Neuroscience 164 (2009) 1833–1844

EFFECT OF RESINIFERATOXIN ON GLUTAMATERGIC SPONTANEOUS EXCITATORY SYNAPTIC TRANSMISSION IN SUBSTANTIA GELATINOSA NEURONS OF THE ADULT RAT SPINAL CORD

C.-Y. JIANG, T. FUJITA, H.-Y. YUE, L.-H. PIAO, T. LIU, al., 2000; Szallasi and Di Marzo, 2000; McGaraughty et T. NAKATSUKA AND E. KUMAMOTO* al., 2003; Cristino et al., 2006; Starowicz et al., 2007a) Department of Physiology, Saga Medical School, 5-1-1 Nabeshima, and plays a role in various physiological functions including Saga 849–8501, Japan motor control, cognition and antinociception (Marinelli et al., 2003; McGaraughty et al., 2003; Starowicz et al., 2008). At Abstract—The transient receptor potential (TRP) vanilloid central synapses, TRPV1 activation by capsaicin, the pun- type 1 (TRPV1) agonist, capsaicin, enhances glutamatergic gent ingredient of hot chili peppers, increases the spontane- spontaneous excitatory synaptic transmission in CNS neu- ous release of L-glutamate from nerve terminals (Sasamura rons. Resiniferatoxin (RTX) has a much higher affinity for et al., 1998; Marinelli et al., 2003; McGaraughty et al., 2003; TRPV1 than capsaicin, but its ability to modulate excitatory Xing and Li, 2007; Starowicz et al., 2007a). transmission is unclear. We examined the effect of RTX on TRPV1 activation in the central terminals of primary- excitatory transmission using the whole-cell patch-clamp technique in substantia gelatinosa (SG) neurons of adult rat afferent neurons increases the spontaneous release of spinal cord slices. Bath-applied RTX dose-dependently in- L-glutamate to substantia gelatinosa (SG) neurons of the creased the frequency, but not the amplitude, of spontaneous spinal dorsal horn (Yang et al., 1998; Guo et al., 1999; excitatory postsynaptic current (sEPSC), independent of its Valtschanoff et al., 2001; Morisset and Urban, 2001; application time. In about a half of the neurons tested, this Hwang et al., 2004), which play a pivotal role in regulating ؊ effect was accompanied by an inward current at 70 mV that nociceptive transmission (Willis and Coggeshall, 1991). was sensitive to glutamate-receptor antagonists. Repeated application of RTX did not affect excitatory transmission. TRPV1 activation also potentiates L-glutamate release in RTX was more potent than capsaicin but showed similar rat superficial medullary dorsal horn neurons (Jennings et efficacy. RTX activity could be blocked by capsazepine or al., 2003). Endogenous agonists for TRPV1 include endo- SB-366791, a TRPV1 antagonist, but not tetrodotoxin, a Na؉- cannabinoids and lipoxygenase metabolites, which have channel blocker, and could be inhibited by pretreatment with similar structures to capsaicin, which is not produced en- capsaicin but not the TRPA1 agonist, allyl isothiocyanate. dogenously (Zygmunt et al., 1999; Hwang et al., 2000; for RTX enhances the spontaneous release of L-glutamate from review see Caterina and Julius, 2001; Starowicz et al., nerve terminals with similar efficacy as capsaicin and pro- duces a membrane depolarization by activating TRPV1 in the 2007b). Tonic TRPV1 activation by endogenous agonists SG, with fast desensitization and slow recovery from desen- in the PAG may activate descending antinociceptive path- sitization. These results indicate a mechanism by which RTX ways in rats (Starowicz et al., 2007a). can modulate excitatory transmission in SG neurons to reg- Resiniferatoxin (RTX) is an ultrapotent TRPV1 agonist, ulate nociceptive transmission. © 2009 IBRO. Published by a capsaicin analog isolated from the dried latex of the Elsevier Ltd. All rights reserved. cactus-like plant, Euphorbia resinifera (Hergenhahn et al., Key words: spinal dorsal horn, TRPV1, whole-cell patch- 1975; Schmidt and Evans, 1979). RTX binds to the same clamp, pain. recognition site as capsaicin in the cytosolic tails of TRPV1 (Jung et al., 2002). RTX has higher binding affinity and is three to four orders of magnitude more potent than capsa- Transient receptor potential vanilloid type 1 (TRPV1) is a icin at producing current responses in oocytes expressing molecular integrator of painful stimuli such as vanilloids, TRPV1, in inhibiting twitch contraction of the vas deferens noxious heat and protons on the peripheral terminals of and in thermoregulation and neurogenic inflammation primary-afferent neurons (Caterina et al., 1997; Caterina (Szallasi and Blumberg, 1989; Maggi et al., 1990; Szallasi and Julius, 2001). TRPV1 is also expressed in several brain et al., 1993, 1995; Szallasi, 1994). [3H]RTX is widely used nuclei such as the hypothalamus, substantia nigra and peri- to examine the distribution of TRPV1 in the nervous sys- aqueductal gray (PAG) (Sasamura et al., 1998; Mezey et tem (Szallasi and Blumberg, 1990; Szallasi, 1994; Acs et *Corresponding author. Tel: ϩ81-952-34-2273; fax: ϩ81-952-34-2013. al., 1996). RTX may also be useful in treating disorders E-mail address: [email protected] (E. Kumamoto). such as neuropathic pain and lower urinary tract dysfunc- Abbreviations: AITC, allyl isothiocyanate; APV, DL-2-amino-5-phos- phonovaleric acid; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; tion that involve excessive TRPV1 activity (Szallasi, 2002). In dorsal root ganglion (DRG) neurons or heterologous DRG, dorsal root ganglion; EC50, effective concentration producing half-maximal response; NMDA, N-methyl-D-aspartate; RTX, resinifera- cells expressing TRPV1, RTX produces a whole-cell cur- toxin; sEPSC, spontaneous excitatory postsynaptic current; SG, sub- rent response and single-channel current activity that per- stantia gelatinosa; TRPA1, transient receptor potential A1; TRPV1, sist after its washout (Raisinghani et al., 2005). It has not transient receptor potential vanilloid type 1; TTX, tetrodotoxin; VH, holding potential. been fully examined yet how RTX affects excitatory trans- 0306-4522/09 $ - see front matter © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2009.09.033

