Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Since the discovery of nociceptin or orphanin FQ (N/OFQ), the endogenous ligand for opioid-like orphan receptor 1, there have been many reports that N/OFQ has both pronociceptive (or hyperalgesic) and antinociceptive properties (Inoue et al., 1999; for review, see Mogil and Pasternak, 2001). On the other hand, the supraspinal administration of antisense oligodeoxynucleotide or antagonist for N/OFQ receptor (NOP) caused an increase in nociceptive threshold (Meunier et al., 1995; Rossi et al., 1997; Zhu et al., 1997; Calo et al., 2000, 2002; Shinkai et al., 2000), whereas NOP^/ mice displayed normal baseline nociceptive responses in some analgesic paradigms (Nishi et al., 1997; Mamiya et al., 1998; Ozaki et al., 2000). These findings suggest that N/OFQ might play differential pain modulatory roles. In most popular analgesic paradigms we use, various nociceptive thermal, mechanical, and chemical stimulations might activate distinct types of fibers at the same time. These fibers might include both pain-stimulatory and -inhibitory ones, according to the gate control theory (Melzack and Wall, 1965). The late nociceptive responses in some paradigms might be modulated to some extent by descending pain-inhibitory mechanisms secondary to the initial nociceptive input (Fields, 1987). Thus, it is important to use nociception tests based on the molecularly distinct nociceptive stimulation, which causes rapid nociceptive behaviors, in the attempt to characterize the modality (or nociceptor)-specific role of specific neurotransmitters or neuropeptides. Algogenic-induced nociceptive flexion (ANF) test in mice would be the one we have recently developed to clarify the distinct roles of such neurotransmitters or neuropeptides in the nociceptor-specific pain regulation. Here, we report the in vivo inhibitory role of spinal N/OFQ-ergic neurons for the pain after polymodal substance P-ergic fiber stimulation, by use of NOP^/ mice and selective NOP antagonists in the ANF test and other known analgesiometric assays such as paw pressure, Hargreaves thermal nociception, and capsaicin tests.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Male ddY mice weighing 20 to 22 g were used. Mutant mice were homozygotes (NOP^/) lacking the genomic NOP gene, heterozygotes (NOP^+/), and its wild-type (NOP^+/+), which have been developed previously (Nishi et al., 1997), and housed in a group of 10 animals. They were kept in a room maintained at 21 ± 2°C with free access to a standard laboratory diet and tap water. Procedures were approved by Nagasaki University Animal Care Committee and complied with the recommendations of the International Association for the Study of Pain (Zimmermann, 1983).

Drugs

The following drugs were used: N/OFQ (Sawady Technology, Tokyo, Japan), substance P (Peptide Institute, Osaka, Japan), bradykinin (Sigma-Aldrich, St. Louis, MO), adenosine triphosphate and capsaicin (Nacalai Tesque, Kyoto, Japan), and MK-801 (Sigma/RBI, Natick, MO). ONO-54918-07 (a stable prostaglandin I₂ agonist) was gift from Ono Pharmaceutical Co., Ltd. (Tokyo, Japan). CP-99994 was generously provided by Pfizer Pharmaceuticals (Sandwich, Kent, UK). 1-[(3R,4R)-1-Cyclooctylmethyl-3-hydroxymethyl-4-piperidyl]-3-ethyl-1,3-dihydro-2H-benzimidazole-2-one (J-113397) was generously provided by Banyu (Tsukuba, Japan) and [Nphe¹]N/OFQ(1-13)NH₂ was a gift from Prof. S. Salvadori and Dr. R. Guerrini (Department of Pharmaceutical Sciences, University of Ferrara, Ferrara, Italy). All drugs except capsaicin were dissolved in physiological saline. Capsaicin was dissolved in 10% ethanol and 10% Tween 80 in physiological saline.

Intrathecal Injection

The i.t. injection was adopted according to the method of Hylden and Wilcox (1980). A 28-gauge stainless steel needle attached to a 50-µl Hamilton microsyringe was inserted between lumbar 5 and 6 in unanesthetized mice, and drugs were given slowly in a volume of 5 µl.

