (2.0–4.0 ml, 2.0–4.0 ml, 20 mg/ ml; n = 6). Randomisation was accomplished by assigning individual treatments to a series of random numbers generated with a spreadsheet programme (Microsoft Excel). The investigator (S.N.) was blinded to identity of the drug solution used. Aqueous solutions were applied to the tongue in a water bath (Fig. 1). After 10 min of recording of any spontaneous discharge, the tongue was immersed in drug solution for 10 min. The solution was removed and the
mechanical threshold reassessed at one minute intervals for 30 min. The tongue was then immersed in hot water (60 8C) to test whether afferent fibres also responded to thermal stimulation. The hot water was removed after 2–3 min to prevent damage to the tongue. The tongue was cooled down with room-temperature saline. Afferent fibres which fired action potentials when hot water was added were categorised as polymodal nociceptors, whereas those which did not were categorised as mechanoreceptive afferent fibres. Mechanical threshold was reassessed at one minute intervals for 10 min. At the end of each experiment, rats were terminated with an overdose of pentobarbital (100 mg/kg). The conduction distance of lingual nerve was approximated to the nearest milimeter by using a suture to trace the anatomical distribution of the nerve from the external meatus to the respective receptive fields on the tongue.
254
archives of oral biology 59 (2014) 251–257
Fig. 2 – Photomicrographs of a section of tongue are shown. The arrows indicate positive labelling for the axonal marker PGP 9.5, the GABAAg2 subunit (GABAA) and the neuropeptide substance P (SP). Note that in this section, the label for the GABAAg2 subunit is primarily seen in the mucosa. The white calibration bar indicates 50 mm.
2.3.
Drug solutions
GABA powder was dissolved in 1% phosphate buffered saline (PBS, pH 7.4) solution to a concentration of 51.5 mg/ml (0.5 M) while muscimol was first dissolved in 0.05 M hydrochloric acid followed by 1% PBS to a concentration of 20 mg/ml (0.175 mM). Solutions of GABA and muscimol were kept frozen ( 20 8C) until used. The concentration of GABA (51.5 mg/ml or 0.5 M) was selected based on previous research which had shown that this concentration inhibited temporomandibular jointjaw muscle reflexes.19,21 The concentration of muscimol (20 mg/ml or 0.175 mM) was chosen based on studies investigating topical applications of muscimol on the peripheral nervous system.24,25
2.4.
Data analysis
The total number of PGP9.5 positive fibres was counted. The number of PGP9.5 positive fibres that co-expressed the GABAA g2 subunit but not substance P (non-petidergic) was counted, and the number of PGP9.5 positive fibres that co-expressed both the GABAA g2 subunit and substance P (peptidergic) was counted. The proportion of GABAA g2 positive fibres with and without substance P was determined by dividing the total number of these fibres by the total number of PGP9.5 positive fibres. A Mann–Whitney rank sum test was used to compare the proportion GABAA g2 subunit expressing fibres in the substance P positive and substance P negative groups. The density of nerve fibres from each section was determined by dividing the total number of PGP 9.5-positive fibres by the area of the tissue section. Spike2 (Cambridge Electronic Devices, UK) was used to sort action potentials in cases where more than one afferent fibre was recorded. The conduction velocity (CV) of each recorded afferent fibre was calculated by dividing the estimated conduction distance (mm) by the latency to evoke an orthodromic action potential. Average mechanical threshold from each 10 min epoch was calculated. Relative mechanical threshold was calculated by dividing the mechanical threshold post treatment(s) by the baseline mechanical threshold. Significant differences in mechanical threshold were analysed with a two-way repeated measures ANOVA with time as the repeated factor and treatment as the non-repeated factor one-
way or one-way ANOVA after hot water application. A Spearman rank order correlation was used to investigate the relationships between conduction velocity and mechanical threshold. In all tests, the level of significance was set at P < 0.05.
3. 3.1.1.
Results GABAA receptor expression
The mean ( SD) nerve fibre density was 4.0 0.3 nerve fibres/ mm2 calculated from averages from four tongue sections/rat from six female rat tongues. It was found that 22 4% of nerve fibres in the tongue were positive for substance P (Fig. 2A). There was no significant difference between the proportions of peptidergic fibres expressing the GABAAg2 (n = 24) (94% [89– 100%]) and non-peptidergic fibres that expressed GABAAg2 (n = 24) (93% [92–96%]) (P > 0.05, Mann–Whitney rank sum test).
