The role of Nav1.9 channel in the development of neuropathic orofacial pain associated with trigeminal neuralgia

Background Trigeminal neuralgia is accompanied by severe mechanical, thermal and chemical hypersensitivity of the orofacial area innervated by neurons of trigeminal ganglion (TG). We examined the role of the voltage-gated sodium channel subtype Nav1.9 in the development of trigeminal neuralgia. Results We found that Nav1.9 is required for the development of both thermal and mechanical hypersensitivity induced by constriction of the infraorbital nerve (CION). The CION model does not induce change on Nav1.9 mRNA expression in the ipsilateral TG neurons when evaluated 9 days after surgery. Conclusions These results demonstrate that Nav1.9 channels play a critical role in the development of orofacial neuropathic pain. New routes for the treatment of orofacial neuropathic pain focussing on regulation of the voltage-gated Nav1.9 sodium channel activity should be investigated.


Background
Trigeminal neuralgia is an excruciating pain syndrome characterized by severe facial pain that can be triggered by light touch of the orofacial surface area producing stabbing, shooting or burning sensations. Trigeminal neuralgia is accompanied by mechanical, thermal and/ or chemical hypersensitivities of the orofacial area and is maintained by impaired signaling in sensory neurons of trigeminal ganglion (TG) [1]. However, little is known about the intracellular mechanisms or altered functioning of TG nociceptors underling the development of orofacial pain.
Studies of the role of Na v 1.9 channels showed that SCN11A null mutant mice developed reduced hyperalgesia in different inflammatory pain models [formalin, carrageenan, complete freund's adjuvant (CFA), prostaglandin E 2 ] [21,22] as well as diminished nociceptive sensitivity triggered by inflammatory mediators [bradykinin, serotonin, adenosine triphosphate (ATP)] [2]. In contrast, Na v 1.9 −/− mice showed unaltered pain-related behaviour in models of DRG-related neuropathic pain of various origins (partial sciatic nerve injury [2], chronic constriction injury (CCI) [23] and spinal nerve transaction [24]). Recent studies of human SCN11A gene mutations demonstrated that several mutations in this gene are associated with either, peripheral neuropathy and episodic chronic pain [25] or congenital inability to experience pain [26]. Despite the established role of Na v 1.9 channels in pain perception, the contribution of this channel subtype to the development of orofacial pain remains unknown, and how cephalic neuropathic pain develops with loss of Na v 1.9 channels has not been studied.
Using global Na v 1.9 knockout mice, we show for the first time that loss of Na v 1.9 channels is associated with the failure to develop orofacial pain in a model of trigeminal neuralgia.

Results
Basal orofacial sensitivity in Na v 1.9 +/+ and Na v 1.9 −/− mice To examine if the loss of Na v 1.9 channels could influence the sensitivity of the orofacial area, basal mechanical and thermal sensitivities were examined in Na v 1.9 +/+ and Na v 1.9 -/mice. Mechanical orofacial sensitivity was evaluated as a functional reaction in response to 0.04 g von Frey filament applied to the animal forehead innervated by the trigeminal nerve (Fig. 1a). A filament of 0.04 g intensity was chosen since I have been found to evoke nociceptive responses in the orofacial area after unilateral constriction of the infraorbital nerve (CION), but present almost no effect in control mice [27]. No differences in basal responses to mechanical stimulation were observed between genotypes (n = 13-14 mice, p > 0.05; Fig. 1c). Groups did not differ with regard to basal thermal sensitivity, measured as a latency of response to radiant heat applied to the vibrissal pad surface (n = 13-14 mice, p > 0.05; Fig. 1b, d). Thus, loss of Na v 1.9 channels does not affect basal orofacial sensitivity, either mechanical or thermal, in mice.