1833 1834 C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844 mission. Here, we examined the effects of RTX on gluta- Statistical analysis matergic spontaneous excitatory synaptic transmission in Numerical data are given as the meanϮstandard error of the SG neurons of adult rat spinal cord slices using the whole- mean (SEM). Statistical significance is determined as PϽ0.05 cell patch-clamp technique. using a Student’s t-test unless otherwise mentioned. In all cases, n refers to the number of neurons studied. EXPERIMENTAL PROCEDURES RESULTS All animal experiments were approved by the Animal Care and Use Committee of Saga University. All efforts were made to Whole-cell patch-clamp recordings were made from a total minimize animal suffering and the number of animal used. of 128 SG neurons. Stable recordings could be obtained from slices maintained in vitro for more than 10 h, and Spinal cord slice preparation recordings were made from single SG neurons for up to Spinal cord slices from adult rats were prepared as described 3 h. All SG neurons tested had resting membrane poten- previously (Yue et al., 2005; Fujita and Kumamoto, 2006; Liu et tials lower than Ϫ60 mV (when measured in current-clamp al., 2008). Male Sprague–Dawley rats (6–8 weeks old) were Ϫ mode), and exhibited sEPSCs at a VH of 70 mV, near the anesthetized with urethane (1.2 g/kg i.p.), and a lumbosacral reversal potential for inhibitory postsynaptic currents (Iy- segment (L1–S3) of the spinal cord was extracted and placed in preoxygenated cold Krebs solution (2–4 °C) preequilibrated with adomi et al., 2000; Liu et al., 2008). The sEPSCs were 95% O and 5% CO . The composition of Krebs solution used was completely blocked by a non-N-methyl-D-aspartate (NMDA) 2 2 ␮ (in mM): NaCl, 117; KCl, 3.6; CaCl2, 2.5; MgCl2, 1.2; NaH2PO4, receptor antagonist, CNQX (10 M), and were not affected ϭ ϩ 1.2; NaHCO3, 25; and glucose, 11 (pH 7.4 when saturated with by a voltage-gated Na -channel blocker, TTX (0.5 ␮M) (Iy- the gas). After cutting all ventral and dorsal roots, the pia-arach- adomi et al., 2000; Yue et al., 2005). Thus, sEPSCs were not noid membrane was removed. The spinal cord was mounted on a contaminated by spontaneous inhibitory postsynaptic cur- microslicer (DTK-1000, Dousaka, Kyoto, Japan) and several 650 rents and occurred without spike propagation from the soma ␮m-thick transverse slices were cut. One of the slices was trans- ferred to a recording chamber (0.5 ml in volume), then completely of presynaptic TTX-sensitive neurons. submerged and superfused at 12–15 ml/min with Krebs solution saturated with 95% O2 and 5% CO2 and maintained at Effects of RTX on glutamatergic spontaneous 36.0Ϯ0.5 °C. The remaining slices (at most five) were stored excitatory transmission in SG neurons under similar conditions until use. RTX (0.5 ␮M) superfused for 1 min enhanced spontane- Patch-clamp recordings from SG neurons ous excitatory transmission in a SG neuron (Fig. 1A). The sEPSC frequency increased gradually over time, peaking The SG can be identified under a stereomicroscope as a translu- around 4 min after RTX addition; this facilitation was ac- cent band across the dorsal horn. Whole-cell voltage-clamp re- companied by a small increase in sEPSC amplitude (Fig. cordings from SG neurons were made at a holding potential (VH) of Ϫ70 mV using a patch-pipette. The recorded neurons were 1B). This increase in sEPSC frequency did not subside for located at the center of SG to avoid recordings from laminae I and at least 10 min after RTX washout. RTX significantly in- III neurons, as reported previously (Yue et al., 2005; Fujita and creased the proportion of sEPSCs with a shorter inter- Kumamoto, 2006). The patch-pipette solution used was com- event interval and a larger amplitude (Fig. 1C); this effect posed of (in mM): K-gluconate, 135; CaCl , 0.5; MgCl , 2; KCl, 5; 2 2 was confirmed in three other neurons and accompanied by ethyleneglycol-bis(aminoethylether) tetraacetate, 5; N-2-hydroxyeth- ylpiperazine-N=-2-ethanesulfonate, 5; and Mg–ATP, 5 (pHϭ7.2). a small inward current. Signals were acquired using an Axopatch 200 B amplifier (Molec- Capsaicin-induced enhancement of spontaneous exci- ular Devices, Sunnyvale, CA, USA). Data were low-pass filtered at tatory transmission in SG neurons shows slow recovery 5 kHz, and digitized at 333 kHz with an A/D converter. The data from desensitization (Yang et al., 1998). Similarly, a sec- were stored and analyzed with a personal computer using ond RTX application 1 or 2 h later did not affect excitatory pCLAMP 8.1 software (Molecular Devices) and Mini Analysis transmission (Fig. 2). The peak amplitude of inward current Program (Synaptosoft, Decatur, GA, USA); detection criteria for spontaneous excitatory postsynaptic currents (sEPSCs) included produced by the second application was smaller than that a 4.5 pA event threshold, with a fast rise time and a decay curve of the first application (Fig. 2A, B). For example, the initial that approximated exponential decay. RTX treatment produced an inward current with a peak amplitude of 6.4 or 15.5 pA, but a second application 1 h Drug application later did not change the holding currents. Overall peak current amplitudes were 14.1Ϯ3.6 pA initially and 2.3Ϯ1.2 Drugs were applied by superfusing a solution containing drugs ϭ without altering the perfusion rate or temperature. The drugs used pA (n 3) 2 h later, although some neurons did not produce were RTX, SB-366791, DL-2-amino-5-phosphonovaleric acid any inward current after RTX. We therefore only added (APV; Sigma Aldrich, St. Louis, MO, USA), allyl isothiocyanate RTX once to a spinal cord slice in subsequent experi- (AITC), capsaicin, capsazepine, tetrodotoxin (TTX; Wako, Osaka, ments. Japan) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; Tocris At around 4 min after RTX addition, when the sEPSC Cookson, Bristol, UK). These drugs (except for capsaicin, TTX frequency increase was maximal (see Fig. 1B), RTX in- and APV where distilled water was used as a solvent) were first Ϯ ϭ Ͻ dissolved in dimethyl sulfoxide at 1000x stocks and then stored at creased sEPSC frequency to 236 20% (n 23; P 0.05) Ϫ20 °C. These drugs were then diluted to the final concentration of control (11.9Ϯ1.6 Hz) in most (88%) of neurons exam- in Krebs solution immediately before use. ined (nϭ26). sEPSC frequency did not change in three C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844 1835