3.2.
Response properties
A total of 60 mechanically-responsive afferent fibres were collected from 27 female Sprague–Dawley. In nine female rats, only the baseline properties of 30 afferent fibres were assessed. Based on their response to 60 8C water, 42 (70% of the population) were categorised as mechanoreceptors and 18 (30% of the population) as polymodal nociceptors. The estimated CVs of the collected fibres ranged from 0.99– 53.13 m/s. There were 41 Ab fibres (>12 m/s), 16 Ad fibres (2– 12 m/s) and 3 C-fibres ( 0.05, Spearman Rank Order Correlation) (Fig. 3). Comparisons of the basic properties between mechanoreceptors and polymodal nociceptors showed no significant differences between the two types of sensory afferent fibres. There was no significant difference in the CVs (mean SEM) when mechanoreceptors (19.48 1.76) and polymodal nociceptors (18.16 2.30) were compared (P > 0.05, Student’s t test). Similarly, the baseline mechanical thresholds (median [interquartile range]) between mechanoreceptors (2.13 [1.41– 3.57]) and polymodal nociceptors (2.11 [1.19–3.03]) were
archives of oral biology 59 (2014) 251–257
PM MR
Baseline Mechanical Threshold (g)
30
also not significantly different (P > 0.05, Mann–Whitney rank sum test).
25
3.2.1.
20 15 10 5 0 0
10
20
30
40
50
60
Conduction Velocity (m/s)
Fig. 3 – The scatter plot shows the relationship between conduction velocity and mechanical threshold for the 60 tongue afferent fibres recorded from female rats. There was no significant relationship between baseline mechanical threshold and conduction velocity (P > 0.05, Spearman rank order correlation). PM: polymodal nociceptors, MR: mechanoreceptors.
Relative Mechanical Threshold
3
2
1
0
G
M
V
B
Relative Mechanical Threshold
0-10 min
2.0
G
M
V
10-20 min
G
M
V
20-30 min
*
1.5
1.0
0.5
0.0
G
M
V
Fig. 4 – (A) The mean relative mechanical threshold (MT) after treatment with GABA (n = 6), muscimol (n = 5) or vehicle (n = 8) solutions is shown. There was no significant effect of treatment on the relative MT. (B) Ten minutes after thermal stimulation (hot water at 60 8C), the relative MT of muscimol-treated fibres was significantly greater than vehicle-treated fibres (*: P < 0.05, one way ANOVA).
Effect of GABA receptor agonists
A total of 8, 10 and 12 fibres (n = 6 female rats per group) were treated with GABA, muscimol or vehicle (PBS) solutions, respectively. There were no significant differences in the conduction velocity (GABA 19 2 m/s, muscimol 26 4 m/s, PBS 22 3 m/s) or baseline mechanical threshold (GABA 1.5 0.5 g, muscimol 3.4 0.8 g, PBS 4.4 2.1 g) of the three groups of afferent fibres. Two way repeated measures ANOVA indicated a significant effect of time, but not treatment or the interaction between treatment and time on the mechanical threshold of afferent fibres examined (Fig. 4A). Post hoc analysis revealed that all three treatments significantly increased mechanical thresholds compared to baseline over the entire 30 min post-assessment period. Mechanical thresholds were reassessed after application of hot water to the bath of 5, 6 and 8 fibres treated with GABA, muscimol and vehicle solutions, respectively (Fig. 4B). The relative mechanical threshold of the muscimol-treated fibres was significantly greater than the vehicle-treated fibres (*: P < 0.05, one way ANOVA, Holm-Sidak post hoc test).
4.
