Inhibition of G protein-coupled P2Y2 receptor induced analgesia in a rat model of trigeminal neuropathic pain

Backgrouds ATP and P2X receptors play important roles in the modulation of trigeminal neuropathic pain, while the role of G protein-coupled P2Y2 receptors and the underlying mechanisms are less clear. The threshold and frequency of action potentials, fast inactivating transient K+ channels (IA) are important regulators of membrane excitability in sensory neurons because of its vital role in the control of the spike onset. In this study, pain behavior tests, QT-RT-PCR, immunohistochemical staining, and patch-clamp recording, were used to investigate the role of P2Y2 receptors in pain behaviour. Results In control rats: 1) UTP, an agonist of P2Y2/P2Y4 receptors, caused a significant decrease in the mean threshold intensities for evoking action potentials and a striking increase in the mean number of spikes evoked by TG neurons. 2) UTP significantly inhibited IA and the expression of Kv1.4, Kv3.4 and Kv4.2 subunits in TG neurons, which could be reversed by the P2 receptor antagonist suramin and the ERK antagonist U0126. In ION-CCI (chronic constriction injury of infraorbital nerve) rats: 1) mRNA levels of Kv1.4, Kv3.4 and Kv4.2 subunits were significantly decreased, while the protein level of phosphorylated ERK was significantly increased. 2) When blocking P2Y2 receptors by suramin or injection of P2Y2R antisense oligodeoxynucleotides both led to a time- and dose-dependent reverse of allodynia in ION-CCI rats. 3) Injection of P2Y2 receptor antisense oligodeoxynucleotides induced a pronounced decrease in phosphorylated ERK expression and a significant increase in Kv1.4, Kv3.4 and Kv4.2 subunit expression in trigeminal ganglia. Conclusions Our data suggest that inhibition of P2Y2 receptors leads to down-regulation of ERK-mediated phosphorylation and increase of the expression of IA–related Kv channels in trigeminal ganglion neurons, which might contribute to the clinical treatment of trigeminal neuropathic pain.