Loss of Na v 1.9 channels alleviates CION-induced orofacial hypersensitivity
Constriction of the infraorbital nerve [25] led to a development of long-lasting mechanical hypersensitivity in Na v 1.9 +/+ . This hypersensitivity developed 7 days after surgery, reached a peak at 2 weeks, and remained persistent until 3 weeks (Fig. 2a). CION-induced mechanical hypersensitivity was robust, showing a significantly increased reaction in response to von Frey filament applied to the forehead of Na v 1.9 +/+ and Na v 1.9 −/− mice. The increase in the response frequency was 26 % (p < 0.05), 35 % (p < 0.01) and 31 % (p < 0.01) on day 7, 14 and 21 after CION respectively compared to shamoperated mice (n = 13-14 mice; Fig. 2a). Strikingly, in Na v 1.9 −/− mice, CION was not accompanied by the development of hypersensitivity of the orofacial area. The responses of Na v 1.9 −/− mice were similar to those observed in the littermates of the sham-operated group over all periods tested. Thus, loss of Na v 1.9 channels abolished the development of CION-induced mechanical hypersensitivity.
Loss of Na v 1.9 channels also abolished the development of CION-induced thermal hypersensitivity. Consistent with previous studies [28], CION produced a long-lasting thermal hypersensitivity in Na v 1.9 +/+ mice, which developed within 1 week after surgery. The CIONinduced decrease in thermal latency in Na v 1.9 +/+ mice was 10 % (p < 0.05) on day 7 and 12 % (p < 0.001) on day 14 post-surgery. However, the latency was not changed in Na v 1.9 −/− mice following CION over all period tested (n = 13-14 mice; Fig. 2b). Together, these results indicate that the Na v 1.9 channel is required for the development of neuropathic orofacial pain.
Expression of Na v 1.9 mRNA in TG on Sham and CION Na v 1.9 +/+ mice As the Na v 1.9 −/− mice did not develop mechanical or thermal hypersensitivity after CION, the next step was to evaluate if the CION model induced any change in the expression of Na v 1.9 mRNA in TG neurons. Sample from ipsilateral TG was processed on day 9 after surgery, for extraction of mRNA. As showed on Fig. 3, no Na v 1.9 mRNA expression difference was observed between the CION (1.06 ± 0.42) and Sham (1.12 ± 0.28) groups in Na v 1.9 +/+ mice (p = 0.9026).

Discussion
Considerable evidence suggests participation of Na v 1.9 channels in the perception of pain. Both gain of function and loss of function pain conditions have been associated with point mutations in Na v 1.9 [25,26]. Two of those mutations led to a reduction in the current threshold and increased firing frequency in response to suprathreshold Fig. 1 Loss of Na v 1.9 does not influence orofacial sensitivity in mice. Mechanical orofacial sensitivity in response to 0.04 g von Frey filament applied to the forehead (a) and thermal orofacial sensitivity in response to radiant heat applied to a vibrissal pad (b) in sham-operated Na v 1.9 +/+ and Na v 1.9 −/− mice. Mechanical (c) and thermal (d) orofacial sensitivities baselines. Mechanical (e) and thermal (f) orofacial sensitivities evaluated before (pre) and at different time-points after surgery. Changes in mechanical (g) and thermal (h) orofacial sensitivities after sham-operation. Values represent mean ± SEM analysed for 13-14 mice. **p < 0.01, ***p < 0.001 compared to that before surgery (pre) (Student paired t test) stimuli [25]. Mutations in SCN11A (R225C and A808G), reported in patients experienced episodic chronic pain, caused an increase in the Na v 1.9 channels-mediated current density and hyperexcitability of nociceptive DRG neurons without changes in the resting membrane potential [29]. Other gain of function mutations lead Na v 1.9 channels to display excessive activity at resting voltages, causing sustained depolarization of nociceptors, impaired generation of action potentials and resulting in a pain-free state [26]. The principal mechanism by which Na v 1.9 channels contribute to painful states includes sensitization of nociceptors triggered by generation of persistent Na + currents at subthreshold voltages that promotes depolarization near the resting membrane potential, reduces the current threshold required to trigger an action potential, and results in hyperexcitability of sensory neurons [7,16,[18][19][20]30]. Here using Na v 1.9 −/− mice, we show that Na v 1.9 channels are also involved in sustained (tonic) firing of nociceptive TG neurons and are required for the development of orofacial neuropathic pain.
The role of the Na v 1.9 channel in maintaining peripheral inflammatory pain has been demonstrated in different studies [21,22,30]. Inflammatory pain of different origins (induced by formalin, carrageenan, CFA, capsaicin, ATP, IL-1β or prostaglandin-E2) was substantially reduced in Na v 1.9 −/− mice [21,22,30], while neuropathic pain, produced by CCI or spinal nerve transection, still developed, showing mechanical allodynia of the hind paw of Na v 1.9 −/− mice [24,31]. Using CION-induced model of orofacial neuropathic pain [28,32], we demonstrate here the absence of both mechanical and thermal hypersensitivity of the orofacial area in Na v 1.9 −/− mice after CION. Loss of Na v 1.9 channels did not change basal orofacial sensitivity (mechanical or thermal), indicating no major effects of Na v 1.9 deletion on normal nociceptive pathways. Our findings of a crucial role for Na v 1.9 channels in the development of neuropathic orofacial pain are in contrast with reports showing no critical involvement of these channels in the development of neuropathic pain in other somatic neuropathic pain models [24,31]. This may reflect distinct mechanisms of neuropathic pain.
Na v 1.9 channels enhance sustained repetitive (tonic) firing in TG neurons. Previous studies showing that Na v 1.9 channels maintain sustained repetitive firing in DRG nociceptors and modulate patterns of firing discharge, including plateau potentials, active a b Fig. 2 Orofacial neuropathic hypersensitivity was attenuated in Na v 1.9 KO mice. Changes in mechanical orofacial sensitivity in response to 0.04 g von Frey filament applied to the forehead (a) and thermal orofacial sensitivity in response to radiant heat applied to a ipsilateral vibrissal pad (b) in Na v 1.9 +/+ and Na v 1.9 −/− mice before surgery (pre) and at different time-point after CION. Values represent mean ± SEM. analysed for 13-14 mice. *p < 0.05, **p < 0.01, ***p < 0.001 compared to the correspondent time-point in sham Na v 1.9 +/+ mice (Two-way ANOVA followed by Bonferroni's test)