In Vivo Nociception Test

Tail-Flick Test. Animals were gently restrained by hand, and a light beam adjusted for 10- to 12-s latency in naive mice was focused onto the blackened dorsal surface of the tail. Latency up to a cut-off time of 20 s was measured (Ueda et al., 2000a).

Paw-Pressure (Digital von Frey) Test. Mice were placed in a Plexiglas chamber on a 6 × 6-mm wire mesh grid floor and were allowed to accommodate for a period of 1 h. A mechanical stimulus was then delivered onto the middle of the plantar surface of the right hind paw by using a 0.8- to 0.9-mm-diameter filament connected to an automatic transducer indicator (model 1601; IITC Inc., Woodland Hills, CA), as described by Doboly et al. (2002). The filament used produces 10 g of force at 5 s, when paw withdrawal is elicited in naive mice. A 20-s cut-off time was used to avoid tissue damage.

Hargreaves Thermal Nociception Test. A thermal beam was focused on the hind limb foot pads of mice placed on a glass surface and the withdrawal response latency measured, with a 20-s cut-off time, as described by Hargreaves et al. (1988).

Capsaicin-Induced ABL Test. The algogenic (intraplantarly, i.pl.)-induced biting and licking test was carried out by use of 0.4 or 0.8 µg of capsaicin, as reported previously (Tan-No et al., 1998). A 28-gauge stainless steel needle attached to a 50-µl Hamilton microsyringe was inserted into the foot pad in unanesthetized mice, and capsaicin was given slowly in a volume of 20 µl. Total duration time showing these behaviors during 5 min after i.pl. injection was summed and used as biting and licking responses (in seconds).

ANF Test. Experiments were performed as described previously (Inoue et al., 1998; Ueda, 1999; Doboly et al., 2002). Briefly, mice were held in a cloth sling with their four limbs hanging free through holes. The sling was suspended on a metal bar. All limbs were tied with strings, and three were fixed to the floor, whereas the other one was connected to an isotonic transducer and recorder. A polyethylene cannula (0.61 mm in outer diameter) filled with drug solution was connected to microsyringe and then carefully inserted into the undersurface of the right hind paw. Because we used light and soft polyethylene cannula, it did not fall off the paw during the experiments. Because the intensity of flexor responses differs from mouse to mouse, we used the biggest response among spontaneous and nonspecific flexor responses occurring immediately after cannulation as the maximal reflex. Algogenic substance injection was i.pl. given every 5 min unless otherwise stated. Algogenic substance-induced nociceptive activity was expressed as the ratio of maximal reflex in each mouse, and in the dose-response experiments, increasing doses of compound were given at 5-min intervals. Average of responses by twice-repeated challenges per each dose was evaluated.

Central Algogenic-Induced SBL Test. The nocifensive behaviors characterized by reciprocal hind limb scratching, caudally directed biting, and licking (SBL behavior) during 5 min after intrathecal injection of algogenic were evaluated (Hylden and Wilcox, 1981; Inoue et al., 1998). Before experiments, mice were adapted to an individual plastic cage for 1 h. Immediately after i.t. injection of algogenic (substance P), each mouse was placed into the transparent cage for behavioral tests. All mice were used for only one experiment by the observer who did not know what kind of pretreatments had been given.

Immunohistochemistry

Immunohistochemistry for NK1 tachykinin receptor using free-floating 30-µm section of spinal cord from 4% paraformaldehyde-perfused NOP^+/+ and NOP^/ mice was performed as described previously (Mantyh et al., 1995).

Western Blot Analysis

SDS-polyacrylamide gel electrophoresis by using 12% polyacrylamide gel and immunoblot analysis were performed as described previously (Yoshida and Ueda, 1999). Thirty micrograms of protein extracted from the dorsal horn of the spinal cord was used. To get equal transfer efficiency, we have applied all samples to the same gel and carried out the immunoblot transfer using the same membrane. Visualization of immunoreactive bands was performed by using an enhanced chemiluminescence substrate for detection of horseradish peroxidase, Super Signaling Substrate (Pierce Chemical, Rockford, IL). The intensities of immunoreactive bands were analyzed by NIH Imaging for Macintosh after scanning exposed films.