Introduction
Trigeminal neuropathic pain disorders, as typical, atypical, or post-therapeutic trigeminal neuralgias, are pain that is either spontaneous or can be elicited by harmless but crucial activities, such as eating and talking, or by light touch to facial skin [1]. The current treatments do not provide long-lasting relief for these frequently treatment-refractory patients due to a limited understanding of their pathophysiology. Chronic constriction nerve injury (CCI) has characteristics of inflammation and nerve injury [2,3]. Previous studies using a chronic constriction nerve injury model of the infraorbital nerve (ION-CCI) have reported it to be a good model that mimics trigeminal neuralgia of humans [4][5][6][7]. The major pathologic changes for trigeminal neuralgia are axonal loss and demyelination in trigeminal root [8]. Constrictive infraorbital nerve injury like constrictive sciatic nerve injury induces demyelination as sources of pathological ectopic firing accompanying mechanical allodynia and heat hyperalgesia [4].
Adenosine 5′-triphosphate (ATP) and uridine 5′-triphosphate (UTP) are released from cells as a consequence of tissue injury and mediate their bio-effects through binding to a large group of cell surface receptors of both P2X or P2Y receptor families [9]. There were early hints that ATP might be involved in pain, including the demonstration of pain produced by injection of ATP into human skin blisters [10,11]. In trigeminal ganglion (TG) neurons, the highly selective distribution of P2X 3 and P2X 2/3 receptors within the nociceptive system has suggested a potential role for ATP as a pain mediator [12,13]. Expression of P2Y 1, 2, 4, and 6 receptors has also been reported in TG neurons [14]. P2Y 2 receptors are typically expressed on small, nociceptive neurons [15]. In vitro studies have demonstrated that co-activation of P2Y 2 receptors and TRPV channels by ATP could underlie ATP-induced pain [16]. UTP, a selective agonist for P2Y 2 and P2Y 4 receptors, activates cutaneous afferent fibers [17], mediates excitation of dorsal root ganglion (DRG) neurons [18] and sensitizes mouse bladder sensory neurons [19]. These results suggest that UTP may be an endogenous nociceptive messenger. However, in vivo studies have shown that UTP significantly alleviates mechanical allodynia in a neuropathic pain model [20,21]. However, the effect of activation of P2Y 2 receptors on neuropathic pain is not clear and requires further study.
Multiple types of voltage-gated ion channels are related to neuronal excitability, such as voltage-gated K + (Kv) channels, which are important regulators of membrane potentials and action potentials in nociceptive sensory neurons [22,23]. In rat small TG neurons, Kv currents have been divided into three types: slow inactivating transient K + current (I D ), fast inactivating transient K + current (I A ) and dominant sustained K + current (I K ) [24]. I A is particularly important in the control of the spike onset, the threshold of the action potential firing, and the firing frequency [25]. Many studies have shown that the Kv1.4, Kv3.4, Kv4.2, and Kv4.3 subunits contribute to the I A channels in DRG neurons [26][27][28], which suggests that I A has the ability to regulate the neuronal activity of nociceptive neurons. After sciatic nerve injury, the expression of Kv1.4 was decreased in small-diameter DRG neurons [28]. Another study showed that activation with the GABA B receptor agonist baclofen inhibited the excitability of TG neurons, which was mediated by potentiation of both I A and I K in rat smalldiameter TG neurons [29]. I A , I K and the total K + currents were significantly reduced in rats with inferior alveolar nerve transection and ION-CCI [3,30]. A recent report demonstrated that P2Y 2 receptors mediate an excitation of DRG neurons through inhibition of K V 7 channels [18].
In this study, we hypothesize that activation of P2Y 2 receptors might mediate trigeminal neuropathic pain through regulating the expression and function of Kv1.4, Kv3.4, Kv4.2, and Kv4.3 subunits. We have used pain behavior tests, quantitative reverse transcription-polymerase chain reaction analysis (QT-RT-PCR), immunohistochemical staining and patch-clamp recording to investigate the role of P2Y 2 receptors in pain behavior, excitability of TG neurons, and modulation of I A channels in rats.

Animals
Experiments were performed on male Sprague-Dawley rats weighing 200-250 g. Rats were kept under standard laboratory conditions with food and water ad libitum. They were housed three per cage and maintained on a 12:12 h light: dark schedule at a constant ambient temperature (24 ± 1°C). All experimental procedures were approved by the Institutional Animal Care and Use Committee at the Second Military Medical University.
For animal behavioral tests, we performed a peripheral target injection to the TG via the infraorbital foramen as described previously by Neubert [31]. Briefly, at day 9 after surgery, rats were anaesthetized with diethyl ether. A sterile stainless steel needle was inserted medial (1-2 mm) to the palpated portion of the zygomatic process through the infraorbital foramen. The needle was positioned at~10°angle relative to the midline of the head. The tip of the needle was advanced approximately 20 mm along the infraorbital canal and subsequently through the foramen rotundum, then the corresponding drugs were injected. The mechanical pain threshold was then determined every 5 min or 12 h after injection.

P2Y 2 receptor antisense oligodeoxynucleotides (AS-ODN)
P2Y 2 receptor AS-ODN was purchased from Invitrogen Company. Oligonucleotides to rats P2Y 2 were synthesized and purified by Integrated DNA Technologies (ADT). The sequence was as follow: antisense 5′-CCAGGAGTCC AGGCCTGCTGCCATTGCC-3′. The sequences were checked for uniqueness using the National Center for Biotechnology Information's Local Alignment Searchtool (BLAST) based on a previous study [32].