RTX enhances spontaneous excitatory transmission in a dose-dependent and application-time independent manner RTX (0.005–2 ␮M) showed less efficacy for enhancing excitatory transmission at higher concentrations (Fig. 3A). RTX increased sEPSC frequency maximally at 0.5 ␮M (Fig. 3B). The Hill equation with data from Ͻ0.5 ␮M indi- cated an effective concentration producing half-maximal ␮ response (EC50) of 0.21 M. RTX did not affect sEPSC amplitude except at 0.5 ␮M, but dose-dependently (0.005–2 ␮M) produced an inward current (Table 1). At 20 min interval in a range of 0.05–0.2 ␮M, the second appli- cation of RTX did not increase sEPSC frequency. sEPSC frequencies in the first and second application of RTX were 233Ϯ28% (nϭ3; PϽ0.05) and 104Ϯ2% (nϭ3; PϾ0.05), respectively, of control (just before RTX application) at 0.2 ␮M; 150Ϯ10% (nϭ3; PϽ0.05) and 105Ϯ2% (nϭ3; PϾ0.05), respectively, of control at 0.1 ␮M; 149Ϯ11% (nϭ3; PϽ0.05) and 103Ϯ3% (nϭ3; PϾ0.05), respectively, of control at 0.05 ␮M (data not shown), where these effects were measured between 3.5 and 4 min after RTX treat- ment. We next examined how the application time of RTX (0.5 ␮M) affected sEPSC frequency. The increase in sEPSC frequency appeared to desensitize after RTX ad- dition. Superfusing RTX for 0.5 min increased sEPSC frequency after its washout (Fig. 4A), and after 2 min treatment, sEPSC frequency increased in the presence of RTX (Fig. 4B). sEPSC frequency and amplitude increases produced by RTX did not depend on the application time between 0.5 and 2 min (Fig. 4C). RTX superfused for 0.5 min produced an inward current (peak amplitude: 6.3 and 14.8 pA; Fig. 4A) in two of the four neurons tested, and the ␮ two remaining neurons had no response. Superfusing RTX Fig. 1. Resiniferatoxin (RTX; 0.5 M) superfused for 1 min increases the Ϯ frequency of spontaneous excitatory postsynaptic currents (sEPSCs) for 2 min produced a peak inward current of 15.4 3.7 pA with a small increase in amplitude in substantia gelatinosa (SG) (nϭ4; Fig. 4B) in four of the five neurons examined and the neurons. (A) Recordings of sEPSCs in the absence and presence of remaining neuron had no response. Thus, the facilitatory RTX. In this and subsequent figures, the duration of drug superfusion effect of RTX on excitatory transmission may desensitize is shown by a horizontal bar above the chart recording, and 4 consec- utive traces of sEPSCs for a period indicated by a short bar below the with time and at higher concentrations. chart recording are shown in an expanded time scale. RTX produces a small peak inward current of 3.5 pA. In this and subsequent record- Effects of RTX on spontaneous excitatory ings, horizontal dotted lines indicate the holding current level in the transmission in the presence of various drugs control. (B) Changes in sEPSC frequency and amplitude (closed and open circles, respectively) were measured every 0.5 min (frequency The TRPV1 antagonist, capsazepine (10 ␮M), blocked the and amplitude in the control: 15.0 Hz and 12.1 pA, respectively). In this and subsequent graphs showing relative values, the control level ability of RTX to increase sEPSC frequency and amplitude (100%) is indicated by a horizontal dotted line. (C) Cumulative histo- (Fig. 5Aa) in all neurons examined, producing sEPSC grams of the amplitude and inter-event interval of sEPSC in the control frequency and amplitude of 105Ϯ10% (nϭ4; PϾ0.05) and (dotted line) and with RTX (continuous line). The histograms were 104Ϯ4% (nϭ4; PϾ0.05), respectively, of control (8.2Ϯ0.9 examined for 0.5 min in the control (435 sEPSC events) and 3 min Ϯ after RTX washout (947 sEPSC events). RTX shifted the inter-event Hz, 11.1 0.9 pA) between 3.5 and 4 min after RTX treat- interval and amplitude distributions to a shorter and larger ones, ment. Capsazepine itself did not affect sEPSC frequency respectively (PϽ0.01; Kolmogorov–Smirnov test). (A–C) were ob- and amplitude (97Ϯ12% and 105Ϯ4%, respectively, of tained from the same neuron. Holding potential (V )ϭϪ70 mV. H control; nϭ4; each PϾ0.05). Capsazepine also inhibited the inward current produced by RTX; in three of four neu- neurons [99Ϯ8% (nϭ3; PϾ0.05) of control (26.7Ϯ2.7 Hz)]. rons tested, its peak amplitude averaged to be 6.6Ϯ2.0 pA sEPSC amplitude increased slightly [114Ϯ3% (nϭ26; (nϭ3; Fig. 5A), a value significantly smaller than that PϽ0.05) of control (12.5Ϯ0.6 pA)] and RTX produced an (10.7Ϯ1.5 pA, nϭ15) in the absence of capsazepine overall peak inward current of 10.7Ϯ1.5 pA (nϭ15) in 58% (PϽ0.05). The remaining neuron did not change holding of 26 neurons tested (Table 1). currents. Another TRPV1 antagonist, SB-366791 (30 ␮M, 1836 C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844

Fig. 2. The facilitatory effect of RTX (0.5 ␮M) on excitatory transmission does not recover for at least 2 h after washout. (A, B) Recordings of sEPSCs showing the effects on excitatory transmission of an initial RTX application (left) and a second RTX application 1 h (A) or 2 h (B) later (right), with left and right recordings obtained from the same neuron. (C) sEPSC frequency (left) and amplitude (right) in the first (1st) or second (2nd) application of RTX, relative to control, measured between 3.5 and 4 min after RTX addition. Control (just before RTX application): the first application, 17.7Ϯ4.6 Hz and 16.1Ϯ1.5 pA; the second application 1 h later, 16.3Ϯ5.3 Hz and 17.7Ϯ2.3 pA (nϭ6; C). Control: the first application, 10.0Ϯ4.4 Hz and 9.9Ϯ0.8 pA; the second application 2 h later, 9.6Ϯ2.3 Hz and 9.2Ϯ1.1 pA (nϭ3; C). Data obtained from the same neuron are connected by a straight line. ϭϪ VH 70 mV.

a concentration enough to depress TRPV1 activation in the (Fig. 5Ab). In the presence of SB-366791, sEPSC fre- rat SG; see Lappin et al., 2006), also inhibited RTX activity quency and amplitude were 132Ϯ13% (nϭ4; PϾ0.05) and 99Ϯ1% (nϭ4; PϾ0.05), respectively, of control (7.1Ϯ1.5 ␮ Table 1. Peak amplitudes of resiniferatoxin (RTX) (0.005–2 M)- Hz, 9.2Ϯ0.4 pA) between 3.5 and 4 min after RTX treat- induced inward currents in SG neurons and the numbers of neurons exhibiting these currents ment, and there was no change in holding currents in one neuron (Fig. 5Ab) and a small inward current of 4.4Ϯ1.3 ϭ Concentration (␮M) Peak amplitude (pA)a Cell number ratiob pA (n 3) in other three neurons, where SB-366791 itself hardly affected sEPSC frequency and amplitude (103Ϯ1% 0.005 4.3Ϯ0.3 3/4 and 101Ϯ1%, respectively, of control). Thus, changes in 0.01 5.0Ϯ0.8 3/4 sEPSC frequency and amplitude and in holding current Ϯ 0.02 16.5 5.2 3/4 produced by RTX are mediated by TRPV1. 0.05 8.9Ϯ1.3 4/9 In contrast, the Naϩ-channel blocker, TTX, did not 0.1 8.6Ϯ0.9 4/4 0.2 12.2Ϯ2.7 7/8 affect RTX activity (Fig. 5B), producing sEPSC frequency 0.5 10.7Ϯ1.5 15/26 and amplitude values of 252Ϯ45% (nϭ5; PϽ0.05) and 1 10.2, 9.6 2/6 126Ϯ9% (nϭ5; PϽ0.05) of control (11.4Ϯ3.1 Hz, 11.9Ϯ 2 20.0, 5.6 2/5 1.5 pA), respectively, between 3.5 and 4 min after RTX a Values are meansϮSEM. addition. In four of the five cells tested, the peak inward b Ratio of the number of cells exhibiting inward current to that of the current amplitude in the presence of TTX was 11.1Ϯ1.0 pA cells examined. (nϭ4), a value comparable to that (10.7Ϯ1.5 pA, nϭ15) in C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844 1837

sient receptor potential A1 (TRPA1), respectively (Yang et al., 1998; Kosugi et al., 2007). We next examined whether the facilitatory effect of RTX interacts with that of capsaicin (2 ␮M) or AITC (100 ␮M); these concentrations can en- hance excitatory transmission (Yang et al., 1998; Kosugi et al., 2007). In neurons where RTX enhanced excitatory transmission, the transmission was not affected by capsa- icin applied for 1 min at 25 min after RTX washout (Fig. 6A). sEPSC frequency and amplitude between 3.5 and 4