Receptor Binding

The dorsal horn of spinal cord was isolated from the mouse and the synaptic membranes were prepared, and membrane binding study using [³H]substance P was carried out, according to Inoue et al. (1988). In saturation binding experiments, the membranes were incubated with [³H]substance P at concentrations varying from 0.1 to 1.2 nM in a final volume of 500 µl for 1 h at 25°C. Binding reaction was terminated by rapid filtration of the incubation mixture through GF/B glass filter (Whatman, Maidstone, UK) presoaked with 0.1% polyethyleneimine. The radioactivity content of the filter was determined using a liquid scintillation counter (LSC-5100; Aloka, Tokyo, Japan) at the efficacy of 50%. Nonspecific binding was determined using 1 µM unlabeled substance P.

RT-PCR

Total RNA was isolated from mouse spinal cord with TRIzol (Invitrogen, Carlsbad, CA), and 1 µg was used for cDNA synthesis with Superscript II reverse transcriptase and random hexamer primers (Invitrogen). The cDNA was used as a template for PCR amplification with TaqDNA polymerase (Takara, Kyoto, Japan) and NK1 primers (5'-CAT CAA CCC AGA TCTC TACC-3' and 5'-AGC TGG AGC TTT CTG TCA TGG-3') or GAPDH primers (5'-GTG AAG GTC GGT GTG AAC GGA TTT-3' and 5'-CAC AGT CTT CTG GGT GGC AGT GAT-3'). PCR amplification was carried out under the condition of 28 cycles (for NK1) at 94°C for 30 s, 51°C for 1 min, and 72°C for 1 min or 24 cycles (for GAPDH) at 94°C for 30 s, 60°C for 1 min, and 72°C for 1 min. Cycle number was optimized for each primer set to ensure that amplifications using template from spinal dorsal horn of NOP^+/+ mice were in the linear amplification range (data not shown). The photograph of electrophoresis of PCR products was analyzed by NIH Image for Macintosh after scanning exposed films.

Statistical Analysis

In the experiment using three types of mice, statistical evaluations were performed using Dunnett's test for multiple comparisons, after one-way analysis of variance. In other experiments, statistical evaluations were performed using Student's t test. The criterion of significance was set at p < 0.05. All results are expressed as the mean ± S.E.M.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Altered Nociceptive Responses upon Various Nociceptive Stimuli in NOP^+/+ Mice. When tail-flick test, a very popular thermal nociception test was adopted, there was no significant change in nociception among NOP^+/+, NOP^+/, and NOP^/ mice (Fig. 1A), being consistent with previous reports (Nishi et al., 1997). In this test, radiant heat stimulus was adjusted for NOP^+/+ mice to show 10- to 11-s latency. Similar degrees of tail-flick latency and nociception were observed in standard mice (data not shown). In the paw pressure test, the average of threshold pressure (in grams) to induce withdrawal response in control NOP^+/+ mice was 10.4 ± 0.7 g (n = 6). As shown in Fig. 1B, there was no significant change in the threshold between NOP^+/+ and NOP^/ mice. In the Hargreaves test, however, the latency for paw withdrawal in NOP^/ mice was significantly lowered (hyperalgesic) to that in NOP^+/+ mice, which show the average latency of 9.2 ± 0.3 s (n = 6; Fig. 1C). Similarly, the capsaicin (0.4 or 0.8 µg)-induced ABL test also showed the hyperalgesia in NOP^/ mice, compared with NOP^+/+ mice (Fig. 1D).