Surgery and behavior test
Chronic constriction injury of the infraorbital branch of trigeminal nerve (ION-CCI) and sham surgery Rats underwent CCI of the right ION as previously described [3]. Briefly, rats were anaesthetized with sodium pentobarbital (50 mg/kg i.p.) and a small incision (approximately 5 mm) was made at the juncture between the zygomatic arch and nasal bone, under the right eye. The muscle was dissected to exposure the infraorbital branch of the TG nerve until it was clearly visible through the incision and two ligatures (4-0 chromic catgut) were loosely tied (with about 2 mm spacing) around it. The loose ligature method was used according to the procedure developed by Imamura et al. [33]: the ligatures reduced the diameter of the nerve by a just noticeable amount and retarded, but did not interrupt the circulation through the superficial vasculature. The incision was sutured with 1.0 silk. Sham-operated rats were treated identically, but no ligatures were applied to the ION. All operations were performed aseptically. After surgery, all rats were maintained in a warm room until they recovered from anesthesia and no antibiotics were administered.

Behavior test
Rats were allowed to acclimate for at least 3 days before use in experiments. All the experiments were carried out between 9:00 and 14:00 in the animal housing room. Before each testing session, animals were placed in individual plastic cages and left to adapt to the environment for at least 15 min. The mechanical pain threshold was tested one day before and every three days after surgery in the two groups. In accordance with our previous study, we injected drugs and performed behavior tests at day 9 after surgery when the mechanical pain threshold of rats was lowest. Each rat received drugs only once and was used in only one experiment. The mechanical pain threshold (PWT) was determined with a rigid von Frey filament coupled with a force transducer (Electrovonfrey, model no: 2391, IITC Inc. Woodland Hills, CA), as previously described [34]. Stimuli were applied within the ION territory, around the center of the vibrissal pad, on the hairy skin surrounding the mystacial vibrissae, ten consecutive times with 2 s each time at 30sec intervals on the nerve-injured side [35]. The PWT was considered as the lowest force of the filaments that produced a brisk head withdrawal, touching or scratching the facial regions upon mechanical stimulation [35]. The mechanical pain threshold were measured every 5 min or 12 h following drug administration. The persons conducting the behavioral measurements were blind to the treatments.

Retrograde labeling of TG neurons innervating the facial skin
TG neurons innervating the facial skin were identified for electrophysiology and immunohistochemistry using the fluorogold (FG)-labeling method [36]. A FG solution (2% in distilled water, 50 μl) was injected into the facial skin in the bilateral region of the whisker pad using a 31-gauge microsyringe.

Cell culture
Control male Sprague-Dawley rats were used for patchclamp recording and RT-PCR analysis. Rats were decapitated and bilateral TG were rapidly removed, de-sheathed, cut and incubated in 2 ml Ca 2+ -and Mg 2+ -free Hanks' balanced salt solution with 10 mM HEPES buffer (pH 7.4) (HBSS; Life Technologies) containing 1.5 mg/ml collagenase (Class II, Worthington Biochemical Corporation, UK) and 6 mg/ml bovine serum albumin (Sigma Chemical Co., Poole, UK) at 37°C in a shaking bath (170 rpm) for 30 min. This was followed by incubation in 2 ml HBSS containing 1 mg/ml trypsin (Sigma) at 37°C in a shaking bath (170 rpm) for 10 min. The solution was replaced with 1 ml growth medium comprising L-15 medium supplemented with 10% bovine serum, 50 ng/ml nerve growth factor, 0.2% NaHCO 3 , 5.5 mg/ml glucose, 200 i.u./ml penicillin and 2 g/ml streptomycin. The ganglia were dissociated into single neurons by gentle mechanical trituration and plated onto 35 mm Petri dishes coated with 10 μg/ml laminin (Sigma) [37]. Cells were maintained at 37°C in a humidified atmosphere containing 5% CO 2 . All neurons were studied after 16 h removal from the animals. For RT-PCR, cultured TG neurons were incubated in growth medium without 10% bovine serum, in order to eliminate glia cells.
Real-time quantitative reverse transcription-polymerase chain reaction analysis RNA extraction and RT-PCR Total RNA was extracted using an RNeasy Mini Kit (QIAGEN; Clifton Hill, Australia). RNA purity was determined using a method of ultraviolet spectrophotometry at a wavelength of 260-280 nm. 2 μg of total RNA was reversely transcribed to complementary DNA in a 20 μl reaction mixture containing 1× reverse transcriptase buffer (

