Enzymatic conversion of ATP to adenosine contributes to ATP-induced inhibition of glutamate release in rat medullary dorsal horn neurons

Neuropharmacology 93 (2015) 94e102

Contents lists available at ScienceDirect

Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm

Enzymatic conversion of ATP to adenosine contributes to ATP-induced inhibition of glutamate release in rat medullary dorsal horn neurons In-Sun Choi a, Jin-Hwa Cho a, Maan-Gee Lee b, c, Il-Sung Jang a, c, * a

Department of Pharmacology, School of Dentistry, Kyungpook National University, Daegu 700-412, Republic of Korea Department of Pharmacology, School of Medicine, Kyungpook National University, Daegu 700-412, Republic of Korea c Brain Science & Engineering Institute, Kyungpook National University, Daegu 700-412, Republic of Korea b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 October 2014 Received in revised form 16 January 2015 Accepted 20 January 2015 Available online 3 February 2015

Purine nucleotides, such as ATP and ADP, activate ionotropic P2X and metabotropic P2Y receptors to regulate neurotransmitter release in the peripheral as well as central nervous system. Here we report another type of ATP-induced presynaptic modulation of glutamate release in rat medullary dorsal horn neurons. Glutamatergic excitatory postsynaptic currents (EPSCs) induced by electrical stimulation of trigeminal tract were recorded from horizontal brain stem slices using a whole-cell patch clamp technique. ATP decreased the amplitude of glutamatergic EPSCs in a reversible and concentration dependent manner and increased the paired-pulse ratio. In addition, ATP reduced the frequency of miniature EPSCs without affecting the current amplitude, suggesting that ATP acts presynaptically to reduce the probability of glutamate release. The ATP-induced decrease in glutamatergic EPSCs was not affected by P2X and P2Y receptor antagonists, but was completely blocked by DPCPX, a selective adenosine A1 receptor antagonist. The ATP-induced decrease in glutamatergic EPSCs was also inhibited by an inhibitor of tissue nonspecific alkaline phosphatase but not by inhibitors of other enzymes such as ecto-nucleoside triphosphate diphosphohydrolases and ecto-5’-nucleotidases. The results suggest that exogenously applied purine nucleotides are rapidly converted to adenosine by specific enzymes, and subsequently act on presynaptic A1 receptors to inhibit glutamate release from primary afferent terminals. This type of modulation mediated by purine nucleotides may play an important role in regulating nociceptive transmission from orofacial tissues. © 2015 Elsevier Ltd. All rights reserved.

Keywords: ATP Presynaptic inhibition Primary afferents Medullary dorsal horn Tissue nonspecific alkaline phosphatases Patch clamp

1. Introduction ATP is an extracellular signaling molecule that acts as a fast neurotransmitter in the peripheral and central nervous system (Burnstock, 2007). ATP is involved in a number of physiological functions, such as neuroprotection, locomotion, development, and pain, through the activation of multiple types of P2 receptors (e.g.,

ionotropic P2X and metabotropic P2Y receptors) expressed on neuronal membranes (Ralevic and Burnstock, 1998; Burnstock, 2007). P2 receptors are also expressed on presynaptic nerve terminals, and regulate the release of a variety of neurotransmitters (Rodrigues et al., 2005; Dorostkar and Boehm, 2008; Gonçalves and Queiroz, 2008). Moreover, extracellular ATP can be rapidly hydrolyzed to ADP, AMP, and adenosine via multiple types of extracellular

Abbreviations: ab-me-ATP, ab-methylene-ATP; APV, DL-2-amino-5-phosphonovaleric acid; ARL67156, 6-N,N-diethyl-b-g-dibromomethylene-D-adenosine-50 -triphosphate; BBG, Brilliant Blue G; Bz-ATP, 20 -30 -O-(4-benzoylbenzoyl)-ATP; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; DRG, dorsal root ganglia; ENT1, equilibrative nucleoside transporter 1; EPSCs, excitatory postsynaptic currents; KeS test, KolmogoroveSmirnov test; mEPSCs, miniature EPSCs; MRS2179, 2’-deoxy-N6-methyl adenosine 30 ,5’diphosphate; MRS2211, 2,2-dimethyl-propionic acid 3-(2-chloro-6-methylaminopurin-9-yl)-2-(2,2-dimethyl-2-[(2-chloro-5-nitrophenyl)azo]-5-hydroxy-6-methyl-3[(phosphonooxy)methyl]-4-pyridinecarboxaldehyde; MRS2395, propionyloxymethyl)-propyl ester; NBMPR, S-(4-nitrobenzyl)-6-thioinosine; NT5E, ecto-5’-nucleotidases; NTPDases, ecto-nucleoside triphosphate diphosphohydrolases; PAP, prostatic acid phosphatase; PPADS, pyridoxalphosphate-6-azophenyl-20 ,4’-disulfonic acid; PPR, pairedpulse ratio; SR95531, 6-imino-3-(4-methoxyphenyl)-1(6H)-pyridazinebutanoic acid; TG, trigeminal ganglia; TNAP, tissue nonspecific alkaline phosphatase; TTX, tetrodotoxin; VDCCs, voltage-dependent Ca2þ channels. * Corresponding author. Department of Pharmacology, School of Dentistry, Kyungpook National University, 188-1, Samduk 2 ga-dong, Jung-gu, Daegu 700-412, Republic of Korea. Tel.: þ82 53 660 6887; fax: þ82 53 424 5130. E-mail address: [email protected] (I.-S. Jang). http://dx.doi.org/10.1016/j.neuropharm.2015.01.020 0028-3908/© 2015 Elsevier Ltd. All rights reserved.