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Altered responses to different modalities of nociceptive stimuli in NOP+/+, NOP+/, or NOP/ mice. Each point of data in all figures was the mean ± S.E.M. from the experiments using at least six mice. Comparison of nociceptive response in various tests among NOP+/+, NOP+/, and NOP/ mice (A and E) or between NOP+/+ and NOP/ mice (B and D, F and I). E and I, results represent the percentage of the maximal reflex, which is the biggest response among spontaneous and nonspecific flexor responses occurring immediately after cannulation. The increasing doses of compound were given every 5 min and the average of two responses by repeated challenges per each dose was evaluated. *, p < 0.05, versus corresponding control. n = 6. BK, bradykinin; SP, substance P; PGI2, prostaglandin I2.

/

+/

/

/

/

Selective Enhancement of Spinal Substance P Responses in NOP^/ Mice. The substance P (i.t.)-induced nocifensive SBL responses, characterized by scratching, biting, and licking to hind paw, were mainly observed at the period of 0 to 5 min after injection. The time period showing the SBL responses during 5 min after the substance P injection was evaluated as the central nociception. In NOP^+/+ mice, marked nocifensive responses were observed with 100 pmol (i.t.) of substance P, and similar results were also obtained in heterozygous NOP^+/ mice (Fig. 2A). In NOP^/ mice, however, markedly enhanced nocifensive responses were observed. The SBL responses by 10 pmol (i.t.) of substance P in NOP^/ mice were equivalent to those by 100 pmol (i.t.) in NOP^+/+ mice.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Enhanced nociceptive responses in NOP/ mice to intrathecally administered substance P, but not NMDA with CP-99994. Results represent SBL responses by the i.t. injection of NMDA or substance P. Each point of data in all figures was the mean ± S.E.M. from the experiments using at least six mice. A, enhanced substance P-induced SBL responses in NOP/ mice. B, enhanced substance P-induced SBL responses in capsaicin-pretreated NOP/ mice. C, NMDA-induced SBL responses in the absence or presence of CP-99994 in NOP+/+, NOP+/, or NOP/ mice. Intrathecal injection of saline was performed to assess a control response. CP-99994 (10 nmol i.t.) was injected 20 min before NMDA injection. Capsaicin (50 mg/kg) or vehicle was injected into the back of newborn (P4) ddY mice. n = 6 to 8. *, p < 0.05, compared with NMDA-treated NOP+/+ mice. #, p < 0.05, compared with saline-treated group in each mouse. §, p < 0.05, compared with NMDA-treated groups in each mouse.

/

/

+/

/

+/

/

Lack of NK1 Receptor Up-Regulation in NOP^/ Mice. The NK1-like immunoreactivity was intensely found in the laminae I of the dorsal horn of spinal cord, but there was no significant change between NOP^/ and NOP^+/+ mice (Fig. 3, A and B). As shown in Fig. 3C, no significant change was also observed in the immunoblot analysis using the dorsal horn region of spinal cord (Fig. 3A). The [³H]substance P binding experiments using dorsal horn membranes revealed that the K_d value of 0.69 ± 0.10 nM and B_max value of 27.79 ± 2.34 pmol/mg protein for NOP^/ mice were quite similar to those for NOP^+/+ mice (K_d value of 0.66 ± 0.05 nM, B_max value of 24.26 ± 1.57 pmol/mg protein), as shown in Fig. 3D. Furthermore, there was also no significant difference in the NK1 gene expression in the dorsal horn by RT-PCR between NOP^/ and NOP^+/+ mice (Fig. 3E).

View larger version (38K): [in this window] [in a new window]   Fig. 3.   No change in NK1-receptor gene expression between NOP+/+ and NOP/ mice. A, schematic representation of the spinal dorsal horn. B, NK1 immunohistochemistry of a coronal section of the spinal cord of NOP+/+ and NOP/minus] mice. Thirty-micrometer sections were used for immunostaining. C, Western blot analysis of NK1 protein levels in the spinal cord of NOP+/+ and NOP/ mice. D, [3H]substance P binding in the spinal cord of NOP+/+ and NOP/ mice. There was no significant change between both preparations. E, quantitative RT-PCR for NK1-receptor gene expression. NK1 gene expressions were analyzed, using GAPDH transcript as a reference. The results represent the relative NK1 receptor expression to GAPDH 1. There was no significant increase in the expression in NOP/ mice, compared with that in NOP+/+ mice. n = 6.