Fig. 3. RTX dose-dependently increases sEPSC frequency with a small increase in amplitude. (A) Chart recordings of sEPSCs in the absence and presence of RTX at 0.005, 0.1 and 2 ␮M; they were obtained from different neurons. (B) sEPSC frequency and amplitude, relative to control, plotted against the logarithm of RTX concentration. This RTX effect was measured between 3.5 and 4 min after RTX addition. sEPSC amplitude was plotted in all neurons examined (nϭ4, 4, 4, 9, 4, 8, 26, 6 and 5 at 0.005, 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1 and 2 ␮M, respectively; control: 12.4Ϯ0.3 pA, nϭ70) while the frequency was only in neurons exhibiting a change being larger than 10% (75, 75, 75, 89, 100, 88, 88, 83 and 100% of the neurons tested at 0.005, 0.01, 0.02, 0.05 0.1, 0.2, 0.5, 1 and 2 ␮M, respectively; control: 12.8Ϯ0.9 Hz, nϭ61). Each point with vertical bars represents the mean and SEM. The SEM of the values without a vertical bar was within the size of symbol. The dose–response curve for sEPSC frequency was drawn ϭ ␮ ϭ according to the Hill equation (EC50 0.21 M; Hill coefficient 0.64). ϭϪ VH 70 mV. the absence of TTX. The remaining neuron did not change holding currents. The non-NMDA receptor antagonist, CNQX (10 ␮M; nϭ5), or CNQX plus an NMDA receptor antagonist, APV (50 ␮M; nϭ5), reduced sEPSC amplitude even with RTX treatment (Fig. 5C, D). In the presence of CNQX, RTX produced a small inward current in neurons tested (nϭ6; see Fig. 5C) with a peak amplitude of 6.0Ϯ1.1 pA, smaller than that in the absence of CNQX (10.7Ϯ1.5 pA, nϭ15; Fig. 4. The facilitatory effect of RTX (0.5 ␮M) on excitatory transmis- PϽ0.05). CNQX and APV completely blocked the change sion does not depend on its application time in a range of 0.5–2 min (A, in holding currents produced by RTX in all neurons exam- B) Recordings of sEPSCs in the absence and presence of RTX for a ined (nϭ5; Fig. 5D). period of 0.5 min (A) or 2 min (B). (C) sEPSC frequency (left) and amplitude (right) with RTX superfused for 0.5, 1 or 2 min, relative to control [8.3Ϯ3.9 Hz, 10.4Ϯ0.5 pA (nϭ4); 11.8Ϯ1.6 Hz, 12.4Ϯ0.7 pA Capsaicin but not AITC inhibits the effect of RTX on (nϭ23); 10.5Ϯ3.1 Hz, 12.9Ϯ0.9 pA (nϭ5), respectively]; shown here spontaneous excitatory transmission are data of neurons exhibiting sEPSC frequency increases larger than 10%, measured between 3.5 and 4 min after RTX addition. In this and Capsaicin and AITC enhance spontaneous excitatory subsequent figures, value in parentheses denotes the number of ϭϪ transmission in SG neurons by activating TRPV1 and tran- neurons examined; ns: not significant. VH 70 mV. 1838 C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844

Fig. 5. RTX (0.5 ␮M) enhances excitatory transmission by directly activating transient receptor potential vanilloid type 1 (TRPV1). (A–D) sEPSC recordings in the absence and presence of RTX plus a TRPV1 antagonist capsazepine (Capz, 10 ␮M; Aa) or SB-366791 (30 ␮M; Ab.), a Naϩ-channel blocker tetrodotoxin (TTX, 0.5 ␮M; B), a non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 ␮M; C) or both CNQX (10 ␮ ␮ ϭϪ M) and an NMDA-receptor antagonist DL-2-amino-5-phosphonovaleric acid (APV, 50 M; D). VH 70 mV. min after capsaicin addition were, respectively, 96Ϯ5% (Fig. 6C). RTX produced a peak inward current of (nϭ4; PϾ0.05) and 101Ϯ2% (nϭ4; PϾ0.05) of control 10.0Ϯ4.4 pA (nϭ3) in three of the four neurons (Fig. 6A) C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844 1839

Fig. 6. Interaction between the effects of RTX (0.5 ␮M) and capsaicin (Caps, 2 ␮M) on spontaneous excitatory transmission in SG neurons. (A, B) Recordings of sEPSCs in the absence and presence of RTX or Caps. In (A), Caps was applied 25 min after RTX washout; in (B), RTX was applied 25 min after Caps washout. In (A) and (B), left and right recordings were obtained from the same neuron. (C) sEPSC frequency (left) and amplitude (right) after RTX alone (from neurons exhibiting sEPSC frequency increases larger than 10%), Caps at 25 min after RTX washout (Caps after RTX), Caps alone, or RTX at 25 min after Caps washout (RTX after Caps), relative to control [11.8Ϯ1.6 Hz, 12.4Ϯ0.7 pA (nϭ23); 15.4Ϯ3.3 Hz, 11.9Ϯ1.4 pA (nϭ4); 17.5Ϯ5.3 Hz, 12.6Ϯ1.7 pA (nϭ4); 17.1Ϯ4.7 Hz, 13.2Ϯ1.7 pA (nϭ4), respectively], measured between 3.5 and 4 min after drug addition. ϭϪ VH 70 mV. and the remaining neuron did not respond to RTX; capsa- AITC (100 ␮M) superfused for 2 min enhanced spon- icin produced a peak inward current of 3.8Ϯ0.1 pA (nϭ3), taneous excitatory transmission in most (75%) of SG neu- a value smaller than that without RTX pretreatment rons tested (nϭ12); sEPSC frequency and amplitude were (PϽ0.05; see below), in three neurons, and the one re- 216Ϯ35% (nϭ9; PϽ0.05) and 118Ϯ9% (nϭ9; PϾ0.05) of maining neuron did not respond to capsaicin (Fig. 6A). control, respectively, around 3 min after AITC addition. However, RTX treatment at 25 min after capsaicin washout AITC also produced a peak inward current of 8.6Ϯ1.2 pA did not change excitatory transmission, with sEPSC fre- (nϭ6) in six of the nine neurons (Fig. 7Aa). In the three quency and amplitude between 3.5 and 4 min after cap- remaining neurons (25%), excitatory transmission was not saicin addition of 192Ϯ16% (nϭ4; PϽ0.05) and 112Ϯ4% affected by AITC (Fig. 7Ab). We tested the effects of AITC (nϭ4; PϾ0.05) of control, respectively, and after RTX ad- and then RTX (0.5 ␮M) on spontaneous excitatory trans- dition 94Ϯ7% (nϭ4; PϾ0.05) and 106Ϯ2% (nϭ4; PϾ0.05) mission in eight neurons (Fig. 7A). In four of six AITC- of control, respectively. Capsaicin produced a peak inward sensitive neurons, RTX increased sEPSC frequency to current of 14.5Ϯ3.3 pA (nϭ4) in all four neurons tested 197Ϯ30% (nϭ4; PϽ0.05) of control (Fig. 7Aa), but not in (Fig. 6B). In three of the four cells, RTX produced a peak the other two (106% and 107% of control). In two neurons inward current of 5.7Ϯ1.1 pA (nϭ3; Fig. 6B), a value insensitive to AITC, where sEPSC frequency was 104 or smaller than that (10.7Ϯ1.5 pA, nϭ15) without capsaicin 100% of control, RTX slightly increased sEPSC frequency pretreatment (PϽ0.05); the remaining neuron did not (147% and 158% of control; Fig. 7Ab). AITC produced a change holding currents. peak inward current of 9.0Ϯ0.3 pA (nϭ3) in three of six 1840 C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844