I.-S. Choi et al. / Neuropharmacology 93 (2015) 94e102

enzymes, including ecto-nucleoside triphosphate diphosphohydrolases (NTPDases), ecto-50 -nucleotidases (NT5E, also known as CD73), prostatic acid phosphatase (PAP), and tissue nonspecific alkaline phosphatase (TNAP) (Dunwiddie et al., 1997; Cunha et al., 1998; Zimmermann, 2000; Zylka et al., 2008; Street et al., 2013). Because NT5E and PAP hydrolyze AMP to adenosine (Zylka et al., 2008; Sowa et al., 2010), these enzymes require ATP or ADP to first be degraded to AMP by NTPDases. In contrast, TNAP can hydrolyze ATP, ADP, and AMP to adenosine (Zimmermann, 2000; Street et al., 2013). The resultant adenosine activates P1 adenosine receptors to modulate the excitability of peripheral and central neurons (Fredholm et al., 2005). ATP is closely involved in nociceptive transmission, because sensory neurons, such as dorsal root ganglion (DRG) and trigeminal ganglion (TG) neurons, and primary afferents express multiple types of P2X, P2Y, and adenosine receptors (Gerevich and Illes, 2004; Donnelly-Roberts et al., 2008; Burnstock, 2009a, 2009b). In addition, endogenous ATP can be released from primary afferent terminals and interneurons onto spinal dorsal horn neurons (Bardoni et al., 1997; Jo and Schlichter, 1999), and it acts presynaptically to modulate excitatory sensory synaptic transmission. For example, ATP acts on presynaptic P2X1 and/or P2X3 receptors and facilitates spontaneous glutamate release onto spinal dorsal horn neurons (Gu and MacDermott, 1997; Nakatsuka and Gu, 2001), suggesting that ATP via the activation of P2X receptors plays a pronociceptive role in the spinal cord. In contrast, the activation of P2Y1 receptors reduces membrane currents mediated by voltagedependent Ca2þ channels (VDCCs) in DRG neurons (Gerevich et al., 2004), suggesting that P2Y1 receptors can inhibit primary afferent synaptic transmission in spinal dorsal horn neurons. Furthermore, since ATP hydrolyzing enzymes are found within the DRG and spinal dorsal horn (Zylka et al., 2008; Sowa et al., 2010), ATP via the activation of P1 receptors might affect nociceptive transmission in the spinal cord. While TG neurons and their afferents also express a number of purinoceptors (Chen et al., 1995; Xiang et al., 1998; Ruan and Burnstock, 2003), much less is known about the functional role of ATP in trigeminal nociceptive transmission from peripheral tissues. In the present study, therefore, we directly addressed the effect of ATP on primary afferentevoked glutamatergic transmission using horizontal brain stem slices. 2. Materials and methods 2.1. Preparation All experiments complied with the guiding principles for the care and use of animals approved by the Council of the Physiological Society of Korea and the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and every effort was made to minimize both the number of animals used and their suffering. Sprague Dawley rats (12e16 d old, either sex) were decapitated under ketamine anesthesia (100 mg/kg, i. p.). The brain stem was dissected and horizontally sliced at a thickness of 400 mm by use of a microslicer (VT1000S; Leica, Nussloch, Germany) in a cold artificial cerebrospinal fluid (ACSF; 120 NaCl, 2 KCl, 1 KH2PO4, 26 NaHCO3, 2 CaCl2, 1 MgCl2 and 10 glucose, saturated with 95% O2 and 5% CO2). Slices were kept in an ACSF saturated with 95% O2 and 5% CO2 at room temperature (22e25  C) for at least 1 h before electrophysiological recording. Thereafter, the slices were transferred into a recording chamber, and both the medullary dorsal horn region and trigeminal roots were identified under an upright microscope (E600FN, Nikon, Tokyo, Japan) with a water-immersion objective (40). The ACSF routinely contained 10 mM SR95531, 1 mM strychnine, 50 mM APV to block GABAA, glycine and NMDA receptors, respectively. The bath was perfused with ACSF at 2 ml/min by the use of a peristaltic pump (MP-1000, EYELA, Tokyo, Japan). 2.2. Electrical measurements All electrical measurements were performed by use of a computer-controlled patch clamp amplifier (MultiClamp 700B; Molecular Devices; Union City, CA, USA). For whole-cell recording, patch pipettes were made from borosilicate capillary glass (1.5 mm outer diameter, 0.9 mm inner diameter; G-1.5; Narishige, Tokyo,

95

Japan) by use of a pipette puller (P-97; Sutter Instrument Co., Novato, CA, USA). The resistance of the recording pipettes filled with internal solution (in mM; 140 CsMeHSO3, 5 TEA-Cl, 5 CsCl, 2 EGTA, 2 Mg-ATP and 10 Hepes, pH 7.2 with Tris-base) was 4e6 MU. Membrane currents were filtered at 2 kHz (MultiClamp Commander; Molecular Devices), digitized at 10 kHz (Digidata 1322A, Molecular Devices), and stored on a computer equipped with pCLAMP 10.0 (Molecular Devices). In wholecell recordings, 10 mV hyperpolarizing step pulses (30 ms in duration) were periodically delivered to monitor the access resistance, and recordings were discontinued if access resistance changed by more than 15%. All electrophysiological experiments were performed at room temperature (22e25  C). To record action potential-dependent glutamatergic excitatory postsynaptic currents (EPSCs), a glass stimulation pipette (~10 mm diameter) filled with a bath solution, was positioned around the trigeminal root (see Choi et al., 2011, 2012). Brief paired pulses (500 ms, 1e5 V, 20 Hz) were applied by the stimulation pipette at a frequency of 0.1 Hz using a stimulator (SEN-7203, Nihon Kohden, Tokyo, Japan) equipped with an isolator unit (SS-701J, Nihon Kohden). 2.3. Data analysis The amplitudes of action potential-dependent glutamatergic EPSCs were calculated by subtracting the baseline from the peak amplitude. The conduction velocity of primary afferents innervating medullary dorsal horn neurons was calculated by dividing the distance between stimulation and recording sites by the latency of EPSCs (see Choi et al., 2012). The effect of drugs on EPSCs was quantified as a percentage change in EPSC amplitude compared to the control values. Miniature EPSCs (mEPSCs) were counted and analyzed using the MiniAnalysis program (Synaptosoft, Inc., Decatur, GA, USA) as described previously (Jang et al., 2002). The average values of both the frequency and amplitude of mEPSCs during the control period (10e20 min) were calculated for each recording, and the frequency and amplitude of all the events during the ATP or ADP application (5 min) were normalized to these values. The effects of these different conditions were quantified as a percentage increase in mEPSC frequency compared to the control values. The inter-event intervals and amplitudes of a large number of synaptic events obtained from the same neuron were examined by constructing cumulative probability distributions and compared using the KolmogoroveSmirnov (KeS) test with Stat View software (SAS Institute, Inc., Cary, NC, USA). The continuous curves for the concentrationeinhibition relationship were fitted using a least-squares fit to the following equation:  
 n I ¼ 1  C n C n þ EC50 ; where I is the ATP- or ADP-induced inhibition of EPSC amplitude, C is the concentration of ATP or ADP, EC50 is the concentration for the half-effective response and n is the Hill coefficient. Numerical values are provided as the mean and standard error of the mean (SEM) using values normalized to the control. Significant differences in the mean amplitude and frequency were tested using the Student's two-tailed paired t-test, using absolute values rather than normalized ones. Values of p < 0.05 were considered significant. 2.4. Drugs The drugs used in the present study were ATP, ATPgS, ADP, ADPbS, UTP, UDP, abme-ATP, Bz-ATP, adenosine, DPCPX, suramin, PPADS, BBG, strychnine, APV, CNQX, ARL67156, MRS2179, MRS2395, S-(4-nitrobenzyl)-6-thioinosine (NBMPR) (from Sigma, St. Louis, MO, USA) and SR95531, MRS2211, TTX, POM-1, baclofen (from Tocris, Bristol, UK), and 2,5-dimethoxy-N-(quinolin-3-yl)benzenesulfonamide (TNAP inhibitor, TNAP-I) (from Merck Milipore, Darmstadt, Germany), adenosine deaminase (from Worthington, Lakewood, NJ, USA). All drugs were applied by bath application. In a subset of experiments, ATP was applied using the ‘Yetube system’ for rapid solution exchange (Murase et al., 1989).

3. Results 3.1. ATP acts presynaptically to inhibit glutamate release in medullary dorsal horn neurons Action potential-dependent EPSCs were recorded from medullary dorsal horn neurons at a VH of 60 mV by electrical stimulation through a glass pipette placed to the spinal trigeminal tract. These EPSCs were mediated by AMPA/KA receptors because 10 mM CNQX, a selective AMPA/KA receptor antagonist, completely blocked all synaptic currents (data not shown, see also Choi et al., 2011). Under these conditions, the effect of ATP on glutamatergic EPSCs evoked by paired stimulation at an interval of 50 ms (20 Hz) was observed. As shown in Fig. 1A and B, bath applied ATP (100 mM), an endogenous P2 receptor agonist, reversibly decreased the first EPSC (EPSC1)