Enhanced Nociception in NOP Antagonist (i.t.)-Treated Mice. The hyperalgesia was also observed in the ANF test with bradykinin and substance P, when 1 pmol of J-113397 (Ozaki et al., 2000; Ueda et al., 2000a) or 1 nmol of [Nphe¹]N/OFQ(1-13)NH₂ (Calo et al., 2000) was i.t. pretreated 20 min before the test (Fig. 4, A and B). The ED₅₀ value for bradykinin and substance P in antagonist-treated mice was 100 to 10,000 times lower than that in vehicle-control mice. However, the i.t. injection of either antagonist alone did not show any gross behavioral changes nor nocifensive responses without stimuli. On the other hand, the SBL responses by i.t. injection of substance P were markedly enhanced by the treatment of these antagonists (Fig. 4C).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Enhancement of nociception driving spinal substance P neurotransmission or substance P-induced nociception by NOP antagonist treatments. A and B, enhanced bradykinin (i.pl.)- or substance P (i.pl.)-induced nociceptive responses in NOP antagonist-treated mice in ANF test. J-113397 (1 pmol) or Nphe1 (1 nmol) was given by i.t. 20 min before several peripheral nociceptive stimulation in the ANF tests. C, enhanced substance P (i.t.)-induced SBL responses in NOP antagonist-treated mice. J-113397 (1 pmol) or Nphe1 (1 nmol) was given by i.t. 20 min before substance P injection. Details are indicated in the legend of Fig. 1. n = 6 to 8.

Characterization of Spinal Transmission in Several Nociception Tests. We tested the spinal antagonism using substance P and glutamate receptor antagonists in various nociception tests. As shown in Table 1, the i.t. injection (3 nmol each) of MK-801 (NMDA receptor antagonists) or -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (CNQX)/kainate receptor antagonist markedly inhibited the tail-flick responses, whereas there was only a little change with CP-99994 (3 nmol). Similar results were obtained with C57/Black mice (data not shown). The paw pressure response in standard mice was blocked by MK-801, but not by CNQX or CP-99994. On the other hand, the Hargreaves thermal nociception was equally and significantly inhibited by i.t. injection with CP-99994 or MK-801, but not by CNQX. Similar results were obtained in capsaicin test. Capsaicin-induced nociception was also equally blocked by CP-99994 or MK-801, but not by CNQX. In the ANF test, lower doses of antagonists (100 pmol i.t.) were used, because the pain intensity used in this test is much weaker than other tests (tail-flick, Hargreaves, paw pressure, and capsaicin tests). Bradykinin- and substance P-induced nociception was blocked by CP-99994, but not by MK-801 and CNQX. On the other hand, adenosine triphosphate- or prostaglandin I₂ agonist-induced nociception was blocked by MK-801, but not by CNQX or CP-99994. All the cases with lack of antagonism (100 pmol i.t.) in ANF tests were reproduced when the antagonist doses were increased to 3 nmol (data not shown).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

It remains to be determined how N/OFQ-ergic neurons play roles in the pain regulation. Several pharmacological analyses revealed that N/OFQ showed both anti- and pronociceptive actions in in vivo studies, whereas it mostly had inhibitory actions in in vitro studies (Meunier, 1997; Borgland et al., 2001; Vaughan et al., 2001). Very small amounts of N/OFQ mRNA are observed in the dorsal root ganglion neurons (Pettersson et al., 2002), and this activity in the spinal cord is reported to originate from intrinsic spinal neurons, rather than primary afferent neurons (Riedl et al., 1996). These findings suggest that the in vivo role of N/OFQ in the spinal cord seems to play a role as an interneuronal transmitter to regulate some modalities of pain.