Whole-cell patch clamp recording
Whole-cell patch-clamp recording was undertaken at room temperature with an Axopatch 200B amplifier (Axon Instruments, Foster City, CA, USA). Membrane potential held at −60 mV, signals were filtered at 2 kHz (−3 dB frequency, Bessel filter, 80 dB per decade), then digitized at 10-50 kHz (Digidata 1320A interface, Axon Instruments). The leak current was subtracted from the potassium currents using Clampfit programs. Patch electrodes had resistance of 2-5 MΩ. For voltagerecordings, the pipette solution contained the following (in mM): K gluconate 120, KCl 10, NaCl 5, MgCl 2 •6H 2 O 2, CaCl 2 •2H 2 O 1, HEPES 10, EGTA 11, Mg-ATP 2, Li-GTP 1 (pH adjusted to 7.4 with KOH). The external solution contained (in mM): NaCl 145, KCl 3, CaCl 2 •2H 2 O 1, MgCl 2 •6H 2 O 2, HEPES 10, glucose 10, (pH adjusted to 7.4 with NaOH). Test solutions bathing the cytoplasmic face of the patch membrane contained (in mM): NMDG 145, TEA 25, KCl 3, MgCl 2 •6H 2 O 0.6, CdCl 2 1, CaCl 2 •2H 2 O 2.5, HEPES 10, glucose 10 (pH adjusted to 7.4 with tris-base and 300 mOsM). CdCl 2 was included to block voltage-gated calcium channels. NMDG and TEA were included to reduce currents from voltagegated sodium channels, I K currents, hyperpolarizationactivated cation channels, and capsaicin-induced inward currents [27,38]. A protocol was used as previous described [39], briefly, a pre-pulse (−120 mV, 100 ms) was followed by test pulses (400 ms) from −60 to +60 mV with 10 mV increments, and only those cells that exhibited minimal outward currents during the pre-pulse were analyzed. For current-recordings, action potentials were recorded under current-recordings. During a 400ms injection of a positive current (ranging from −40 to 450 pA), a single action potential could be evoked, depending on the type of neuron (Aβ-, Aδ-and C-units) [3], for example Aδ-units were frequently encountered at a later period after ION-CCI. Cultured TG neurons with soma diameters ranging from 18 to 39 μm were used for action potential recording, for they are consistent with nociceptive Aδ-and C-neurons [39]. Those neurons with retrograde labelling were used for I A recording. The amplitude of the I A was measured at the peak. Whole-cell current-voltage (I-V) curves for individual neurons were generated by calculating the peak outward current at each testing potential and normalizing to the cell capacitance.

Immunofluorescence histochemistry
Rats were anaesthetized with sodium pentobarbital (50 mg/kg i.p.) and perfused transcardially with 0.1 M phosphate buffer solution (PBS), pH 7.4 and subsequently with fresh 4% paraformaldehyde phosphate buffer (PB) solution. After the perfusion, TGs were harvested and fixed in 4% paraformaldehyde for 4 hours. They were then transferred into 20% sucrose for at least 3 days. Series frozen transverse sections (10 μm thick) were made through the TG with a cryostat (Leica, CM1850, Germany), collected and then washed 3 × 5 min in cold PBS. The preparations were then preincubated in antiserum solution 1 (10% normal bovine serum, 0.2% Triton X-100, 0.4% sodium azide in 0.01 mol/l PBS pH 7.

Statistical analysis
All data are presented as means ± SEM. The electrophysiological data were analyzed using the clampfit 9.0 and origin 7.0. For current-clamp recording, differences between the means of action potentials were tested for significance using unpaired Student's t-tests. For voltageclamp recording and animal behavior test, differences between the means were tested for significance using repeated measures ANOVA followed by Dunnett's analysis. For RT-PCR results, differences among groups were tested for significance using two way ANOVA followed by Dunnett's analysis. For Western-blot and immunofluorescence histochemical results, differences among groups were tested for one-way ANOVA followed by Tukey's HSD and unpaired Student's t-tests. Differences were considered as statistically significant when the p value was lower than 0.05 (p < 0.05).