96

I.-S. Choi et al. / Neuropharmacology 93 (2015) 94e102

Fig. 1. Effect of ATP on primary afferent-evoked glutamatergic EPSCs. A, A typical time course of EPSC1 amplitude (a) and PPR (EPSC2/EPSC1; b) before, during and after application 100 mM ATP. The amplitudes and PPRs of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. B, ATP-induced changes in EPSC1 amplitude (a) and PPR (b). Each column was normalized to the control and represents the mean and SEM from 16 experiments. **; p < 0.01. C, Concentrationeresponse relationships of ATP. The EC50 values for EPSC1 amplitude and PPR calculated from curve fitting results were 50.3 mM and 45.8 mM, respectively. Each point and error bar represents the mean and SEM from 6 to 16 experiments. D, A typical time course of EPSC1 amplitude before, during and after the repetitive application of 100 mM ATP. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. E, A scatter plot of the extent of ATP (100 mM)-induced inhibition of EPSCs against the calculated conduction velocity of primary afferents innervating medullary dorsal horn neurons. The continuous line is the least-squares linear fit (r2 ¼ 0.10, n ¼ 99).

amplitude to 47 ± 4% of the control (n ¼ 16, p < 0.01), and increased the paired-pulse ratio (PPR; EPSC2/EPSC1) from 0.69 ± 0.07 to 1.28 ± 0.15 (n ¼ 16, p < 0.01). In addition, ATP clearly affected glutamatergic EPSCs in a concentration-dependent manner, in which the EC50 values for EPSC1 amplitude and PPR were 50.3 mM and 45.8 mM, respectively (Fig. 1C). The ATP-induced inhibition of glutamatergic transmission was reproducible during repeated applications with a time interval of 25 min (first; 59 ± 6%, second; 61 ± 5%, third; 59 ± 6% of the control, n ¼ 6, Fig. 1D). In medullary dorsal horn neurons responding to ATP (99 of 136 neurons tested; 72.8%), the calculated conduction velocity of primary afferentevoked EPSCs was 1.18 ± 0.07 m/s (range; 0.22e3.01 m/s, n ¼ 99, Fig. 1E). This calculated conduction velocity was similar to that

estimated from Ad- and C-fibers innervating dorsal horn neurons of the spinal cord (Nakatsuka et al., 2000), suggesting that most of the EPSCs recorded in this study originated from Ad- and/or C-fibers. We also examined the effect of ATP on glutamatergic mEPSCs in the presence of 300 nM TTX. ATP (100 mM) significantly decreased the frequency of glutamatergic mEPSCs (55 ± 6% of the control, n ¼ 5, p < 0.01), without affecting the current amplitude (95 ± 4% of the control, n ¼ 5, p ¼ 0.38 (Supplementary Fig. S1). In addition, ATP (100 mM) significantly shifted the cumulative distribution of the inter-event interval to the right (p < 0.01, KeS test) without affecting the cumulative distribution of the current amplitude (p ¼ 0.76, KeS test), consistent with a decrease in the frequency of glutamatergic mEPSCs (Supplementary Fig. S1).

Fig. 2. Effects of P2X receptor agonists on glutamatergic EPSCs. Aa, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP and 100 mM ab-meATP. b, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP in the absence and presence of 20 mM PPADS. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. Ba, ATP- or ab-me-ATP-induced changes in EPSC1 amplitude. b, ATP-induced changes in EPSC1 amplitude in the absence and presence of PPADS. Each column was normalized to the control and represents the mean and SEM from 5 to 7 experiments. **; p < 0.01, n.s; not significant. Ca, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP and 100 mM Bz-ATP. b, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP in the absence and presence of 10 mM BBG. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. Da, ATP- or Bz-ATP-induced changes in EPSC1 amplitude. b, ATP-induced changes in EPSC1 amplitude in the absence and presence of BBG. Each column was normalized to the control and represents the mean and SEM from 6 experiments. **; p < 0.01, n.s; not significant.

I.-S. Choi et al. / Neuropharmacology 93 (2015) 94e102

97

3.2. P2X receptors are not responsible for the ATP-induced inhibition of glutamate release

that ATP reduces glutamatergic transmission via mechanisms independent on P2X receptors.

ATP action can be mediated by P2 receptors, e.g., P2X and P2Y receptors. Although all P2X receptor subtypes except P2X7 are expressed on TG neurons (Xiang et al., 1998), P2X3 receptors are predominantly localized on small nociceptive TG neurons as well as primary afferent fibers (Chen et al., 1995; Xiang et al., 1998; Kim et al., 2008). Therefore, we examined the effect of ab-me-ATP, a selective P2X1, P2X3, and heteromeric P2X2/3 receptor agonist, on glutamatergic EPSCs (Ralevic and Burnstock, 1998). However, abme-ATP (100 mM) had no inhibitory effect on glutamatergic EPSCs (93 ± 5% of the control, n ¼ 5, p ¼ 0.35, Fig. 2Aa and Ba). We also examined the effects of PPADS and suramin, nonselective P2X and P2Y receptor antagonists (Ralevic and Burnstock, 1998), on the ATPinduced inhibition of glutamate release. PPADS (20 mM) had no effect on the basal amplitude of glutamatergic EPSCs (95 ± 3% of the control, n ¼ 7, p ¼ 0.11, Fig. 2Ab and Bb). In the presence of PPADS, however, ATP was still able to decrease EPSC1 amplitude (70 ± 5% of the PPADS condition, n ¼ 7, p < 0.01, Fig. 2Ab and Bb). Suramin (100 mM) also had no effect on the basal amplitude of glutamatergic EPSCs (88 ± 3% of the control, n ¼ 5, p ¼ 0.06, data not shown). In the presence of suramin, ATP was still able to decrease EPSC1 amplitude (49 ± 6% of the suramin condition, n ¼ 5, p < 0.01, data not shown). Although P2X7 receptors are not expressed on TG neurons, ATP can act on P2X7 receptors expressed on glia to affect neuronal activities (Jarvis, 2010). In order to examine the possible involvement of glial P2X7 receptors in the ATP-induced inhibition of glutamate release, we examined the effect of Bz-ATP, a P2X7 and P2Y11 receptor agonist (Abbracchio et al., 2006), on glutamatergic EPSCs. Bz-ATP (100 mM) had no effect on the amplitude of glutamatergic EPSCs (99 ± 3% of the control, n ¼ 6, p ¼ 0.44, Fig. 2Ca and Da). In addition, the ATP-induced inhibition of glutamate release was not affected by 10 mM BBG, a P2X7 receptor antagonist (51 ± 7% of the BBG condition, n ¼ 6, p < 0.01, Fig. 2Cb and Db). The results suggest

3.3. P2Y receptors are not responsible for the ATP-induced inhibition of glutamate release The inhibitory action of ATP on action potential-dependent glutamate release could be mediated by P2Y receptors. Among eight P2Y receptor subtypes, e.g., P2Y1, 2, 4, 6, 11, 12, 13, 14, ATP can activate P2Y1, P2Y2, P2Y4, P2Y11, P2Y12, and P2Y13 receptor subtypes (Abbracchio et al., 2006). Therefore, we examined the effects of selective agonists for these P2Y receptor subtypes on glutamatergic EPSCs. UTP (100 mM), which activates P2Y2, P2Y4, P2Y6 and P2Y11 receptor subtypes (Abbracchio et al., 2006), failed to reduce glutamatergic EPSCs (94 ± 3% of the control, n ¼ 4, p ¼ 0.07, data not shown). In addition, UDP (100 mM), a P2Y4 and P2Y6 receptor agonist (Abbracchio et al., 2006), had no inhibitory effect on glutamatergic EPSCs (99 ± 4% of the control, n ¼ 6, p ¼ 0.88, data not shown). In contrast, ADP, which activates P2Y1, P2Y6, P2Y11, P2Y12 and P2Y13 receptor subtypes (Abbracchio et al., 2006), significantly reduced glutamatergic EPSCs. As shown in Fig. 3A and B, ADP (100 mM) decreased EPSC1 amplitude to 50 ± 8% of the control (n ¼ 6, p < 0.01), and increased the PPR from 0.79 ± 0.13 to 1.39 ± 0.31 (n ¼ 6, p < 0.01), suggesting that ADP also acts presynaptically to decrease the probability of glutamate release. The inhibitory effect of ADP on glutamatergic EPSCs was highly concentration-dependent (EC50 ¼ 45.8 mM, Supplementary Fig. S2). ADP (100 mM) also significantly decreased the frequency of glutamatergic mEPSCs (60 ± 9% of the control, n ¼ 5, p < 0.01), without affecting the current amplitude (103 ± 2% of the control, n ¼ 5, p ¼ 0.20) (Supplementary Fig. S2). In addition, ADP (100 mM) significantly shifted the cumulative distribution of the inter-event interval to the right (p < 0.01, KeS test) without affecting the cumulative distribution of the current amplitude (p ¼ 0.83, KeS test), consistent with a decrease in the frequency of glutamatergic mEPSCs (Supplementary Fig. S2). In medullary dorsal horn neurons