Fig. 3
Expression of Na v 1.9 mRNA in TG on Sham and CION Na v 1.9 +/+ mice 9 days after surgery. Values represent mean ± SEM analysed for 6-7 mice (Student paired t test) hyperpolarizing responses, and sustained oscillatory bursting discharges [16,18,33]. It is intriguing that gain of function Na v 1.9 mutations in humans may result in enhanced pain or a complete loss of pain sensation [25,26]. In transgenic mice, inflammatory pain is associated with enhanced Na v 1.9 persistent sodium channel activity [30]. In the present study, neuropathic pain is dependent on the activity of Na v 1.9 in a model of trigeminal neuralgia. A critical role for Na v 1.9 in regulating pain thresholds is thus clear, but the precise mechanisms that contribute to the different pain phenotypes remains to be established.

Conclusions
Our study demonstrates for the first time that the Na v 1.9 channel plays a critical role in the development of neuropathic orofacial pain associated with trigeminal neuralgia. Understanding how orbital nerve damage leads to myofascial pain through the modulation of Na v 1.9 activity may provide new approaches to the treatment of trigeminal neuralgia.

Animals
C57Bl/6 mice (2-3 month-old) were used in accordance with the protocols approved by the UK Home Office and UCL ethics committee under a Home Office project license. All efforts were made to minimise animal suffering and to reduce the number of animals used. Experiments were conducted using both male and female wild type littermate and global Na v 1.9 knockout mice bred from heterozygous (Na v 1.9 +/− ) animals. The generation and characterization of the Na v 1.9 null mutant line was previously described [34]. Mouse colonies were genotyped by PCR using ear biopsy samples. Only homozygous Na v 1.9-null (Na v 1.9 −/− ) and wild type (Na v 1.9 +/+ ) mice were used in these studies. Genomic DNA was extracted using a lysis buffer containing (in mM) 67 Tris, 16 (NH 4 ) 2 SO 4 , 6.7 MgCl 2 , 100 β-mercaptoethanol, and 0.5 % Triton X-100 and 0.05 mg/ml proteinase K. The primers used were: 5′-AAC AGTCTTACGCTGTTCCGATG-3′ (sense), 5′-ATGTG GCACTGGGCTTGAACTC-3′ (antisense), 5′-CTCGTC GTGACCCATGGCGAT-3′ (Neomycin FW). PCR was performed in one reaction and resulted in a 276 bp fragment product for WT and a 600 bp band for knockout mice.