Fig. 7. Effects of a transient receptor potential A1 (TRPA1) agonist allyl isothiocyanate (AITC; 100 ␮M) and RTX (0.5 ␮M) on spontaneous excitatory transmission in the same SG neuron (Aa, b, B). Recordings of sEPSCs in the absence or presence of AITC or RTX, where left and right recordings were obtained from the same neuron. In (A), RTX was applied 25 min after AITC washout; in (B), AITC was applied 25 min after RTX washout. In (Ac), sEPSC frequency (left) and amplitude (right) after AITC or RTX, relative to control (AITC: 15.4Ϯ2.8 Hz, 12.6Ϯ0.6 pA; RTX: 20.0Ϯ3.8 Hz, 15.5Ϯ1.2 pA; nϭ8), measured between 2.5 and 3 min after AITC addition or between 3.5 and 4 min after RTX addition. Data obtained from the same neuron ϭϪ are connected by a straight line. VH 70 mV. neurons (Fig. 7Aa). One of the three neurons had a peak rons; one of the two neurons produced a peak inward inward current of 15.5 pA, produced by RTX (Fig. 7Aa), current of 6.4 pA in response to RTX and the other did not comparable to that (10.7Ϯ1.5 pA, nϭ15) without AITC (Fig. 7Ab). The effects of AITC and RTX on sEPSC fre- pretreatment; the two remaining neurons did not change quency and amplitude are summarized in Fig. 7Ac. holding currents. Both AITC and RTX did not change hold- We further examined the effect of AITC (100 ␮M) on ing currents in the three remaining neurons. AITC did not spontaneous excitatory transmission in eight SG neurons change sEPSC frequency or holding currents in two neu- at 25 min after the treatment of spinal cord slices with RTX C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844 1841