Fig. 3. Effect of ADP on primary afferent-evoked glutamatergic EPSCs. A, A typical time course of EPSC1 amplitude (a) and PPR (b) before, during and after application 100 mM ATP and 100 mM ADP. Insets represent typical traces of the numbered region. B, ADP-induced changes in EPSC1 amplitude (a) and PPR (b). Each column was normalized to the control and represents the mean and SEM from 6 experiments. **; p < 0.01. C, A typical time course of EPSC1 amplitude before, during and after application 100 mM ADP in the absence and presence of 10 mM MRS2179, a selective P2Y1 receptor antagonist. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. D, ADP-induced changes in EPSC1 amplitude in the absence and presence of 10 mM MRS2179 (a; n ¼ 6), 100 mM MRS2211 (b; n ¼ 6), and 30 mM MRS2395 (c; n ¼ 6). **; p < 0.01, n.s; not significant.

98

I.-S. Choi et al. / Neuropharmacology 93 (2015) 94e102

responding to ADP (41 of 41 neurons that responded to ATP), the calculated conduction velocity of primary afferent-evoked EPSCs was 1.39 ± 0.1 m/s (range; 0.46e2.81 m/s, n ¼ 41, Supplementary Fig. S2). These pharmacological results suggest that P2Y1, P2Y12 or P2Y13 receptor subtypes might be involved in the ADP- and ATPinduced inhibition of glutamate release. Therefore, we further examined whether the ADP-induced inhibition of glutamate release is blocked by P2Y receptor subtype selective antagonists, such as MRS2179 (KD ¼ 109 nM for P2Y1 receptors, Boyer et al., 1998), MRS2395 (KD ¼ 3.6 mM for P2Y12 receptors, Xu et al., 2002) and MRS2211 (IC50 z 1 mM for P2Y13 receptors, Kim et al., 2005). The application of each antagonist had no effect on the basal glutamatergic transmission (data not shown). However, ADP (100 mM) decreased EPSC1 amplitude even in the presence of each antagonist (69 ± 9% of the 10 mM MRS2179 condition, n ¼ 6, p < 0.01; 63 ± 10% of the 100 mM MRS2395 condition, n ¼ 6, p < 0.01; 55 ± 11% of the 30 mM MRS2211 condition, n ¼ 6, p < 0.01, Fig. 3C and D). 3.4. ATP inhibits glutamate release via a mechanism dependent on adenosine A1 receptors The above results suggest that either ATP or ADP decreases primary afferent-evoked glutamatergic transmission via a mechanism independent on the activation of P2X and/or P2Y receptors, indicating that ATP action would be mediated by adenosine A1 receptors after enzymatic hydrolysis (Zimmermann, 2000; Masino et al., 2002). Therefore, we examined the effect of DPCPX, a selective A1 receptor antagonist, on the ATP-induced inhibition of glutamate release. The ATP-induced inhibition of EPSCs (49 ± 5% of the control, n ¼ 9, p < 0.01) was completely blocked by 100 nM DPCPX (98 ± 5% of the DPCPX condition, n ¼ 9, p ¼ 0.92, Fig. 4A). In addition, the ADP-induced inhibition of EPSCs (46 ± 7% of the control, n ¼ 7, p < 0.01) was clearly blocked by 100 nM DPCPX (91 ± 5% of the DPCPX condition, n ¼ 7, p ¼ 0.08, Fig. 4B). It should be noted, however, that ADP-induced inhibition of EPSCs was not blocked by 10 mM MRS2179, a P2Y1 receptor antagonist (see Fig. 3C

and D). On the other hand, DPCPX (100 nM) had no effect on the basal EPSC1 amplitude (104 ± 4% of the control, n ¼ 8, p ¼ 0.28) or PPR (0.53 ± 0.04 and 0.58 ± 0.05 for the control and DPCPX conditions, respectively, n ¼ 8, p ¼ 0.96, Supplementary Fig. S3), consistent with a lack of tonic inhibition by endogenous extracellular adenosine. We also found that ATP significantly decreased EPSC1 amplitude even in the presence of 10 mM MRS2179 (63 ± 8% of the MRS2179 condition, n ¼ 7, p < 0.01, data not shown). The results suggest that either ATP or ADP decreases primary afferentevoked glutamatergic transmission in an adenosine A1 receptordependent manner. However, the contamination of adenosine within the ATP or ADP solution could be negligible, because we did not detect any adenosine in either solution using HPLC analysis (data not shown). To further elucidate whether ATP and ADP/adenosine act on the same receptor, we examined the effects of ADP and adenosine on glutamatergic EPSCs after the application of ATP, If ATP and ADP/adenosine act on different receptors to decrease glutamatergic EPSCs, ADP and/or adenosine should show a synergistic inhibitory effect on glutamatergic EPSCs. At a 100 mM concentration, ATP, ADP, and adenosine showed a similar and near maximal inhibitory effect on glutamatergic EPSCs (Supplementary Fig. S3). ATP (100 mM) decreased EPSC1 amplitude to 48 ± 8% of the control (n ¼ 5, p < 0.01, Fig. 4C and D). In the continued presence of 100 mM ATP, ADP (100 mM) did not further decrease EPSC1 amplitude (45 ± 9% of the control, n ¼ 5, p ¼ 0.57, Fig. 4C and D). In addition, in the continued presence of both 100 mM ATP and ADP, adenosine (100 mM) did not further decrease EPSC1 amplitude (42 ± 8% of the control, n ¼ 5, p ¼ 0.15, Fig. 4C and D). However, it is possible that the inhibitory action of ATP on glutamatergic EPSCs was maximal, such that either ADP or adenosine would not be able to further decrease glutamatergic EPSCs. To exclude this possibility, we examined the effect of baclofen, a selective GABAB receptor agonist, on glutamatergic EPSCs after the cumulative application of ATP, ADP and adenosine. Baclofen (10 mM) greatly decreased EPSC1 amplitude to 6 ± 2% of the control (n ¼ 5, p < 0.01, 4C and D) even in the presence of ATP, ADP, and adenosine.

Fig. 4. Effect of DPCPX on the ATP- and ADP-induced decrease in EPSCs. Aa, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP in the absence and presence of 100 nM DPCPX. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. b, ATP-induced changes in EPSC1 amplitude in the absence and presence of DPCPX. Each column was normalized to the control and represents the mean and SEM from 9 experiments. **; p < 0.01, n.s; not significant. Ba, A typical time course of EPSC1 amplitude before, during and after application 100 mM ADP in the absence and presence of 100 nM DPCPX. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. b, ADP-induced changes in EPSC1 amplitude in the absence and presence of DPCPX. Each column was normalized to the control and represents the mean and SEM from 7 experiments. **; p < 0.01, n.s; not significant. C, A typical time course of EPSC1 amplitude before, during and after the cumulative application of 100 mM ATP, 100 mM ADP, 100 mM adenosine, and 10 mM baclofen. Insets represent typical traces of the numbered region. D, Agonist-induced changes in EPSC1 amplitude. Each column was normalized to the control and represents the mean and SEM from 5 experiments. **; p < 0.01, n.s; not significant.