Infraorbital nerve constriction
To produce the painful condition of trigeminal neuralgia, we used a model of unilateral CION [25] according to a method previously described for rats [28,35] with some modifications. Briefly, mice were anesthetized with an intramuscular (i.m.) injection of a mixture of ketamine (Fort Dodge Animal Health LTD, UK) and medetomidine (Orion Pharma, UK) in the doses of 50 and 0.5 mg/kg, respectively. One incision was made below the right eye 1 mm caudal to the myofascial (vibrissal) pads. The superior lip elevator and anterior superficial masseter muscles were dissected to expose the rostral end of the infraorbital nerve, as it emerges from the infraorbital fissure. Extra care was taken to prevent facial nerve damage. Two silk ligatures (No. 8.0, Ethicon) were loosely tied at a distance of 2 mm around the infraorbital nerve, producing a development of orofacial hypersensitivity and preventing infraorbital nerve destruction [35]. The incision was closed with polypropylene sutures (No. 6.0, Ethicon). After surgery, all animals were injected i.m. with atipamezole (5 mg/kg, Orion Pharma) and were maintained in a warmed area until full recovery from anaesthesia. Animals operated on an identical manner with no ligature applied to the nerve were used as control (sham).

Measurement of mechanical hyperalgesia
Mice were acclimatized in an individual animal enclosure (4″ × 4″) chamber for at least 1 h. Sham-operated or CION-injured animals were submitted to a repeated mechanical stimulus (0.04 g von Frey filament) applied to the forehead surface innervated by the trigeminal nerve (Fig. 1a). Each trial was repeated 10 times at 30-s interval at least. The occurrence of head reactions (attack/escape or head withdrawal) was expressed for each trial as the mean of percentage reactions that represents an index of mechanical nociceptive sensitivity [27].
To determine basal mechanical sensitivity, the mechanical stimulus was applied to the forehead of animals 1 day before surgery. A development of mechanical hyperalgesia following CION was estimated at different days after surgery.

Measurement of thermal hyperalgesia
Thermal hyperalgesia of the orofacial area was measured as previously described [28,36]. Each animal was held in front of a radiant heat source positioned 1 cm from the vibrissal pad surface (Fig. 1b). The intensity of thermal stimulus applied to the vibrissal pad skin was adjusted to temperature of ~50 °C (15 s cut off ). Once the animal started to withdraw its head or vigorously flick a snout, the heat beam was turned off. The time between the start of the beam and a functional response was defined as a latency of response. A reduction in the response latency reflected thermal hyperalgesia. To determine basal thermal sensitivity, a heat stimulus was applied to the ipsilateral side of the snout 1 day before surgery. Basal latency of responses was typically from 10 to 15 s. The development of thermal hyperalgesia following CION was measured at different days after surgery.

Na v 1.9 expression analysis real-time PCR
After animal euthanasia (with a CO 2 chamber following by cervical dislocation), TG was quickly removed and placed in buffer RTL and RNA was isolated using RNeasy mini kit (Quiagen) according to the manufacturer instructions. The RNA was quantified by Nanodrop and converted to cDNA using iScript ™ reverse transcription supermix for qRT-PCR (BioRad) according to the manufacturer instructions. PCR amplifications were performed using a Mastercycler ep realplex (Eppendorf ). All samples were run in triplicate in a final volume of 20 µl containing 10 ng of cDNA, 1 µM of primers and SsoAdvanced ™ universal SYBR ® Green supermix (BioRad), according to manufacturer protocol. Prior to PCR, an 8 min enzyme activation step was done at 95 °C. The PCR protocol consisted of 10 s denaturation at 95 °C, 5 s at annealing temperature at 63 °C, 10 s elongation at 72 °C for 40 cycles.
The annealing temperature was confirmed by melting curve. The primers sequences used were the following: TTCCACTCTACGTACCTTCCGAGT (forward) and ATTCCCATGAAGAGCTGCTGACCA (reverse) for Na v 1.9; TGCGACTTCAACAGCAACTC (forward) and CTTGCTCAGTGTCCTTGCTG (reverse) for GAPDH. Na v 1.9 and GAPDH cDNA relative amounts were calculated function of the samples cycle threshold (Ct) using a standard concentration curve constructed with a serial dilution (10 times) from 100 to 0.01 ng of cDNA mix. For each sample, the relative amount of Na v 1.9 cDNA was normalized by the GAPDH cDNA amount.

Statistical analysis
Data were analyzed using GraphPad Prism 5. The results are presented as mean ± SEM with n referring to the number of animals tested. Multiple groups were compared using two-way analysis of variance [37] followed by Bonferroni post hoc test or Student paired t test as indicated.
Values of p less than 0.05 was considered as statistically significant for either test used.