(0.5 ␮M) for 2 min. In five neurons where RTX increased leading to enhanced spontaneous release of L-glutamate. sEPSC frequency [237Ϯ47% (nϭ5; PϽ0.05) of control], TRPV1 has a high Ca2ϩ permeability (Caterina et al., AITC did not affect sEPSC frequency [98Ϯ4% (nϭ5; 1997), and TRPV1 activation mobilizes Ca2ϩ from intra- PϾ0.05) of control; Fig. 7B]. Three of the five neurons cellular Ca2ϩ stores in TRPV1-transfected HEK293 cells exhibited a peak inward current of 6.4Ϯ1.4 pA (nϭ3), (Marshall et al., 2003). Moreover, TRPV1 activation by produced by RTX, and the two remaining neurons did not capsaicin at synapses in DRG/spinal cord co-cultures pro- change holding currents, while AITC did not produce any longs the elevation of intraterminal Ca2ϩ levels and in- inward current in all of the five neurons. Three of the eight creases L-glutamate release (Medvedeva et al., 2008). neurons tested did not respond to both RTX and AITC. RTX-mediated presynaptic enhancement and inward current recovered from desensitization slowly. RTX showed a bell-shaped dose–response for changes in sEPSC fre- DISCUSSION quency, and this presynaptic effect did not depend on RTX Superfusing RTX increases sEPSC frequency with a min- perfusion time in the range of 0.5–2 min, suggesting that imal change in sEPSC amplitude in about 90% of SG this presynaptic effect rapidly desensitized at higher doses neurons, with about half of the neurons producing an in- and with increased treatment time. A slow recovery from ward current at Ϫ70 mV. This increase in sEPSC fre- desensitization was seen at a low concentration such as quency is consistent with the observation that RTX re- 0.05 ␮M at 20 min after washout of RTX. This may be due duces the inter-event interval of sEPSC, indicating presyn- to a high affinity of RTX for TRPV1 (Szallasi et al., 1993). aptic activity. RTX produced presynaptic enhancement The EC50 value for RTX to increase sEPSC frequency was and inward currents similar to those induced by capsaicin 0.21 ␮M (using values below 0.5 ␮M), less than the in rat SG neurons (Yang et al., 1998, 2000; Morisset and amount (1 ␮M; Yang et al., 1998) needed for capsaicin to Urban, 2001) and substantia nigra dopaminergic neurons increase sEPSC frequency in SG neurons, being consis- (Marinelli et al., 2003). The presynaptic effect produced by tent with a higher affinity for TRPV1 (Szallasi et al., 1993). RTX as well as capsaicin was resistant to TTX, indicating RTX produced an inward current in about a half of the a direct action of RTX (Yang et al., 1998), although there neurons tested, but required lower concentrations (0.005– was a difference in TTX effect between the inward currents 0.5 ␮M) than capsaicin (Table 1). However, the maximal produced by RTX and capsaicin (see below). The presyn- effects of RTX (0.5 ␮M) and capsaicin (2 ␮M) on sEPSC aptic effect of RTX (0.5 ␮M) appeared to be slower in onset frequency and holding current were comparable (Fig. 3B, than that of capsaicin (2 ␮M; Fig. 6B and Yang et al., Table 1 and Yang et al., 1998). Consistent with this obser- 1998). The difference in onset is similar for inward currents vation, RTX (0.3 ␮M) and capsaicin (10 ␮M) show similar induced by capsaicin (1 ␮M) and RTX (0.1 ␮M) in cultured increases in sEPSC frequency in substantia nigra neurons rat DRG neurons (Raisinghani et al., 2005). Presynaptic (Marinelli et al., 2003), as well as similar activity in con- facilitation in SG neurons as well as inward current produced tracting the rat isolated urinary bladder (Maggi et al., in DRG neurons persisted for at least 10 min after RTX 1990). These comparable maximal effects may occur be- washout (Fig. 1B and Raisinghani et al., 2005). The result cause TRPV1 rapidly desensitizes after activation. that the RTX effects were observed in some but not all SG RTX-induced inward current was resistant to TTX, in- neurons tested may be due to the fact that the SG is com- dicating that RTX may have activated TRPV1 in postsyn- prised of a heterogenous cell group (Grudt and Perl, 2002). aptic neurons. TRPV1 exists in both pre- and postsynaptic Although RTX has TRPV1-independent activity on volt- neurons in the superficial laminae of the rat spinal dorsal age-gated ion channels expressed in clonal neuroendo- horn (Valtschanoff et al., 2001) and mouse brain (Cristino crine cells (Sugimoto et al., 2008), both pre- and postsyn- et al., 2006). However, RTX-induced inward current was aptic actions produced by RTX in SG neurons are inhibited blocked by CNQX and APV treatment, indicating that the by the TRPV1 antagonist capsazepine or SB-366791, in- inward current resulted from activation of non-NMDA and dicating TRPV1 involvement. This idea is supported by the NMDA receptors by L-glutamate released from the primary- observations that capsaicin inhibits the effects of RTX on afferent terminals after TRPV1 activation. NMDA receptors excitatory transmission and vice versa, i.e., cross-desen- are present in the SG (Tölle et al., 1993) and respond to sitization, and is consistent with the existence of TRPV1 in exogenous NMDA and dorsal root stimulation (Yajiri et al., the central terminals of primary-afferent neurons (Guo et 1997). AITC activates NMDA receptors by promoting ex- al., 1999; Valtschanoff et al., 2001; Hwang et al., 2004). A cess of L-glutamate released from nerve terminals (Kosugi similar cross-desensitization of RTX and capsaicin activity et al., 2007). Capsaicin produces an inward current by occurs for the release of substance P or calcitonin gene- activating interneurons via non-NMDA receptors after L- related peptide from the central terminals of primary-affer- glutamate is released from primary-afferent central termi- ent neurons in the rat spinal cord, as well as in contraction nals, leading to the release of neurotransmitters other than involving transmitter release from the peripheral terminals substance P that produce the inward current. Thus, the of primary-afferent neurons in the rat isolated urinary blad- inward current produced by capsaicin can be blocked by der (Maggi et al., 1990). TTX and CNQX, and is resistant to APV (Yang et al., 1998, TRPV1 activation increases extracellular Ca2ϩ influx 2000). Even if RTX binds to the same TRPV1 site as and subsequent Ca2ϩ mobilization from intracellular Ca2ϩ capsaicin (see Szallasi and Blumberg, 1990), the activity stores in the central terminals of primary-afferent neurons, on excitatory transmission may be different. RTX but not 1842 C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844 capsaicin mobilizes Ca2ϩ from inositol 1,4,5-trisphosphate- with fast desensitization and slow recovery from desensi- sensitive Ca2ϩ stores in TRPV1-transfected HEK293 cells tization; this maximal effect is similar to capsaicin. (Marshall et al., 2003). Understanding the differences in inward currents produced by RTX and capsaicin in SG REFERENCES neurons requires further work. Superfusing AITC (100 ␮M) for 2 min did not affect the Acs G, Palkovits M, Blumberg PM (1996) Specific binding of [3H]res- ability of RTX to increase sEPSC frequency or produce iniferatoxin by human and rat preoptic area, locus ceruleus, medial hypothalamus, reticular formation and ventral thalamus membrane inward currents 25 min after washout of AITC (Fig. 7A). In preparations. Life Sci 59:1899–1908. neurons where AITC showed little facilitatory effect on exci- Akopian AN, Ruparel NB, Jeske NA, Hargreaves KM (2007) Transient tatory transmission, RTX also weakly increased sEPSC fre- receptor potential TRPA1 channel desensitization in sensory neu- quency (Fig. 7Ab, c), suggesting that TRPV1 and TRPA1 rons is agonist dependent and regulated by TRPV1-directed inter- may correlate with each other, as suggested by in situ nalization. J Physiol 583:175–193. hybridization studies (Story et al., 2003; Kobayashi et al., Bron R, Klesse LJ, Shah K, Parada LF, Winter J (2003) Activation of Ras is necessary and sufficient for upregulation of vanilloid recep- 2005). On the other hand, the pretreatment with RTX (0.5 tor type 1 in sensory neurons by neurotrophic factors. Mol Cell ␮M) for 2 min resulted in the lack of the ability of AITC to Neurosci 22:118–132. enhance excitatory transmission 25 min after washout of Brown DC, Iadarola MJ, Perkowski SZ, Erin H, Shofer F, Laszlo KJ, RTX. Unlike RTX, capsaicin (2 ␮M) did not affect the Olah Z, Mannes AJ (2005) Physiologic and antinociceptive effects facilitatory effect of AITC (Kosugi et al., 2007). Such a of intrathecal resiniferatoxin in a canine bone cancer model. An- esthesiology 103:1052–1059. difference between RTX and capsaicin actions may be due Caterina MJ, Julius D (2001) The vanilloid receptor: a molecular to a distinction in the affinity for TRPV1 between them. gateway to the pain pathway. Annu Rev Neurosci 24:487–517. Akopian et al. (2007) have reported a cross desensitization Caterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD, between TRPV1 and TRPA1 in producing inward currents Julius D (1997) The capsaicin receptor: a heat-activated ion chan- in cultured rat sensory neurons at an interval of 3 min. nel in the pain pathway. Nature 389:816–824. Further study is required to know an interaction between Chan CLH, Facer P, Davis JB, Smith GD, Egerton J, Bountra C, Williams NS, Anand P (2003) Sensory fibres expressing capsaicin TRPV1 and TRPA1 in nerve terminals. receptor TRPV1 in patients with rectal hypersensitivity and faecal urgency. Lancet 361:385–391. TRPV1 and clinical use of RTX in the nervous Cristino L, de Petrocellis L, Pryce G, Baker D, Guglielmotti V, Di Marzo system V (2006) Immunohistochemical localization of cannabinoid type 1 and vanilloid transient receptor potential vanilloid type 1 receptors Inflammatory hyperalgesia and neuropathic pain develop in the mouse brain. Neuroscience 139:1405–1415. due to TRPV1 over-expression or ectopic expression in Culshaw AJ, Bevan S, Christiansen M, Copp P, Davis A, Davis C, both the CNS and peripheral nervous system, in animals Dyson A, Dziadulewicz EK, Edwards L, Eggelte H, Fox A, Gentry and in humans (Hudson et al., 2001; Walker et al., 2003; C, Groarke A, Hallett A, Hart TW, Hughes GA, Knights S, Kotsonis P, Lee W, Lyothier I, McBryde A, McIntyre P, Paloumbis G, Pan- Gavva et al., 2005; Culshaw et al., 2006). TRPV1-knock esar M, Patel S, Seiler M-P, Yaqoob M, Zimmermann K (2006) out mice exhibit reduced hyperalgesia in inflammatory pain Identification and biological characterization of 6-aryl-7-isopropy- models (Davis et al., 2000). Nerve growth factor, a key lquinazolinones as novel TRPV1 antagonists that are effective in mediator of inflammatory pain, upregulates TRPV1 ex- models of chronic pain. J Med Chem 49:471–474. pression in rat DRG neurons through the small GTPase Davis JB, Gray J, Gunthorpe MJ, Hatcher JP, Davey PT, Overend P, Ras (Bron et al., 2003). TRPV1 is expressed in neurons Harries MH, Latcham J, Clapham C, Atkinson K, Hughes SA, Rance K, Grau E, Harper AJ, Pugh PL, Rogers DC, Bingham S, involved in the sensation of cancer pain and is a potential Randall A, Sheardown SA (2000) Vanilloid receptor-1 is essential treatment target for this pain (for review see Prevarskaya for inflammatory thermal hyperalgesia. Nature 405:183–187. et al., 2007). Selective blockade of TRPV1 attenuates Fujita T, Kumamoto E (2006) Inhibition by endomorphin-1 and endo- bone cancer pain in the mouse (Ghilardi et al., 2005). morphin-2 of excitatory transmission in adult rat substantia gelati- Intrathecal administration of RTX in dogs with bone cancer nosa neurons. Neuroscience 139:1095–1105. Gavva NR, Tamir R, Qu Y, Klionsky L, Zhang TJ, Immke D, Wang J, produces a prolonged antinociceptive response (Brown et Zhu D, Vanderah TW, Porreca F, Doherty EM, Norman MH, Wild al., 2005). Upregulated TRPV1 expression in primary-af- KD, Bannon AW, Louis J-C, Treanor JJS (2005) AMG 9810 [(E)- ferent fibers occurs in disease states such as inflammatory 3-(4-t-butylphenyl)-N-(2,3-dihydrobenzo[b][1,4]dioxin-6-yl)acrylamide], disease and irritable bowel syndrome (Yiangou et al., a novel vanilloid receptor 1 (TRPV1) antagonist with antihyperal- 2001; Chan et al., 2003). Peripheral inflammation upregu- gesic properties. J Pharmacol Exp Ther 313:474–484. lates TRPV1 involved in enhancing excitatory transmission Ghilardi JR, Röhrich H, Lindsay TH, Sevcik MA, Schwei MJ, Kubota K, Halvorson KG, Poblete J, Chaplan SR, Dubin AE, Carruthers NI, in rat SG neurons (Lappin et al., 2006). TRPV1 activation Swanson D, Kuskowski M, Flores CM, Julius D, Mantyh PW (2005) in the CNS can also mediate antinociception (Starowicz et Selective blockade of the capsaicin receptor TRPV1 attenuates al., 2007a) and RTX blockade of abnormal TRPV1 activity bone cancer pain. J Neurosci 25:3126–3131. may have beneficial effects. Grudt TJ, Perl ER (2002) Correlations between neuronal morphology and electrophysiological features in the rodent superficial dorsal horn. J Physiol 540:189–207. CONCLUSION Guo A, Vulchanova L, Wang J, Li X, Elde R (1999) Immunocytochem- ical localization of the vanilloid receptor 1 (VR1): relationship to In conclusion, RTX acts on TRPV1 in glutamatergic nerve neuropeptides, the P2X3 purinoceptor and IB4 binding sites. Eur terminals in the SG to spontaneously release L-glutamate J Neurosci 11:946–958. C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844 1843