In the ANF test, we used several algogenics to stimulate distinct nociceptive fibers. This ANF test is more sensitive to produce algogenic-induced nociception, because we observed nociceptive responses in much lower doses compared with that in another test (Kato et al., 2002). From a series of experiments using ANF test (Inoue et al., 1999; Ueda et al., 2000b; Rashid et al., 2003), we have proposed three different types of nociceptive fibers based on the sensitivity to neonatal capsaicin and spinal antagonism. In this diagram, bradykinin and substance P stimulate neonatal capsaicin-sensitive polymodal C (we call it type I)-fibers, which use substance P and NK1 receptor for primary afferent pain transmission in the spinal cord, whereas adenosine triphosphate stimulates the capsaicin-sensitive (we call it type II) fibers, which use glutamate and NMDA receptor for the pain transmission. On the other hand, prostaglandin I₂ agonist stimulates the capsaicin-insensitive (we call it type III) fibers, which use glutamate and NMDA receptor for the pain transmission. These type I and II fibers may be suitable for the substance P-containing, nerve growth factor-sensitive neurons and P2X₃ receptor-expressing, glial-derived neurotrophic factor-sensitive fiber, which are proposed by Snider and McMahon (1998), respectively. As expected, N/OFQ (i.pl.)-induced flexor responses were abolished in NOP^/ mice (Fig. 1E). The nociceptive responses by both bradykinin and substance P, on the other hand, were markedly potentiated in NOP^/ mice (Fig. 1, F and G). On the other hand, there was no significant change in the responses by adenosine triphosphate or prostaglandin I₂ agonist, which do not use the substance P transmission, but glutamate-1 (Fig. 1, F and G; Table 1). These results might suggest the view that substance P-mediated nociception is negatively regulated by spinal N/OFQ-ergic system, rather than glutamate nociception is, as shown in the working hypothesis (Fig. 5). To prove this hypothesis, the peripheral stimulation-selective release of N/OFQ from the spinal cord should be detected as a future subject.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Working hypothesis of in vivo pain-inhibitory role of N/OFQ neuron in the dorsal horn of the spinal cord. The hypothetical diagram is based on the sensitivity to neonatal capsaicin and spinal antagonism. In this diagram, bradykinin and substance P stimulate neonatal capsaicin-sensitive polymodal C (type I)-fibers, which use substance P and NK1 receptor for primary afferent pain transmission in the spinal cord, whereas adenosine triphosphate stimulates the capsaicin-sensitive (type II) fibers, which use glutamate and NMDA receptor for the pain transmission. On the other hand, prostaglandin I2 agonist stimulates the capsaicin-insensitive (type III) fibers, which use glutamate and NMDA receptor for the pain transmission. The pain inhibitory role through NOP or N/OFQ-ergic neuron in the dorsal horn of spinal cord could be attributed to the action on the second-order neuron for type I substance P fibers. Cells indicated with open or closed circle represent stimulatory or inhibitory neurons, respectively. Details are indicated in the text.

In addition, our hypothesis was also supported by the results (Fig. 1, A-D) that the hyperalgesia in NOP^/ mice was observed in the Hargreaves thermal nociception test and capsaicin tests, which drive spinal substance P system in part for pain transmission, but not in the tail-flick and paw pressure test without substance P system (Table 1). The lack of hyperalgesia in the tail-flick test in NOP^/ mice is consistent with the previous report (Nishi et al., 1997). In the previous study (Nishi et al., 1997), NOP^/ mice did not show the hyperalgesia in acid-induced writhing responses, which are sensitive to neonatal capsaicin treatment (Ikeda et al., 2001). However, because the acid-induced writhing responses were not affected in mice lacking the gene encoding tachykinin 1 (Zimmer et al., 1998), spinal substance P transmission is unlikely involved in this test. All these results strongly suggest that the involvement of spinal substance P transmission in the nociception test is closely related to the hyperalgesia in NOP^/ mice.