I.-S. Choi et al. / Neuropharmacology 93 (2015) 94e102

3.5. Hydrolysis of ATP to adenosine by TNAP is responsible for the ATP-induced inhibition of glutamate release These results suggest that ATP, after enzymatic hydrolysis, decreases primary afferent-evoked glutamatergic transmission via a mechanism dependent on the activation of adenosine A1 receptors rather than P2X and/or P2Y receptors. ATP hydrolysis is mediated by various enzymes including NTPDases, NT5E, and TNAP (Gonçalves and Queiroz, 2008). Therefore, we first examined the effects of ATPgS and ADPbS, which are breakdown-resistant ATP and ADP analogs, respectively, on glutamatergic EPSCs. The extent of inhibition of EPSC1 amplitude induced by either ATPgS (100 mM) or ADPbS (100 mM) was similar to that induced by ATP (100 mM) or ADP (100 mM) (Fig. 5A and B). However, as both drugs can be hydrolyzed by the aforementioned enzymes (Cunha et al., 1998), we further examined the effects of inhibitors for enzymes involved in the hydrolysis of ATP on the ATP-induced inhibition of glutamate release. A previous study had shown that ARL67156, an NTPDase inhibitor, efficiently blocks the hydrolysis of ATP even at a 10 mM concentration (Nakatsuka and Gu, 2001). The application of ARL67156 (100 mM) did not affect the basal amplitude of glutamatergic EPSCs (94 ± 4% of the control, n ¼ 5, p ¼ 0.10, Fig. 5C). However, ATP decreased EPSC1 amplitude even in the presence of 100 mM ARL67156 (45 ± 5% of the ARL67156 condition, n ¼ 5, p < 0.01, Fig. 5C). Since ARL67156 is known to be less effective on NTPDase 2 (Müller et al., 2006; Wall et al., 2008), we also examined the effect of POM-1, another NTPDase inhibitor (Wall et al., 2008), on the ATP-induced inhibition of glutamate release. The application of POM-1 (30 mM) slightly decreased the basal amplitude of glutamatergic EPSCs (85 ± 7% of the control, n ¼ 6, p ¼ 0.07, Fig. 5D). In the continued presence of 30 mM POM-1, however, ATP (100 mM) still decreased EPSC1 amplitude (42 ± 2% of the POM-1 condition, n ¼ 6, p < 0.01, Fig. 5D). We also examined the effect of ab-me-ADP, an NT5E inhibitor, on ATP-induced inhibition of glutamate release.

99

The application of both 100 mM ARL67156 and 100 mM ab-me-ADP did not affect the basal amplitude of glutamatergic EPSCs (96 ± 6% of the control, n ¼ 3, p ¼ 0.64, data not shown) or PPR (0.62 ± 0.12 and 0.62 ± 0.09 for the control and ARL67156 plus ab-me-ADP condition, respectively, n ¼ 3, p ¼ 0.84, data not shown). In the presence of both 100 mM ARL67156 and 100 mM ab-me-ADP, however, ATP (100 mM) still decreased EPSC1 amplitude (63 ± 18% of the ARL67156 plus ab-me-ADP condition, n ¼ 3, p ¼ 0.18, data not shown). In addition to NTPDase and NT5E, several enzymes such as PAP and TNAP might be involved in the generation of adenosine (Zylka et al., 2008; Sowa et al., 2010; Street et al., 2013). However, because PAP cannot hydrolyze ATP to generate adenosine at a neutral pH (Sowa et al., 2010), we examined the effect of a TNAP inhibitor (TNAP-I, Dahl et al., 2009) on ATP-induced inhibition of glutamate release. The application of TNAP-I (50 mM) did not affect the basal amplitude of glutamatergic EPSCs (100 ± 4% of the control, n ¼ 7, p ¼ 0.70, Fig. 6A and Ba). However, the extent of ATP-induced decrease in EPSC1 amplitude (57 ± 5% of the control, n ¼ 7, p < 0.01) was significantly reduced in the continued presence of 50 mM TNAP-I (84 ± 2% of the TNAP-I condition, n ¼ 7, p < 0.05, Fig. 6A and Ba). We also examined the effect of AMP on glutamatergic EPSCs, as TNAP efficiently hydrolyzes AMP to generate adenosine (Street et al., 2013). The application of TNAP-I (50 mM) also significantly reduced the AMP-induced decrease in EPSC1 amplitude (58 ± 4% of the control for AMP, n ¼ 7, p < 0.01), and this decrease was significantly reduced in the presence of 50 mM TNAP-I (84 ± 2% of the TNAP-I condition, n ¼ 7, p < 0.05, Fig. 6A and Bb). On the other hand, TNAP-I did not affect the adenosine-induced inhibition of glutamate release, as adenosine (100 mM) still inhibited EPSC1 amplitude even in the presence of 50 mM TNAP-I (51 ± 3% of the TNAP-I condition, n ¼ 7, p < 0.01, Fig. 6C and D). We also examined the effect of adenosine deaminase on the ATP-induced decrease in EPSC1 amplitude. The application of adenosine

Fig. 5. Effects of NTPDase inhibitors on the ATP-induced decrease in EPSCs. A, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP and 100 mM ATPgS. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. B, A typical time course of EPSC1 amplitude before, during and after application 100 mM ADP and 100 mM ADPbS. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. C, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP in the absence and presence of 100 mM ARL67156. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. b, ATP-induced changes in EPSC1 amplitude in the absence and presence of ARL67156. Each column was normalized to the control and represents the mean and SEM from 5 experiments. **; p < 0.01, n.s; not significant. Da, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP in the absence and presence of 30 mM POM-1. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. b, ATPinduced changes in EPSC1 amplitude in the absence and presence of POM-1. Each column was normalized to the control and represents the mean and SEM from 6 experiments. **; p < 0.01, n.s; not significant.

100

I.-S. Choi et al. / Neuropharmacology 93 (2015) 94e102

Fig. 6. Effect of TNAP-I on the ATP-induced decrease in EPSCs. A, A typical time course of EPSC1 amplitude before, during and after application 100 mM ATP or 100 mM AMP in the absence and presence of 50 mM TNAP-I. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. B, ATP- (a) or AMP (b)-induced changes in EPSC1 amplitude in the absence and presence of 50 mM TNAP-I. Each column was normalized to the control and represents the mean and SEM from 7 experiments. *; p < 0.05, **; p < 0.01, n.s; not significant. C, A typical time course of EPSC1 amplitude before, during and after application 100 mM adenosine in the absence and presence of 50 mM TNAP-I. The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. D, Adenosine-induced changes in EPSC1 amplitude in the absence and presence of 50 mM TNAP-I. Each column was normalized to the control and represents the mean and SEM from 7 experiments. **; p < 0.01, n.s; not significant. E, A typical time course of EPSC1 amplitude before, during and after application of 100 mM ATP in the absence and presence of adenosine deaminase (ADA, 2 unit/ml). The amplitudes of 6 EPSCs were averaged and plotted. Insets represent typical traces of the numbered region. F, ATP-induced changes in EPSC1 amplitude in the absence and presence of adenosine deaminase. Each column was normalized to the control and represents the mean and SEM from 4 experiments. **; p < 0.01, n.s; not significant.