A
255
Discussion
It has been reported that BMS patients who sucked a clonazepam tablet reported improvement in their oral pain, however, it is unclear what mechanism might underlie this effect.14 In the present study it was discovered that a large proportion of nerve fibres in the tongue express GABAA receptors containing the g2 subunit (GABAAg2). Benzodiazepines increase chloride currents through the GABAA receptor by binding to the g subunit, which suggests that it is possible that the pain relieving effects of clonazepam in BMS are mediated, at least in part, through an action on peripheral GABAA receptors.18,26,27 This argument is strengthened by the finding that topical application of the GABAA agonist muscimol was able to increase the mechanical threshold of tongue afferent fibres after exposure to hot water. This finding suggests that activation of peripheral GABAA receptors decreases the sensitivity of tongue afferent fibres, perhaps explaining how topical clonazepam exerts its analgesic effect. The high proportion of GABAAg2 subunit expression by tongue afferent fibres is consistent with a report that essentially all neuronal cell bodies in the rat trigeminal ganglion express the GABAAg1, GABAAg2 or GABAAg3 subunits.28 In the current study, axonal fibres on the most superficial aspect of the mucosa i.e. nerve fibres in the epithelial cells around and in between tongue papilla were quantified. GABAA receptors have also been discovered in the mandible, the palate and the salivary glands.29 Taken together, such an abundance of peripheral GABAA receptors in the oral cavity indicates that these receptors are likely to have important functional roles and are likely to be accessible for local pharmacologic manipulation. One of these functions is likely modification of nociceptive input, since 25% of GABAAg2 subunit expressing nerve fibres co-expressed substance P, a marker of small diameter afferent fibres 30–32.
256
archives of oral biology 59 (2014) 251–257
In this study, 25% of tongue afferent fibres were categorised as polymodal in intact female rats. In contrast, in vitro singleunit recordings of male rat lingual nerve reported 73% of their recorded fibres were polymodal nociceptors.33 However, only afferent fibres with conduction velocities
Available online at www.sciencedirect.com
ScienceDirect journal homepage: http://www.elsevier.com/locate/aob
Peripheral GABAA receptor activation modulates rat tongue afferent mechanical sensitivity Sun Nee Tan a, Esther Song a, Xu-Dong Dong a,b, Rishi Kumar Somvanshi a, Brian E. Cairns a,c,* a
Faculty of Pharmaceutical Sciences, The University of British Columbia, Vancouver, BC, Canada College of Stomatology, Tianjin Medical University, Tianjin 300071, China c Center for Sensory-Motor Interaction, Faculty of Medicine, Aalborg University, Aalborg, Denmark b
article info
abstract
Article history:
Objectives: The expression of GABAA receptors and the effects of GABAA receptor agonists
Received 30 June 2013
on the response properties of tongue afferent fibres were investigated in female rats to
Received in revised form
determine if peripheral GABA receptors might be a target of topical benzodiazepines when
7 November 2013
used for pain relief in burning mouth syndrome patients.
Accepted 27 November 2013
Design: Nerve fibres in tongue sections from six female rats were identified using protein gene product 9.5, and the co-expression of the g subunit of GABAA receptor and substance P
Keywords:
assessed in the nerve fibres. In vivo extracellular recordings of trigeminal ganglion neurons
-Aminobutyric acid (GABA)
that innervate the tongue were undertaken in 27 anesthetised female rats and their
Burning mouth syndrome
responses to mechanical and thermal stimulation characterised before and after topical
Benzodiazepine
application of GABA, the GABAA receptor selective agonist muscimol or vehicle control.
Tongue
Results: The vast majority of tongue nerve fibres examined (95%) expressed the g subunit of
Mucosa
GABAA receptor. Bath application of muscimol, but not GABA, significantly increased the
Pain
mechanical thresholds of tongue afferent fibres compared to vehicle, but only after the
Nerve fibre
tongue had been heated with 60 8C water. Conclusions: GABAA receptors are present on tongue nerve fibres and their activation alters the mechanical sensitivity these fibres. These findings suggest that topical application of benzodiazepines to the oral mucosa may decrease pain in burning mouth syndrome through a local action on peripheral GABAA receptors. # 2013 Elsevier Ltd. All rights reserved.
1.