It should be important how spinal N/OFQ-ergic system acts on the substance P-mediated pain transmission. One of these questions is which presynaptic (primary afferent) nerve terminal or/and postsynaptic (second-order) spinal neuron is the site for N/OFQ, because we have previously reported that N/OFQ given i.t. exerts nocifensive actions in the femtomolar dose range through an substance P release from primary substance P fibers, whereas analgesic actions in the nanomolar dose range through an inhibition of substance P actions on the second-order neuron (Inoue et al., 1998, 1999). These findings raise the question which pain inhibitory or stimulatory responses are observed in NOP^/ mice. However, as far as we have observed in the present study, there is no evidence for the pain inhibitory responses obtained. All the data we obtained using various nociception tests show the hyperalgesia in Hargreaves, capsaicin, and ANF tests using bradykinin and substance P, all of which use spinal substance P transmission, whereas no significant changes were indicated in tail-flick, paw pressure, and ANF tests using adenosine triphosphate and prostaglandin I₂ agonist, which do not use substance P transmission. On the other hand, the nocifensive responses by substance P (i.t.) were also potentiated in NOP^/ mice with or without neonatal capsaicin pretreatment to degenerate C-fibers. All results strongly suggest that there is a discrepancy between pharmacological actions and physiological roles of spinal N/OFQ in the pain regulation.

Another question is whether the genetic deletion of NOP causes some changes in the sensitivity to substance P. As shown in Fig. 3, B to E, the immunoreactivity for NK1 receptor, substance P binding activity, and gene expression at the dorsal horn of spinal cord showed no significant difference between NOP^/ and NOP^+/+mice. Taking into consideration the fact that N/OFQ exerts inhibitory actions through G_i/o mechanisms on various cells in vitro (Meis and Pape, 1998; Zeilhofer et al., 2000), all these findings strongly suggest that the N/OFQ-ergic interneuron plays a role as a recurrent inhibitory interneuron in vivo (Fig. 5).

Here, we demonstrated that intrathecally administrated NMDA caused nocifensive responses through a spinal substance P release (Fig. 2C). This finding raises a question why NMDA (i.t.)-induced nocifensive responses are enhanced in NOP^/ mice, although there is no change in adenosine triphosphate (i.pl.)-induced responses, which are mediated by spinal NMDA receptors. In the previous and present studies, we reported adenosine triphosphate- or its analog (i.pl.)-induced responses were blocked by the intrathecal injection of NMDA receptor antagonist MK-801, but not by substance P receptor antagonist CP-99994 (Ueda et al., 2000b; Table 1). This finding suggests that glutamate released from nociceptive fibers stimulated by adenosine triphosphate (i.pl.) unlikely presynaptically activates the fibers to be stimulated by substance P (i.pl.). Snider and McMahon (1998) supports this view in the review, in which the nociceptive fibers containing substance P (type I) innervate lamina I and II (outer) regions in the dorsal horn, whereas P2X₃ (adenosine triphosphate receptor)-expressing fibers (type II) innervate lamina II (inner) region. Although the in vivo role of presynaptic NMDA receptor on type I nociceptive fibers remains unclear, the algogenic (i.pl.)-induced nociceptive responses through type I fibers unlikely involve this NMDA mechanisms, because they are blocked by substance P antagonist, but not by NMDA antagonist (Ueda et al., 2000b; Table 1).

In the present study, we demonstrated that the postsynaptic supersensitization of substance P (i.t.)-induced nociceptive responses in NOP^/ mice is attributed to the lack of inhibitory N/OFQ-ergic interneurons downstream to substance P-responsive neurons. Here, we used two chemically different NOP antagonists, the nonpeptide J-113397 and the N/OFQ-related peptide [Nphe¹]N/OFQ(1-13)NH₂. Although the potency of the peptide antagonist was about 1000-fold lower than that of J-113397, the two agents produced superimposable results, strongly suggesting that their action is exclusively due to NOP blockade.

In summary, the present study demonstrates that N/OFQ plays an inhibitory role in the pain transmission through polymodal substance P fibers. However, this does not necessarily mean that NOP agonists could behave as potent spinal analgesics, because several potent NK1 receptor antagonists have no significant analgesic actions in clinic (for reviews, see Hill, 2000; Villanueva, 2000). Clinical availability of NOP-selective and -potent ligands should be rather discussed in terms of the potency of chronic pain suppression.