Hergenhahn M, Adolf W, Hecker E (1975) Resiniferatoxin and other Prevarskaya N, Zhang L, Barritt G (2007) TRP channels in cancer. esters of novel polyfunctional diterpenes from Euphorbia resinifera Biochim Biophys Acta 1772:937–946. and unispina. Tetrahedron Lett 19:1595–1598. Raisinghani M, Pabbidi RM, Premkumar LS (2005) Activation of tran- Hudson LJ, Bevan S, Wotherspoon G, Gentry C, Fox A, Winter J sient receptor potential vanilloid 1 (TRPV1) by resiniferatoxin. (2001) VR1 protein expression increases in undamaged DRG J Physiol 567:771–786. neurons after partial nerve injury. Eur J Neurosci 13:2105–2114. Sasamura T, Sasaki M, Tohda C, Kuraishi Y (1998) Existence of Hwang SJ, Burette A, Rustioni A, Valtschanoff JG (2004) Vanilloid capsaicin-sensitive glutamatergic terminals in rat hypothalamus. receptor VR1-positive primary afferents are glutamatergic and con- NeuroReport 9:2045–2048. tact spinal neurons that co-express neurokinin receptor NK1 and Schmidt RJ, Evans FJ (1979) Investigations into the skin-irritant prop- glutamate receptors. J Neurocytol 33:321–329. erties of resiniferonol ortho esters. Inflammation 3:273–280. Hwang SW, Cho H, Kwak J, Lee S-Y, Kang C-J, Jung J, Cho S, Min Starowicz K, Cristino L, Di Marzo V (2008) TRPV1 receptors in the KH, Suh Y-G, Kim D, Oh U (2000) Direct activation of capsaicin central nervous system: potential for previously unforeseen thera- receptors by products of lipoxygenases: endogenous capsaicin- peutic applications. Curr Pharm Des 14:42–54. like substances. Proc Natl Acad SciUSA97:6155–6160. Starowicz K, Maione S, Cristino L, Palazzo E, Marabese I, Rossi F, de Iyadomi M, Iyadomi I, Kumamoto E, Tomokuni K, Yoshimura M (2000) Novellis V, Di Marzo V (2007a) Tonic endovanilloid facilitation of Presynaptic inhibition by baclofen of miniature EPSCs and IPSCs glutamate release in brainstem descending antinociceptive path- in substantia gelatinosa neurons of the adult rat spinal dorsal horn. ways. J Neurosci 27:13739–13749. Pain 85:385–393. Starowicz K, Niga S, Di Marzo V (2007b) Biochemistry and pharma- Jennings EA, Vaughan CW, Roberts LA, Christie MJ (2003) The cology of endovanilloids. Pharmacol Ther 114:13–33. actions of anandamide on rat superficial medullary dorsal horn Story GM, Peier AM, Reeve AJ, Eid SR, Mosbacher J, Hricik TR, neurons in vitro. J Physiol 548:121–129. Earley TJ, Hergarden AC, Andersson DA, Hwang SW, McIntyre P, Jung J, Lee S-Y, Hwang SW, Cho H, Shin J, Kang Y-S, Kim S, Oh U Jegla T, Bevan S, Patapoutian A (2003) ANKTM1, a TRP-like (2002) Agonist recognition sites in the cytosolic tails of vanilloid channel expressed in nociceptive neurons, is activated by cold receptor 1. J Biol Chem 277:44448–44454. temperatures. Cell 112:819–829. Kobayashi K, Fukuoka T, Obata K, Yamanaka H, Dai Y, Tokunaga A, Sugimoto K, Kissin I, Strichartz G (2008) A high concentration of Noguchi K (2005) Distinct expression of TRPM8, TRPA1, and TRPV1 resiniferatoxin inhibits ion channel function in clonal neuroendo- mRNAs in rat primary afferent neurons with A␦/C-fibers and colocal- crine cells. Anesth Analg 107:318–324. ization with Trk receptors. J Comp Neurol 493:596–606. Szallasi A (1994) The vanilloid (capsaicin) receptor: receptor types Kosugi M, Nakatsuka T, Fujita T, Kuroda Y, Kumamoto E (2007) and species differences. Gen Pharmacol 25:223–243. Activation of TRPA1 channel facilitates excitatory synaptic trans- Szallasi A (2002) Vanilloid (capsaicin) receptors in health and disease. mission in substantia gelatinosa neurons of the adult rat spinal Am J Clin Pathol 118:110–121. cord. J Neurosci 27:4443–4451. Szallasi A, Blumberg PM (1989) Resiniferatoxin, a phorbol-related Lappin SC, Randall AD, Gunthorpe MJ, Morisset V (2006) TRPV1 diterpene, acts as an ultrapotent analog of capsaicin, the irritant antagonist, SB-366791, inhibits glutamatergic synaptic transmis- constituent in red pepper. Neuroscience 30:515–520. sion in rat spinal dorsal horn following peripheral inflammation. Eur Szallasi A, Blumberg PM (1990) Specific binding of resiniferatoxin, an J Pharmacol 540:73–81. ultrapotent capsaicin analog, by dorsal root ganglion membranes. Liu T, Fujita T, Nakatsuka T, Kumamoto E (2008) Phospholipase A 2 Brain Res 524:106–111. activation enhances inhibitory synaptic transmission in rat substan- Szallasi A, Di Marzo V (2000) New perspectives on enigmatic vanilloid tia gelatinosa neurons. J Neurophysiol 99:1274–1284. receptors. Trends Neurosci 23:491–497. Maggi CA, Patacchini R, Tramontana M, Amann R, Giuliani S, Santi- Szallasi A, Goso C, Blumberg PM, Manzini S (1993) Competitive cioli P (1990) Similarities and differences in the action of resinif- inhibition by capsazepine of [3H]resiniferatoxin binding to central eratoxin and capsaicin on central and peripheral endings of pri- mary sensory neurons. Neuroscience 37:531–539. (spinal cord and dorsal root ganglia) and peripheral (urinary blad- Marinelli S, Di Marzo V, Berretta N, Matias I, Maccarrone M, Bernardi der and airways) vanilloid (capsaicin) receptors in the rat. J Phar- G, Mercuri NB (2003) Presynaptic facilitation of glutamatergic syn- macol Exp Ther 267:728–733. apses to dopaminergic neurons of the rat substantia nigra by Szallasi A, Nilsson S, Farkas-Szallasi T, Blumberg PM, Hökfelt T, endogenous stimulation of vanilloid receptors. J Neurosci 23: Lundberg JM (1995) Vanilloid (capsaicin) receptors in the rat: 3136–3144. distribution in the brain, regional differences in the spinal cord, Marshall ICB, Owen DE, Cripps TV, Davis JB, McNulty S, Smart D axonal transport to the periphery, and depletion by systemic va- (2003) Activation of vanilloid receptor 1 by resiniferatoxin mobilizes nilloid treatment. Brain Res 703:175–183. calcium from inositol 1,4,5-trisphosphate-sensitive stores. Br J Tölle TR, Berthele A, Zieglgänsberger W, Seeburg PH, Wisden W Pharmacol 138:172–176. (1993) The differential expression of 16 NMDA and non-NMDA McGaraughty S, Chu KL, Bitner RS, Martino B, El Kouhen R, Han P, receptor subunits in the rat spinal cord and in periaqueductal gray. Nikkel AL, Burgard EC, Faltynek CR, Jarvis MF (2003) Capsaicin J Neurosci 13:5009–5028. infused into the PAG affects rat tail flick responses to noxious heat Valtschanoff JG, Rustioni A, Guo A, Hwang SJ (2001) Vanilloid recep- and alters neuronal firing in the RVM. J Neurophysiol 90:2702–2710. tor VR1 is both presynaptic and postsynaptic in the superficial Medvedeva YV, Kim M-S, Usachev YM (2008) Mechanisms of prolonged laminae of the rat dorsal horn. J Comp Neurol 436:225–235. presynaptic Ca2ϩ signaling and glutamate release induced by TRPV1 Walker KM, Urban L, Medhurst SJ, Patel S, Panesar M, Fox AJ, activation in rat sensory neurons. J Neurosci 28:5295–5311. McIntyre P (2003) The VR1 antagonist capsazepine reverses me- Mezey É, Tóth ZE, Cortright DN, Arzubi MK, Krause JE, Elde R, Guo chanical hyperalgesia in models of inflammatory and neuropathic A, Blumberg PM, Szallasi A (2000) Distribution of mRNA for va- pain. J Pharmacol Exp Ther 304:56–62. nilloid receptor subtype 1 (VR1), and VR1-like immunoreactivity, in Willis WD Jr, Coggeshall RE (1991) Structure of the dorsal horn. In: the central nervous system of the rat and human. Proc Natl Acad Sensory mechanisms of the spinal cord, 2nd ed, pp 94–115. New SciUSA97:3655–3660. York: Plenum Press. Morisset V, Urban L (2001) Cannabinoid-induced presynaptic inhibi- Xing J, Li J (2007) TRPV1 receptor mediates glutamatergic synaptic tion of glutamatergic EPSCs in substantia gelatinosa neurons of input to dorsolateral periaqueductal gray (dl-PAG) neurons. J Neu- the rat spinal cord. J Neurophysiol 86:40–48. rophysiol 97:503–511. 1844 C.-Y. Jiang et al. / Neuroscience 164 (2009) 1833–1844