Introduction
Burning mouth syndrome is a chronic intraoral pain condition for which no medical or dental cause has yet been identified.1–3 The prevalence of this syndrome in the general population has been estimated to be around 1%.4–9 Burning mouth syndrome is characterised by symptoms of ‘‘burning-like’’ pain most commonly experienced on the anterior of the tongue, as well as the anterior hard palate and lips, although
other sites in the mucosa of the oral cavity and throat can also be involved.3,10–12 Burning mouth syndrome may go into remission after a period of years,13 and thus conservative medical treatment aimed at symptom resolution is the normal therapeutic avenue. Unfortunately, burning mouth syndrome symptoms have proven quite resistant to conventional pharmacotherapy. Recently, it was reported that a non-conventional approach that involved sucking a 1 mg tablet of the benzodiazepine clonazepam, without swallowing, three times a day for 14 days
* Corresponding author. Tel.: +1 604 822 7715; fax: +1 604 822 3035. E-mail address: [email protected] (B.E. Cairns). 0003–9969/$ – see front matter # 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.archoralbio.2013.11.015
252
archives of oral biology 59 (2014) 251–257
was associated with significant symptomatic improvement in pain scores.14 Benzodiazepines are anxiolytic drugs thought to mediate their effects in the central nervous system by increasing the ability of g-amino-butyric acid (GABA) to open the GABAA receptor; a ligand-gated chloride channel. The interaction between benzodiazepines and the GABAA receptor occurs at the interface between the a and g subunits.15–18 Activation of peripheral GABAA receptors in orofacial tissues is not excitatory and appears to exert anti-nociceptive actions in some animal models of acute craniofacial pain.19–21 These findings suggest that topical application of benzodiazepines to the oral mucosa may decrease pain in burning mouth syndrome through a local action on peripheral GABAA receptors to decrease the excitability of nociceptors.14 However, evidence for a direct effect of GABAA receptor activation on nociceptors in the oral cavity is lacking. In the present study, the expression of GABAA receptors in tongue nerve fibres and the effects of GABA and the GABAA receptor selective agonist, muscimol, on the response properties of tongue afferent fibres were investigated in female rats. It was hypothesised that GABAA receptors would be widely expressed by nerve fibres in the tongue and that activation of these receptors would alter the mechanical threshold of these nerve fibres.
2.
Materials and methods
2.1.
Immunohistochemistry
Tongues from six female Sprague–Dawley rats were removed and frozen with liquid nitrogen. The first 5 mm of frozen tongue tissue, measured from the tip of the tongue was cut coronally and cut into 10 mm-thick sections with a cryotome. A total of four sections from each tongue were chosen for analysis, with each section separated from the previous section by a distance of at least 50 mm. The sections were mounted on glass slides and stored at 20 8C. Sections were incubated in 0.2% Triton-X100 and 10% normal goat serum (NGS) in phosphate-buffered saline (PBS) for 1 h at room temperature. The slides were then washed with PBS three times and incubated overnight at 4 8C with primary antibodies. Axonal fibres were identified using rabbit polyclonal antibody against rat protein gene product (PGP) 9.5 (1:2000; Abcam Inc., Cambridge, MA, USA). PGP 9.5 is an axonal marker and served to identify nerve fibres. The g2 subunit of GABAA receptors were labelled with the use of goat polyclonal antibody (1:50, Santa Cruz Biotechnology, Inc., CA, USA). Substance P expression, which identifies neuropeptide containing afferent fibres, was assessed through the use of a guinea pig polyclonal antibody (1:700; Abcam Inc., Cambridge, MA). Alexa Fluor 555 donkey anti-rabbit IgG antibody (1:700; Invitrogen, NY, USA) was used to visualise the PGP 9.5immunoreactivity (IR); Alexa Fluor 488 donkey anti-goat IgG antibody (1:700; Invitrogen, NY, USA) was used to visualise GABAA g2-IR and Alexa Fluor 633 goat anti-guinea pig IgG antibody (1:700; Invitrogen, NY, USA) was used to visualise the substance P-IR. The entire mucosal area of each section was imaged using a Leica TCS SPE confocal microscope. Some tissue sections were
incubated without primary antibodies to assess the degree of non-specific labelling. In addition, hemaglutinin-tagged cDNA for g2 subunit of GABAA receptor was transfected into HEK-293 cells and Western blots run on the protein extract from these cells. The blots revealed that GABAA g2 antibody labelled a single band at 65 kDa corresponding to molecular weight of the GABAA g2 A nerve fibre was defined as any fibril-shaped fluorescence of at least 4.0 mm in length which stained positive for PGP9.5 antibody. Nerve fibres were considered positively-stained for any of the antibodies used when the intensity of the fluorescence signals were more than 2 standard deviations (SDs) above the mean background intensity. The number of neuronal fibres on the mucosae, with or without colocalisation with SP and GABAA Rg2 were counted using Image J software programme (National Institutes of Health Image). Images of the entire tissue sections were captured using a digital camera (Olympus) and areas of the tissue sections were calculated using Image J software programme (National Institutes of Health Image).
2.2.