deaminase (2 units/ml) did not affect the basal EPSC1 amplitude (100 ± 4% of the control, n ¼ 4, p ¼ 0.58, Fig. 6E and F). However, the extent of ATP-induced inhibition of EPSCs (61 ± 2% of the control, n ¼ 4, p < 0.01) was significantly reduced in the presence of adenosine deaminase (89 ± 3% of the adenosine deaminase condition, n ¼ 4, p ¼ 0.25, Fig. 6E and F). Further, since nucleoside transporters including equilibrative nucleoside transporters 1 (ENT1) are known to control glutamatergic transmission in central neurons (Cunha et al., 1998; Ackley et al., 2003), we examined the effect of NBMPR, a selective ENT1 inhibitor, on the ATP-induced inhibition of EPSCs. The extent of ATP-induced inhibition of EPSCs (59 ± 8% of the control, n ¼ 6, p < 0.01) was not affected by 1 mM NBMPR (57 ± 8% of the NBMPR condition, n ¼ 6, p < 0.01, Supplementary Fig. S3). 4. Discussion In the present study, we investigated the effect of ATP on glutamate release from nociceptive Ad- and C-fibers onto medullary dorsal horn neurons. We found that exogenously applied ATP decreased glutamatergic EPSC amplitude and increased the PPR during paired-pulse stimulation, and that ATP decreased glutamatergic mEPSC frequency without affecting the current amplitude, although further studies are needed to elucidate whether the origin of mEPSCs is primary afferents or excitatory interneurons. In contrast, previous studies performed with spinal dorsal horn neurons have shown that ATP acts on presynaptic P2X1 and/or P2X3 receptors to increase glutamate release onto spinal dorsal horn neurons (Gu and MacDermott, 1997; Nakatsuka and Gu, 2001). Therefore, it was rather unexpected that, in our study, ATP reduced rather than enhanced glutamate release from primary afferents onto medullary dorsal horn neurons, and that ab-me-ATP had no facilitatory effect on glutamatergic EPSCs. This is because multiple

types of P2X receptors, especially P2X3 receptors, are expressed on nociceptive TG neurons and their afferent fibers (Collo et al., 1996; Xiang et al., 1998; Kim et al., 2008). Lack of facilitatory effect of ATP on glutamate release might be due to the fast desensitization of P2X3 receptors, such that slow bath application of ATP might desensitize P2X3 receptors prior to their full activation (Piper and Docherty, 2000). Similarly, Davies and North (2009) showed that fast puff application of ab-me-ATP, via the activation of putative P2X2/3 receptors, facilitates spontaneous glutamate release onto medullary dorsal horn neurons of mice. However, we found that the fast application of ATP using the “Y-tube system” also showed a similar inhibition of glutamatergic EPSCs (Supplementary Fig. S4). Similarly, another study showed that bath application of ab-meATP enhances spontaneous glutamate release onto deep (lamina V) but not superficial (lamina II) dorsal horn neurons (Jennings et al., 2006). Alternatively, the activation of presynaptic P2X receptors might decrease glutamate release by depolarizing presynaptic terminals, as a presynaptic depolarization can inhibit action potentialdependent neurotransmitter release by inactivating Naþ channels (Jang et al., 2001; Dorostkar and Boehm, 2008). However, this should be not the case because the ATP-induced decrease in EPSC amplitude was not blocked by PPADS or suramin, nonselective P2X receptor antagonists. In addition, because Bz-ATP had no effect on glutamatergic EPSCs, the involvement of P2X7 receptors in the ATPinduced decrease in EPSC amplitude can be excluded. Taken together, these results suggest that presynaptic P2X receptors are not responsible for the ATP-induced inhibition of glutamate release from primary afferents onto medullary dorsal horn neurons in the present experimental conditions. ATP is also a general agonist for P2Y receptors, and therefore the inhibitory action of ATP on action potential-dependent glutamate release could be mediated by P2Y receptors. In fact, mRNAs for

I.-S. Choi et al. / Neuropharmacology 93 (2015) 94e102

most of the P2Y receptor subtypes are expressed in TG neurons (Ceruti et al., 2008). Of these, P2Y1 and P2Y4 receptors are expressed on a subpopulation of CGRP-positive and/or NF200positive TG neurons, suggesting that P2Y1 and P2Y4 receptors are likely to be expressed on nociceptive C- and/or Ad-fibers (Ruan and Burnstock, 2003). Although the functional roles of P2Y receptors in nociceptive transmission are still unknown in the trigeminal system, P2Y1 receptors are involved in ADP-induced analgesia in DRG neurons (Gerevich et al., 2004). In the present study, we found that selective antagonists for P2Y1, P2Y12, and P2Y13 receptors did not block the ADP-induced decrease in EPSC amplitude. In addition, nonselective P2Y receptor antagonists, such as suramin and PPADS, did not block the ATP-induced decrease in EPSC amplitude. Thus, presynaptic P2Y receptors, at least P2Y1, P2Y12, and P2Y13 receptors, are not involved in the ATP-induced decrease in glutamate release from primary afferents onto medullary dorsal horn neurons. Given that both ATP and ADP decrease glutamate release from primary afferents via mechanisms independent on the activation of P2X or P2Y receptors, another possibility to explain the present results would be that ATP action is mediated by adenosine A1 receptors after enzymatic hydrolysis. In fact, we found that functional A1 receptors act presynaptically to inhibit glutamate release at the same synapses (Supplementary Fig. S3, see also Choi et al., 2011), suggesting that ATP might act on presynaptic A1 receptors to inhibit glutamate release after enzymatic hydrolysis to adenosine. This possibility is further supported by the present results showing that the A1 receptor antagonist DPCPX clearly blocked the ATP- and ADP-induced decrease in EPSC amplitude. To date, several enzymes, including NTPDases, NT5E, PAP, and TNAP, are known to be involved in the enzymatic hydrolysis of ATP. However, since NT5E converts AMP to adenosine, ATP and/or ADP must first be hydrolyzed to AMP by NTPDases. In contrast, TNAP can directly hydrolyze ATP and/or ADP to adenosine (Abbracchio et al., 2009; Street et al., 2013). In the present study, we found that NTPDase blockers, such as ARL67156 and POM-1, had no effect on the ATP-induced decrease in glutamatergic EPSCs, and that the ATP-induced decrease in glutamatergic EPSCs was not affected in the presence of both ARL67156 and ab-me-ADP. The results suggest that both NTPDases and NT5E might be not responsible for the enzymatic hydrolysis of ATP to adenosine. In contrast, we found that the ATP-induced decrease in glutamatergic EPSCs was significantly blocked by a TNAP inhibitor, suggesting that TNAP might be a major enzyme involved in the hydrolysis of ATP or ADP to adenosine. Although the ATP-induced inhibition of glutamatergic EPSCs was not completely blocked by TNAP-I, this might be due to the incomplete inhibition of TNAP by 50 mM TNAP-I, as this drug is known to inhibit TNAP in a non-competitive manner (Dahl et al., 2009). Alternatively, other enzymes hydrolyzing ATP such as PAP might be involved in the enzymatic hydrolysis of ATP to adenosine. Although a recent study has shown that PAP and NT5E in addition to TNAP act redundantly to generate adenosine in the dorsal spinal cord (Street et al., 2013), the involvement of PAP in the ATP-induced decrease in glutamatergic EPSCs might be negligible because PAP cannot hydrolyze ATP to generate adenosine at a neutral pH (Sowa et al., 2010). On the other hand, although we found that Bz-ATP had no inhibitory effect on glutamatergic transmission in medullary dorsal horn neurons, a previous study showed that Bz-ATP could be converted into adenosine to activate A1 receptors and decrease glutamate release in hippocampal neurons (Kukley et al., 2004). This discrepancy might be due to differential catabolism of Bz-ATP between the two regions, as the conversion of Bz-ATP into adenosine seems to require nucleoside transporters as well as intracellular esterase in hippocampal neurons (Kukley et al., 2004). It is widely accepted that ATP has pronociceptive effects, as it can directly activate multiple types of ionotropic P2X receptors