UTP induces hyperalgesia in control rats
We investigated the role of UTP, an agonist of P2Y 2 and P2Y 4 receptors on control rats. UTP (100 nM, 50 μl) significantly decreased the mechanical pain threshold of the whisker pad 20 min after injection and this remained at least for 3 h ( Figure 1A, n = 8 for each group, p < 0.01); it then declined 9 h after the injection ( Figure 1A).

UTP enhances the excitability of small-diameter TG neurons in control rats
Based on the results of the animal behavioral test, we explored the effect of P2Y 2 receptor activation on the electrophysiological properties of small-diameter TG neurons of rats. Action potentials generated in these TG neurons by square-pulse stimulation are illustrated in Figure 1B. Incubation of TG neurons from control rats with UTP 30 or 100 μM for 16 h, caused dosedependent decrease in the mean threshold intensities for evoking action potentials (UTP30 group: 96.3 ± 21.2 pA; UTP100 group: 42.5 ± 13.1 pA; control group: 171.3 ± 23.2 pA; n = 8, Figure 1B,C, p < 0.05 or p < 0.01 vs control). The decrease of mean threshold intensities for evoking action potentials by UTP 30 μM was reversed by co-incubated with suramin 100 μM (UTP30 group: 96.3 ± 21.2 pA; suramin group: 177.5 ± 11.8 pA; n = 8, Figure 1B,C, p < 0.05). As shown in Figure 1C, the mean number of spikes evoked in the UTP-incubated TG neurons during depolarizing step pulses at 2 × threshold was significantly higher in the UTP-incubated TG neurons than those in the control neurons but did not show any dose-dependent changes (control: 1 spikes/400 ms,; UTP-incubated TG neurons (30 or 100 μM): 3.1 ± 0.9 spikes/400 ms or 3.0 ± 0.8 spikes/400 ms; n = 8, p < 0.05). The increase of mean number of spikes by UTP 30 μM was blocked by co-incubated with suramin 100 μM ( Figure 1B,C, n = 8, p < 0.05 vs UTP 30 group).
Activation of P2Y 2 receptors mediates a functional inhibition of I A channels by UTP in FG-labeled small-diameter TG neurons in control rats FG-labeled TG neurons are illustrated in Figure 2A. We observed whether activation of P2Y 2 receptors could functionally inhibit I A subunits in these TG neurons. For voltage-clamp experiments, typical waveforms of depolarization-activated I A are shown in Figure 2B. After incubation with UTP (30 μM) for 16 h, the mean peak amplitude of I A was significantly suppressed compared with that of control (0.12 ± 0.01 nA vs 0.06 ± 0.01 nA, n = 12, p < 0.01). The suppression of peak amplitudes of I A by UTP (30 μM) was then blocked by co-application of suramin ( Figure 2B, con: 0.14 ± 0.01 nA, n = 9; UTP: 0.09 ± 0.01 nA, n = 20, p < 0.05; suramin: 0.13 ± 0.01 nA, n = 9, p < 0.01). We did not see any dose-dependent changes in I A when using UTP 100 μM (0.08 ± 0.01 nA, n = 20).