In vivo electrophysiology
Female Sprague–Dawley rats (n = 27; 260–340 g; Charles River Inc., Wilmington, MA, USA) were used. Rats were prepared to allow recording of trigeminal ganglion neurons under surgical anaesthesia (O2: 0.3–0.4 l/min; isoflurane 2.5–3.0%;).22 A tracheal cannula was inserted to initiate artificial ventilation and the carotid artery was catheterised to measure blood pressure. The rat’s head was placed in a stereotaxic frame, the skin over the dorsal surface of the skull was reflected and a trephination was made on the right side of the skull. A parylene-coated tungsten-recording electrode was lowered into the trigeminal ganglion via the trephination to record action potentials from single trigeminal ganglion neurons.23 A suture was placed in the middle of the tongue to stretch and hold it in place (Fig. 1). In some experiments, a thermal probe was inserted into the tongue to measure changes in tongue temperature. Throughout the entire experiment, heart rate, mean blood pressure and core body temperature were continuously monitored and kept within physiological ranges of 300–400 beats/min, 60–80 mm Hg and 36.8–37.1 8C, respectively. Upon completion of all surgical procedures, the isoflurane level was reduced to 1.5–2.5% to achieve a depth of anaesthesia characterised by continued absence of a toe pinch reflex. All procedures were performed in adherence with the principles of the Canadian Council on Animal Care and were approved by the University of British Columbia Animal Care Committee. All efforts were made to minimise the number of animals used and their suffering. A blunt probe was used to mechanically stimulate the tongue to identify mechanoreceptive afferent fibres (Fig. 1). Upon identifying one or more afferent fibres, a stimulating electrode was placed in contact with the tongue and moved until increasing electrical current (1–10 mA, 0.5 ms duration at 1 Hz frequency) evoked single orthodromic action potentials of fixed latency. Then, an electronic von Frey hair (model 160IC, IITC) was used to determine the baseline mechanical threshold of the tongue afferent fibres. The von Frey hair was applied at one minute intervals for 10 min. The mechanical
archives of oral biology 59 (2014) 251–257
253
Fig. 1 – An electronic Von Frey hair was used to determine the MTs of afferent fibres before and after topical drug applications. A rubber dam sealed the water bath to prevent leaking of drug solutions so that tongue could be immersed in the bath solution for 10 min each time. A suture was used to fix tongue in a tensed position. A. Method to record and calculate mechanical threshold (MT) from a tongue afferent fibre. At one minute intervals throughout the recording, mechanical pressure was applied using the electronic Von Frey instrument to the afferent receptive field until an action potential (AP) (top trace, near B) was fired. The lowest magnitude of mechanical pressure required generate an AP from this sensory fibre was taken as its MT (from bottom bar; value ii minus value i). B. An example of an afferent fibre with no spontaneous discharge. A template was created by the Spike2 software to identify all action potentials (APs) fired by this particular afferent fibre (upper bar). The baseline tongue temperature for this sensory fibre was 23.3 8C. As temperature gradually rose, the lowest temperature when this fibre started firing APs was taken as its thermal threshold i.e. 42.0 8C.
thresholds determined from these 10 consecutive mechanical stimuli were averaged to calculate the baseline mechanical threshold. Nine females were used to assess baseline properties only, the remaining 18 female rats were randomly assigned to receive a 10-min topical bath application of either vehicle control (2.0–4.0 ml, 1% PBS; n = 6), aqueous GABA solution (2.0– 4.0 ml, 0.5 M; n = 6) or muscimol (2.0–4.0 ml, 2.0–4.0 ml, 20 mg/ ml; n = 6). Randomisation was accomplished by assigning individual treatments to a series of random numbers generated with a spreadsheet programme (Microsoft Excel). The investigator (S.N.) was blinded to identity of the drug solution used. Aqueous solutions were applied to the tongue in a water bath (Fig. 1). After 10 min of recording of any spontaneous discharge, the tongue was immersed in drug solution for 10 min. The solution was removed and the
mechanical threshold reassessed at one minute intervals for 30 min. The tongue was then immersed in hot water (60 8C) to test whether afferent fibres also responded to thermal stimulation. The hot water was removed after 2–3 min to prevent damage to the tongue. The tongue was cooled down with room-temperature saline. Afferent fibres which fired action potentials when hot water was added were categorised as polymodal nociceptors, whereas those which did not were categorised as mechanoreceptive afferent fibres. Mechanical threshold was reassessed at one minute intervals for 10 min. At the end of each experiment, rats were terminated with an overdose of pentobarbital (100 mg/kg). The conduction distance of lingual nerve was approximated to the nearest milimeter by using a suture to trace the anatomical distribution of the nerve from the external meatus to the respective receptive fields on the tongue.