101

including P2X3 and P2X2/3 at peripheral terminals of sensory neurons (Collo et al., 1996; Xiang et al., 1998; Chizh and Illes, 2001). Further, ATP acts on presynaptic P2X receptors to enhance glutamatergic transmission in the spinal and medullary dorsal horn area (Bardoni et al., 1997; Nakatsuka and Gu, 2001; Jennings et al., 2006; Davies and North, 2009). ATP and/or ADP also exert pronociceptive or antinociceptive effects by activating metabotropic P2Y receptors (for review, Gerevich and Illes, 2004). Furthermore, the purine nucleoside adenosine, via the activation of A1 receptors, inhibits glutamatergic transmission in the spinal and medullary dorsal horn area (Lao et al., 2004; Choi et al., 2011). These observations suggest that adenine nucleotides and adenosine are closely involved in the modulation of nociceptive transmission from peripheral tissues. In this study, we have shown that exogenously applied ATP or ADP inhibited primary afferent glutamatergic transmission onto medullary dorsal horn neurons. This inhibition depended on adenosine A1 receptors, rather than P2X and P2Y receptors, and required the enzymatic conversion of ATP and/or ADP to adenosine. The resultant changes in primary afferent transmission may play an important role in the regulation of nociceptive transmission from orofacial tissues. At this stage, it is still unknown whether endogenous ATP, after enzymatic conversion, can inhibit primary afferent glutamatergic transmission onto medullary dorsal horn neurons. Previous studies have shown that ATP can be endogenously released from primary afferent terminals as well as interneurons in the spinal dorsal horn region (Bardoni et al., 1997; Jo and Schlichter, 1999). In contrast, we found that the blockade of enzymatic conversion of ATP to adenosine as well as inhibition of adenosine A1 receptors had no effect on basal glutamate release, consistent with a minor physiological role of endogenous ATP and/or adenosine in the modulation of glutamatergic transmission within the medullary dorsal horn. Further studies are needed to determine the physiological and pharmacological relevance of endogenous ATP for nociceptive transmission from orofacial tissues.

Acknowledgment This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (2008-0062282). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.neuropharm.2015.01.020. References Abbracchio, M.P., Burnstock, G., Boeynaems, J.M., Barnard, E.A., Boyer, J.L., Kennedy, C., Knight, G.E., Fumagalli, M., Gachet, C., Jacobson, K.A., Weisman, G.A., 2006. International Union of Pharmacology LVIII: update on the P2Y G protein-coupled nucleotide receptors: from molecular mechanisms and pathophysiology to therapy. Pharmacol. Rev. 58, 281e341. Abbracchio, M.P., Burnstock, G., Verkhratsky, A., Zimmermann, H., 2009. Purinergic signalling in the nervous system: an overview. Trends. Neurosci. 32, 19e29. Ackley, M.A., Governo, R.J., Cass, C.E., Young, J.D., Baldwin, S.A., King, A.E., 2003. Control of glutamatergic neurotransmission in the rat spinal dorsal horn by the nucleoside transporter ENT1. J. Physiol. (Lond.) 548, 507e517. Bardoni, R., Goldstein, P.A., Lee, C.J., Gu, J.G., MacDermott, A.B., 1997. ATP P2X receptors mediate fast synaptic transmission in the dorsal horn of the rat spinal cord. J. Neurosci. 17, 5297e5304. Boyer, J.L., Mohanram, A., Camaioni, E., Jacobson, K.A., Harden, T.K., 1998. Competitive and selective antagonism of P2Y1 receptors by N6-methyl 2'-deoxyadenosine 3',5'-bisphosphate. Br. J. Pharmacol. 124, 1e3. Burnstock, G., 2007. Physiology and pathophysiology of purinergic neurotransmission. Physiol. Rev. 87, 659e797. Burnstock, G., 2009a. Purines and sensory nerves. Handb. Exp. Pharmacol. 194, 333e392.

102

I.-S. Choi et al. / Neuropharmacology 93 (2015) 94e102

Burnstock, G., 2009b. Purinergic receptors and pain. Curr. Pharm. Des. 15, 1717e1735. Ceruti, S., Fumagalli, M., Villa, G., Verderio, C., Abbracchio, M.P., 2008. Purinoceptormediated calcium signaling in primary neuron-glia trigeminal cultures. Cell. Calcium 43, 576e590. Chen, C.C., Akopian, A.N., Sivilotti, L., Colquhoun, D., Burnstock, G., Wood, J.N., 1995. A P2X purinoceptor expressed by a subset of sensory neurons. Nature 377, 428e431. Chizh, B.A., Illes, P., 2001. P2X receptors and nociception. Pharmacol. Rev. 53, 553e568. Choi, I.S., Cho, J.H., Jang, I.S., 2011. A1 receptors inhibit glutamate release in rat medullary dorsal horn neurons. Neuroreport 22, 711e715. Choi, I.S., Cho, J.H., An, C.H., Jung, J.K., Hur, Y.K., Choi, J.K., Jang, I.S., 2012. 5-HT1B receptors inhibit glutamate release from primary afferent terminals in rat medullary dorsal horn neurons. Br. J. Pharmacol. 167, 356e367. Collo, G., North, R.A., Kawashima, E., Merlo-Pich, E., Neidhart, S., Surprenant, A., Buell, G., 1996. Cloning of P2X5 and P2X6 receptors and the distribution and properties of an extended family of ATP-gated ion channels. J. Neurosci. 16, 2495e2507. ~o, A.M., Ribeiro, J.A., 1998. Inhibition by ATP of hippocampal Cunha, R.A., Sebastia synaptic transmission requires localized extracellular catabolism by ectonucleotidases into adenosine and channeling to adenosine A1 receptors. J. Neurosci. 18, 1987e1995. Dahl, R., Sergienko, E.A., Su, Y., Mostofi, Y.S., Yang, L., Simao, A.M., Narisawa, S., Brown, B., Mangravita-Novo, A., Vicchiarelli, M., Smith, L.H., O'Neill, W.C., n, J.L., Cosford, N.D., 2009. Discovery and validation of a series of aryl Milla sulfonamides as selective inhibitors of tissue-nonspecific alkaline phosphatase (TNAP). J. Med. Chem. 52, 6919e6925. Davies, A.J., North, R.A., 2009. Electrophysiological and morphological properties of neurons in the substantia gelatinosa of the mouse trigeminal subnucleus caudalis. Pain 146, 214e221. Donnelly-Roberts, D., McGaraughty, S., Shieh, C.C., Honore, P., Jarvis, M.F., 2008. Painful purinergic receptors. J. Pharmacol. Exp. Ther. 324, 409e415. Dorostkar, M.M., Boehm, S., 2008. Presynaptic ionotropic receptors. Handb. Exp. Pharmacol. 184, 479e527. Dunwiddie, T.V., Diao, L., Proctor, W.R., 1997. Adenine nucleotides undergo rapid, quantitative conversion to adenosine in the extracellular space in rat hippocampus. J. Neurosci. 17, 7673e7682. Fredholm, B.B., Chen, J.F., Cunha, R.A., Svenningsson, P., Vaugeois, J.M., 2005. Adenosine and brain function. Int. Rev. Neurobiol. 191e270. €der, W., Franke, H., Wirkner, K., No €renberg, W., Gerevich, Z., Borvendeg, S.J., Schro Fürst, S., Gillen, C., Illes, P., 2004. Inhibition of N-type voltage-activated calcium channels in rat dorsal root ganglion neurons by P2Y receptors is a possible mechanism of ADP-induced analgesia. J. Neurosci. 24, 797e807. Gerevich, Z., Illes, P., 2004. P2Y receptors and pain transmission. Purinergic. Signal 1, 3e10. Gonçalves, J., Queiroz, G., 2008. Presynaptic adenosine and P2Y receptors. Handb. Exp. Pharmacol. 184, 339e372. Gu, J.G., MacDermott, A.B., 1997. Activation of ATP P2X receptors elicits glutamate release from sensory neuron synapses. Nature 389, 749e753. Jang, I.S., Jeong, H.J., Katsurabayashi, S., Akaike, N., 2002. Functional roles of presynaptic GABAA receptors on glycinergic nerve terminals in the rat spinal cord. J. Physiol. (Lond) 541, 423e434. Jang, I.S., Rhee, J.S., Kubota, H., Akaike, N., Akaike, N., 2001. Developmental changes in P2X purinoceptors on glycinergic presynaptic nerve terminals projecting to rat substantia gelatinosa neurones. J. Physiol. (Lond) 536, 505e519. Jarvis, M.F., 2010. The neural-glial purinergic receptor ensemble in chronic pain states. Trends. Neurosci. 33, 48e57. Jennings, E.A., Christie, M.J., Sessle, B.J., 2006. ATP potentiates neurotransmission in the rat trigeminal subnucleus caudalis. Neuroreport 17, 1507e1510. Jo, Y.H., Schlichter, R., 1999. Synaptic corelease of ATP and GABA in cultured spinal neurons. Nat. Neurosci. 2, 241e245. Kim, Y.C., Lee, J.S., Sak, K., Marteau, F., Mamedova, L., Boeynaems, J.M., Jacobson, K.A., 2005. Synthesis of pyridoxal phosphate derivatives with antagonist activity at the P2Y13 receptor. Biochem.. Pharmacol. 70, 266e274.