Activation of P2Y 2 receptors mediates an inhibition of I A channels through ERK pathways on small-diameter TG neurons in control rats
Western blot results showed that the level of ERK1/2 phosphorylation was significantly increased in the ipsilateral TG after ION-CCI compared with that from the sham groups ( Figure 7A, n = 5 for each group, p <0.05). It has been reported that ERK activation contributes to changes in membrane excitability as a result of direct or indirect phosphorylation of kinases, key receptors, and ion channels [30]. ERK inhibitors (U0126 or PD98059) enhance A-type potassium currents in dorsal horn neurons 2 were significantly decreased in the saline group of ION-CCI rats compared with the sham rats. They were reversed after P2Y 2 receptor antisense oligodeoxynucleotides treatment. n = 5-9 rats, *, p < 0.05, **p < 0.01 compared with saline groups. There was no difference in the levels of Kv4.3 mRNA among the groups. n = 6-8 rats, p > 0.05. of the spinal cord [40,41], indicating the possible modulation of ERK in potassium channels. We therefore tested whether the presence of an ERK inhibitor, U0126, would modulate the effect of UTP on trigeminal pain perception. In TG from ION-CCI rats, treatment with P2Y 2 receptor AS-ODN (15 μg/50 μl) significantly decreased ERK expression at the protein level ( Figure 7B, n = 5 for each group, p < 0.01). The inhibitory effect of UTP on I A was significantly reversed ( Figure 8A). The mean peak amplitude of was reversed to 0.15 ± 0.03 nA in the U0126 group (100 μM, n = 11), which was significantly different from that of the UTP group (30 μM, 0.09 ± 0.01 nA, n = 20, p < 0.05, Figure 8B). Further, in the cultured ION-CCI TG neurons, in the presence of U0126 (100 μM), the mean threshold intensities for evoking action potentials was significantly increased (control: 85 ± 14 pA; U0126: 182.9 ± 12.1 pA; n = 8, Figure 8C,D, p < 0.01 vs control), while the number of action potentials was significantly decreased (Figure 8C,D, n = 7, p < 0.05 vs control).

Discussion
Four major findings arise from this study, 1) P2Y 2 receptors and Kv1. These results provide evidence that the down-regulation of I A -related potassium channels by activation of P2Y2Rs in TG neurons potentiates neuronal excitability which then contributes to trigeminal neuropathic pain.

Activation of P2Y 2 receptors enhances TG neuron excitability through suppression of I A channels in control rats
Growing evidence indicates that P2X and P2Y receptormediated signaling critically contributes to the development and maintenance of neuropathic pain [12,13]. Here, we have demonstrated that activation of P2Y 2 receptors leads to a significant increase in the excitability of TG neurons. Previous studies have shown that an increase in membrane excitability in DRG neurons was a cellular-correlate of enhanced nociceptive behavior [18,42,43]. ATP is a non-selective agonist for several ionotropic P2X and metabotropic P2Y receptor subtypes [44]. Usually, ATP released from healthy cells plays a ** ** A B Figure 7 Role of ERK pathway in activation of P2Y 2 receptors mediates an inhibition of I A channels on small-diameter TG neurons in control rats. (A) Comparison of the phosphorylation of ERK1/2 in TG from sham and ION-CCI rats. Western blot results showed that the level of ERK1/2 phosphorylation was significantly increased in the ipsilateral TG after ION-CCI, compared with that from the sham group. n = 5 for each group *, p < 0.05. (B) In TG from ION-CCI rats, treatment with P2Y 2 receptor antisense oligodeoxynucleotides (15 μg/50 μl) significantly decreased the expression of ERK protein in TG. n = 5 for each group, **, p < 0.01.