254
archives of oral biology 59 (2014) 251–257
Fig. 2 – Photomicrographs of a section of tongue are shown. The arrows indicate positive labelling for the axonal marker PGP 9.5, the GABAAg2 subunit (GABAA) and the neuropeptide substance P (SP). Note that in this section, the label for the GABAAg2 subunit is primarily seen in the mucosa. The white calibration bar indicates 50 mm.
2.3.
Drug solutions
GABA powder was dissolved in 1% phosphate buffered saline (PBS, pH 7.4) solution to a concentration of 51.5 mg/ml (0.5 M) while muscimol was first dissolved in 0.05 M hydrochloric acid followed by 1% PBS to a concentration of 20 mg/ml (0.175 mM). Solutions of GABA and muscimol were kept frozen ( 20 8C) until used. The concentration of GABA (51.5 mg/ml or 0.5 M) was selected based on previous research which had shown that this concentration inhibited temporomandibular jointjaw muscle reflexes.19,21 The concentration of muscimol (20 mg/ml or 0.175 mM) was chosen based on studies investigating topical applications of muscimol on the peripheral nervous system.24,25
2.4.
Data analysis
The total number of PGP9.5 positive fibres was counted. The number of PGP9.5 positive fibres that co-expressed the GABAA g2 subunit but not substance P (non-petidergic) was counted, and the number of PGP9.5 positive fibres that co-expressed both the GABAA g2 subunit and substance P (peptidergic) was counted. The proportion of GABAA g2 positive fibres with and without substance P was determined by dividing the total number of these fibres by the total number of PGP9.5 positive fibres. A Mann–Whitney rank sum test was used to compare the proportion GABAA g2 subunit expressing fibres in the substance P positive and substance P negative groups. The density of nerve fibres from each section was determined by dividing the total number of PGP 9.5-positive fibres by the area of the tissue section. Spike2 (Cambridge Electronic Devices, UK) was used to sort action potentials in cases where more than one afferent fibre was recorded. The conduction velocity (CV) of each recorded afferent fibre was calculated by dividing the estimated conduction distance (mm) by the latency to evoke an orthodromic action potential. Average mechanical threshold from each 10 min epoch was calculated. Relative mechanical threshold was calculated by dividing the mechanical threshold post treatment(s) by the baseline mechanical threshold. Significant differences in mechanical threshold were analysed with a two-way repeated measures ANOVA with time as the repeated factor and treatment as the non-repeated factor one-
way or one-way ANOVA after hot water application. A Spearman rank order correlation was used to investigate the relationships between conduction velocity and mechanical threshold. In all tests, the level of significance was set at P < 0.05.
3. 3.1.1.
Results GABAA receptor expression
The mean ( SD) nerve fibre density was 4.0 0.3 nerve fibres/ mm2 calculated from averages from four tongue sections/rat from six female rat tongues. It was found that 22 4% of nerve fibres in the tongue were positive for substance P (Fig. 2A). There was no significant difference between the proportions of peptidergic fibres expressing the GABAAg2 (n = 24) (94% [89– 100%]) and non-peptidergic fibres that expressed GABAAg2 (n = 24) (93% [92–96%]) (P > 0.05, Mann–Whitney rank sum test).
3.2.
Response properties
A total of 60 mechanically-responsive afferent fibres were collected from 27 female Sprague–Dawley. In nine female rats, only the baseline properties of 30 afferent fibres were assessed. Based on their response to 60 8C water, 42 (70% of the population) were categorised as mechanoreceptors and 18 (30% of the population) as polymodal nociceptors. The estimated CVs of the collected fibres ranged from 0.99– 53.13 m/s. There were 41 Ab fibres (>12 m/s), 16 Ad fibres (2– 12 m/s) and 3 C-fibres ( 0.05, Spearman Rank Order Correlation) (Fig. 3). Comparisons of the basic properties between mechanoreceptors and polymodal nociceptors showed no significant differences between the two types of sensory afferent fibres. There was no significant difference in the CVs (mean SEM) when mechanoreceptors (19.48 1.76) and polymodal nociceptors (18.16 2.30) were compared (P > 0.05, Student’s t test). Similarly, the baseline mechanical thresholds (median [interquartile range]) between mechanoreceptors (2.13 [1.41– 3.57]) and polymodal nociceptors (2.11 [1.19–3.03]) were
archives of oral biology 59 (2014) 251–257
PM MR
Baseline Mechanical Threshold (g)
30
also not significantly different (P > 0.05, Mann–Whitney rank sum test).