Kim, Y.S., Paik, S.K., Cho, Y.S., Shin, H.S., Bae, J.Y., Moritani, M., Yoshida, A., Ahn, D.K., Valtschanoff, J., Hwang, S.J., Moon, C., Bae, Y.C., 2008. Expression of P2X3 receptor in the trigeminal sensory nuclei of the rat. J. Comp. Neurol. 506, 627e639. Kukley, M., Stausberg, P., Adelmann, G., Chessell, I.P., Dietrich, D., 2004. Ecto-nucleotidases and nucleoside transporters mediate activation of adenosine receptors on hippocampal mossy fibers by P2X7 receptor agonist 2'-3'-O-(4benzoylbenzoyl)-ATP. J. Neurosci. 24, 7128e7139. Lao, L.J., Kawasaki, Y., Yang, K., Fujita, T., Kumamoto, E., 2004. Modulation by adenosine of Ad and C primary-afferent glutamatergic transmission in adult rat substantia gelatinosa neurons. Neuroscience 125, 221e231. Masino, S.A., Diao, L., Illes, P., Zahniser, N.R., Larson, G.A., Johansson, B., Fredholm, B.B., Dunwiddie, T.V., 2002. Modulation of hippocampal glutamatergic transmission by ATP is dependent on adenosine A1 receptors. J. Pharmacol. Exp. Ther. 303, 356e363. €llich, A., Stephan, H., 2006. PolMüller, C.E., Iqbal, J., Baqi, Y., Zimmermann, H., Ro yoxometalatesea new class of potent ecto-nucleoside triphosphate diphosphohydrolase (NTPDase) inhibitors. Bioorg. Med. Chem. Lett. 16, 5943e5947. Murase, K., Ryu, P.D., Randic, M., 1989. Excitatory and inhibitory amino acids and peptide-induced responses in acutely isolated rat spinal dorsal horn neurons. Neurosci. Lett. 103, 56e63. Nakatsuka, T., Ataka, T., Kumamoto, E., Tamaki, T., Yoshimura, M., 2000. Alteration in synaptic inputs through C-afferent fibers to substantia gelatinosa neurons of the rat spinal dorsal horn during postnatal development. Neuroscience 99, 549e556. Nakatsuka, T., Gu, J.G., 2001. ATP P2X receptor-mediated enhancement of glutamate release and evoked EPSCs in dorsal horn neurons of the rat spinal cord. J. Neurosci. 21, 6522e6531. Piper, A.S., Docherty, R.J., 2000. One-way cross-desensitization between P2X purinoceptors and vanilloid receptors in adult rat dorsal root ganglion neurones. J. Physiol. (Lond) 523, 685e696. Ralevic, V., Burnstock, G., 1998. Receptors for purines and pyrimidines. Pharmacol. Rev. 50, 413e492. Rodrigues, R.J., Almeida, T., Richardson, P.J., Oliveira, C.R., Cunha, R.A., 2005. Dual presynaptic control by ATP of glutamate release via facilitatory P2X1, P2X2/3, and P2X3 and inhibitory P2Y1, P2Y2, and/or P2Y4 receptors in the rat hippocampus. J. Neurosci. 25, 6286e6295. Ruan, H.Z., Burnstock, G., 2003. Localisation of P2Y1 and P2Y4 receptors in dorsal root, nodose and trigeminal ganglia of the rat. Histochem. Cell. Biol. 120, 415e426. Sowa, N.A., Taylor-Blake, B., Zylka, M.J., 2010. Ecto-5'-nucleotidase (CD73) inhibits nociception by hydrolyzing AMP to adenosine in nociceptive circuits. J. Neurosci. 30, 2235e2244. Street, S.E., Kramer, N.J., Walsh, P.L., Taylor-Blake, B., Yadav, M.C., King, I.F., Vihko, P., n, J.L., Zylka, M.J., 2013. Tissue-nonspecific alkaline Wightman, R.M., Milla phosphatase acts redundantly with PAP and NT5E to generate adenosine in the dorsal spinal cord. J. Neurosci. 33, 11314e11322. Wall, M.J., Wigmore, G., Lopat ar, J., Frenguelli, B.G., Dale, N., 2008. The novel NTPDase inhibitor sodium polyoxotungstate (POM-1) inhibits ATP breakdown but also blocks central synaptic transmission, an action independent of NTPDase inhibition. Neuropharmacology 55, 1251e1258. Xiang, Z., Bo, X., Burnstock, G., 1998. Localization of ATP-gated P2X receptor immunoreactivity in rat sensory and sympathetic ganglia. Neurosci. Lett. 256, 105e108. Xu, B., Stephens, A., Kirschenheuter, G., Greslin, A.F., Cheng, X., Sennelo, J., Cattaneo, M., Zighetti, M.L., Chen, A., Kim, S.A., Kim, H.S., Bischofberger, N., Cook, G., Jacobson, K.A., 2002. Acyclic analogues of adenosine bisphosphates as P2Y receptor antagonists: phosphate substitution leads to multiple pathways of inhibition of platelet aggregation. J. Med. Chem. 45, 5694e5709. Zimmermann, H., 2000. Extracellular metabolism of ATP and other nucleotides. Naunyn. Schmiedeb. Arch. Pharmacol. 362, 299e309. Zylka, M.J., Sowa, N.A., Taylor-Blake, B., Twomey, M.A., Herrala, A., Voikar, V., Vihko, P., 2008. Prostatic acid phosphatase is an ectonucleotidase and suppresses pain by generating adenosine. Neuron 60, 111e122.