physiological role [45]. In pathological conditions, ATP release can be evoked from sensory neurons and it produces fast excitatory potentials in DRG cells [46]. Thus, endogenously released ATP from damaged cells [47] may contribute to the ectopic firing of Aβ and Aδ neurons and lead to the development of allodynia [48]. In the present study, ATP caused a large decrease in the mean threshold intensities for evoking action potentials and a significant increase in the mean number of spikes in control TG neurons, which is consistent with a previous study [3]. Although there are (controversial) conflicting outcomes following the use of UTP via P2Y 2 receptors for neuropathic pain [16][17][18][19]48], this study demonstrated that UTP caused a large decrease in the mean threshold intensities for evoking action potentials and a significant increase in the mean number of spikes in control TG neurons. UTP has a similar effect on sensory neurons and thus plays a key role in the development of mechanical allodynia [48]. These results suggest that nucleotides enhance the excitability of TG neurons, probably via both P2Y 2 and P2Y 4 receptors, because UTP is a P2Y 2 /P2Y 4 receptor agonist.
Kv channels are crucial in the control of neuronal excitability, and their down-regulation leads to an increase of neuronal excitability [26,49,50]. Homomeric Kv1.4 channels predominate in Aδ and C fibers arising from small-diameter DRG neurons [28]. Morgan et al. [51] reported that Kvl.4 and Kv4.2, which form transient (Atype) K + channels, may regulate synaptic transmission via presynaptic or postsynaptic mechanisms, respectively. The present electrophysiological study found that UTP mediated a functional inhibition of I A channels in FG-labeled small-diameter TG neurons in control rats. UTP-induced depression of I A was blocked by suramin, hence, the P2Y 2 nucleotide receptor must have contributed for the following reasons: (1) UTP, a P2Y 2 /P2Y 4 receptor agonist enhanced the excitability of TG neurons and inhibited I A . (2) ATP and UTP were about equipotent as observed for rat P2Y 2 and P2Y 4 receptors [52].
Inhibition of I A can increase the firing frequency and broaden the action potential leading to increased Ca 2+ influx and neurotransmitter release [33,50,54]  by nociceptive DRG neurons where Kv4.3 appeared selectively in the soma of a subset of non-peptidergic nociceptive DRG neurons, and reduced expression of Kv4.3 in pain-sensing neurons may induce neuropathic pain [26]. Hu et al. [55] found that genetic elimination of Kv4.2 reduced I A and increased excitability of dorsal horn neurons. The expression of mRNA for Kv1.4, Kv3.4, Kv4.2, and Kv4.3 was markedly reduced in diabetic neuropathic rats [27]. Combined with our electrophysiological data, the down-regulation of I A subunits, including mRNA for Kv1.4, Kv3.4, Kv4.2, and Kv4.3, after application of UTP, could account for the reduced I A observed in UTP-incubated small diameter TG neurons from control rats. Suramin reversed the UTP-induced effect on TG neurons in control rats, further suggesting that P2Y 2 receptors were involved.