Yajiri Y, Yoshimura M, Okamoto M, Takahashi H, Higashi H (1997) A Yiangou Y, Facer P, Dyer NHC, Chan CLH, Knowles C, Williams NS, novel slow excitatory postsynaptic current in substantia gelatinosa Anand P (2001) Vanilloid receptor 1 immunoreactivity in inflamed neurons of the rat spinal cord in vitro. Neuroscience 76:673–688. human bowel. Lancet 357:1338–1339.

Yang K, Kumamoto E, Furue H, Yoshimura M (1998) Capsaicin facilitates Yue H-Y, Fujita T, Kumamoto E (2005) Phospholipase A2 activation by excitatory but not inhibitory synaptic transmission in substantia gelati- melittin enhances spontaneous glutamatergic excitatory transmission nosa of the rat spinal cord. Neurosci Lett 255:135–138. in rat substantia gelatinosa neurons. Neuroscience 135:485–495. Yang K, Kumamoto E, Furue H, Li Y-Q, Yoshimura M (2000) Capsa- Zygmunt PM, Petersson J, Andersson DA, Chuang H-h, Sørgård M, Di icin induces a slow inward current which is not mediated by sub- Marzo V, Julius D, Högestätt ED (1999) Vanilloid receptors on stance P in substantia gelatinosa neurons of the rat spinal cord. sensory nerves mediate the vasodilator action of anandamide. Neuropharmacology 39:2185–2194. Nature 400:452–457.

(Accepted 15 September 2009) (Available online 22 September 2009)