25
3.2.1.
20 15 10 5 0 0
10
20
30
40
50
60
Conduction Velocity (m/s)
Fig. 3 – The scatter plot shows the relationship between conduction velocity and mechanical threshold for the 60 tongue afferent fibres recorded from female rats. There was no significant relationship between baseline mechanical threshold and conduction velocity (P > 0.05, Spearman rank order correlation). PM: polymodal nociceptors, MR: mechanoreceptors.
Relative Mechanical Threshold
3
2
1
0
G
M
V
B
Relative Mechanical Threshold
0-10 min
2.0
G
M
V
10-20 min
G
M
V
20-30 min
*
1.5
1.0
0.5
0.0
G
M
V
Fig. 4 – (A) The mean relative mechanical threshold (MT) after treatment with GABA (n = 6), muscimol (n = 5) or vehicle (n = 8) solutions is shown. There was no significant effect of treatment on the relative MT. (B) Ten minutes after thermal stimulation (hot water at 60 8C), the relative MT of muscimol-treated fibres was significantly greater than vehicle-treated fibres (*: P < 0.05, one way ANOVA).
Effect of GABA receptor agonists
A total of 8, 10 and 12 fibres (n = 6 female rats per group) were treated with GABA, muscimol or vehicle (PBS) solutions, respectively. There were no significant differences in the conduction velocity (GABA 19 2 m/s, muscimol 26 4 m/s, PBS 22 3 m/s) or baseline mechanical threshold (GABA 1.5 0.5 g, muscimol 3.4 0.8 g, PBS 4.4 2.1 g) of the three groups of afferent fibres. Two way repeated measures ANOVA indicated a significant effect of time, but not treatment or the interaction between treatment and time on the mechanical threshold of afferent fibres examined (Fig. 4A). Post hoc analysis revealed that all three treatments significantly increased mechanical thresholds compared to baseline over the entire 30 min post-assessment period. Mechanical thresholds were reassessed after application of hot water to the bath of 5, 6 and 8 fibres treated with GABA, muscimol and vehicle solutions, respectively (Fig. 4B). The relative mechanical threshold of the muscimol-treated fibres was significantly greater than the vehicle-treated fibres (*: P < 0.05, one way ANOVA, Holm-Sidak post hoc test).
4.
A
255
Discussion
It has been reported that BMS patients who sucked a clonazepam tablet reported improvement in their oral pain, however, it is unclear what mechanism might underlie this effect.14 In the present study it was discovered that a large proportion of nerve fibres in the tongue express GABAA receptors containing the g2 subunit (GABAAg2). Benzodiazepines increase chloride currents through the GABAA receptor by binding to the g subunit, which suggests that it is possible that the pain relieving effects of clonazepam in BMS are mediated, at least in part, through an action on peripheral GABAA receptors.18,26,27 This argument is strengthened by the finding that topical application of the GABAA agonist muscimol was able to increase the mechanical threshold of tongue afferent fibres after exposure to hot water. This finding suggests that activation of peripheral GABAA receptors decreases the sensitivity of tongue afferent fibres, perhaps explaining how topical clonazepam exerts its analgesic effect. The high proportion of GABAAg2 subunit expression by tongue afferent fibres is consistent with a report that essentially all neuronal cell bodies in the rat trigeminal ganglion express the GABAAg1, GABAAg2 or GABAAg3 subunits.28 In the current study, axonal fibres on the most superficial aspect of the mucosa i.e. nerve fibres in the epithelial cells around and in between tongue papilla were quantified. GABAA receptors have also been discovered in the mandible, the palate and the salivary glands.29 Taken together, such an abundance of peripheral GABAA receptors in the oral cavity indicates that these receptors are likely to have important functional roles and are likely to be accessible for local pharmacologic manipulation. One of these functions is likely modification of nociceptive input, since 25% of GABAAg2 subunit expressing nerve fibres co-expressed substance P, a marker of small diameter afferent fibres 30–32.
256
archives of oral biology 59 (2014) 251–257
In this study, 25% of tongue afferent fibres were categorised as polymodal in intact female rats. In contrast, in vitro singleunit recordings of male rat lingual nerve reported 73% of their recorded fibres were polymodal nociceptors.33 However, only afferent fibres with conduction velocities