The involvement of P2Y 2 receptors in mechanical allodynia in ION-CCI rats
In this study, we found the expression of Kv1.4, Kv3.4, Kv4.2 and Kv4.3 on P2Y 2 receptor-positive TG neurons significantly decreased after ION-CCI compared with those in the sham group. Expression of P2Y 2 receptors, Kv1.4, Kv3.4, Kv4.2 and Kv4.3 was significantly reduced in ION-CCI rats. These data imply that I A channel expression levels of nociceptors and nerve ligation-induced neuropathic pain could be closely related.
The present study showed that activation of P2Y 2 receptors could suppress I A channels in control rats, which might be one of the mechanisms of hyperexcitability of TG neurons after UTP application. We hypothesized that block of P2Y 2 receptors could relieve trigeminal neuropathic pain. Firstly, we confirmed that suramin led to a time-and dose-dependent decrease in pain-related behavior of ION-CCI rats. Some similar observations were reported concerning the analgesic effects of suramin in animal pain models [56,57]. Because suramin is an antagonist of P2Y receptors except P2Y 4 and P2Y 6 receptors [57], the results suggest that P2Y 1 , P2Y 2 , P2Y 11 , P2Y 13 and P2Y 14 receptors could affect pain-related behavior in ION-CCI rats. Considering the effect of UTP in control rats, we concluded that P2Y 2 receptors were probably involved in ION-CCI-induced pain behavior. Secondly, injection of P2Y 2 receptor AS-ODN significantly alleviated mechanical hypersensitivity 6 h after injection, which remained until 120 h. The results further support that block of P2Y 2 receptors could relieve trigeminal neuropathic pain.
To test whether there is a correlation between mechanical sensitivity and I A channel expression, we measured the mRNA levels of the I A -related potassium channels, Kv1. 4 subunits were markedly reduced after ION-CCI, which were then reversed after selective knockdown of P2Y 2 receptor gene expression. It has been reported that there is a close relationship between P2Y and Kv channels. ATP and UTP reversibly inhibited the voltage-gated K + currents in Xenopus embryo spinal neurons [58]. KCNQ1/KCNE1 K + channels and P2Y 4 receptors are coexpressed from the time of birth in the apical membrane of rat strial marginal cells [59]. Purinergic P2Y agonists suppress M currents (I M ), which are generated by Kv7 [18,60]. Our results suggest that activation of P2Y 2 receptors could result in the development of mechanical hypersensitivity, a major symptom of neuropathic pain, which could be as a result of the suppression of the mRNA expression of Kv1.4, Kv3.4 and Kv4.2 subunits.
In the present study, the expressions of Kv4.3 in mRNA and protein levels were decreased after application of UTP in cultured TG neurons from control rats ( Figure 6A,B), but did not change in TG after ION-CCI ( Figure 6C). This may be because: (1) Kv4.3 channels were not prominent in the development of allodynia in ION-CCI rats; and (2) an increase of Kv4.3 channels in glial cells surrounding the neurons in TG compensated for the changes in TG neurons after ION-CCI. Expression of I A -related K V channels, such as Kv4.1, in glial cells suggests that glial cells also play an important role in chronic pain [61,62]. Further research of Kv channels on TG glial cells is required to explain how I A channels are involved in trigeminal neuropathic pain. ERK1/2 is the downstream kinase for the effect of P2Y 2 receptors on I A channels P2Y 2 receptors are G protein-coupled receptors that usually activate PLC-β via G αq , which results in the release of intracellular Ca 2+ and activation of PKC [63]. These events further activate extracellular signalregulated kinase (ERK), including ERK1 and ERK2 [64][65][66]. ERK and Kv4.2 have a functional link at both the cellular and behavioral levels [67]. Phosphorylation of Kv4.2 by PKC enhanced ERK phosphorylation of the channel in vitro. These findings suggest the possibility that Kv4.2 is a locus for PKC and ERK cross-talk [68]. Kv4.3 positive neurons also expressed ERK2 and mGluR5, suggesting that Kv4.3 subunits could be involved in pain modulation [69]. In line with the previous report [70], we found that ION-CCI significantly increased the level of ERK1/2 phosphorylation in TGs. Evidence provided in this study further suggests that the inhibition of I A channels through P2Y 2 receptors is modulated by ERK signaling after ION-CCI. First, I A was significantly inhibited by UTP, which could be reversed when ERK signaling was blocked by U0126. Second, in ION-CCI rats, the expression of ERK in protein level was increased and the mRNA expressions of Kv1.4, Kv3.4 and Kv4.2 subunits were decreased, which were then reversed by P2Y 2 receptor AS-ODN treatment. A recent study has shown that the PI3K/Akt signaling pathway can be activated by P2Y 2 receptors [71]. The PI3K/Akt signaling pathway and Kv channels are both involved in the same disease [72]. Although we could not exclude that other pathways contribute to this effect of UTP, the ERK signaling pathway might be one of the downstream pathways for the effect of P2Y 2 receptors on I A channels, which might contribute to the development of trigeminal neuropathic pain.
In the present study, the effect of UTP on mechanical pain threshold in normal rats started from 10 min, suggesting the pathway without alterations of gene expression. The possibilities could be through facilitating homomeric P2X 2 [19], P2X 3 [19,73], or TRPV1 receptors [16,74]. Further, the long-term effect of UTP (more than 30 min) in pain behavior study and antisense oligodeoxynucleotides effect on ION-CCI rats indicate the alterations of gene expression. Although the underlying mechanisms are not fully understood, inhibition of P2Y 2 receptors leads to down-regulation of ERK-mediated phosphorylation and increase of the expression of I A -related Kv channels in trigeminal ganglion neurons, which might contribute to the clinical treatment of trigeminal neuropathic pain. Taken together, these data suggest that P2Y 2 receptors on TG might play an important role in initiating and maintaining the allodynia in trigeminal neuropathic pain.