Role of glutamatergic receptors located in the nucleus raphe magnus on antinociceptive effect of morphine microinjected into the nucleus cuneiformis of rat

Available online at www.sciencedirect.com

Neuroscience Letters 427 (2007) 44–49

Role of glutamatergic receptors located in the nucleus raphe magnus on antinociceptive effect of morphine microinjected into the nucleus cuneiformis of rat Abbas Haghparast a,b,∗ , Ava Soltani-Hekmat b , Abbas Khani a , Alireza Komaki c a

b

Neuroscience Research Center, Shaheed Beheshti Medical University, P.O. Box 19615-1178, Tehran, Iran Department of Physiology & Pharmacology, School of Medicine and Neuroscience Research Center, Kerman University of Medical Sciences, Kerman, Iran c Department of Physiology, School of Medicine, Hamedan University of Medical Sciences, Hamedan, Iran Received 21 June 2007; received in revised form 2 September 2007; accepted 4 September 2007

Abstract Neurons in the nucleus cuneiformis (CnF), located just ventrolateral to the periaqueductal gray, project to medullary nucleus raphe magnus (NRM), which is a key medullary relay for descending pain modulation and is critically involved in opioid-induced analgesia. Previous studies have shown that antinociceptive response of CnF-microinjected morphine can be modulated by the specific subtypes of glutamatergic receptors within the CnF. In this study, we evaluated the role of NMDA and kainate/AMPA receptors that are widely distributed within the NRM on morphineinduced antinociception elicited from the CnF. Hundred and five male Wistar rats weighing 250–300 g were used. Morphine (10, 20 and 40 ␮g) and NMDA receptor antagonist, MK-801 (10 ␮g) or kainate/AMPA receptor antagonist, DNQX (0.5 ␮g) in 0.5 ␮l saline were stereotaxically microinjected into the CnF and NRM, respectively. The latency of tail-flick response was measured at set intervals (2, 7, 12, 17, 22, 27 min after microinjection) by using an automated tail-flick analgesiometer. The results showed that morphine microinjection into the CnF dose-dependently causes increase in tail-flick latency (TFL). MK-801 microinjected into the NRM, just 1 min before morphine injection into the CnF, significantly attenuated antinociceptive effects of morphine. On the other hand, DNQX microinjected into the NRM, significantly increased TFL after local application of morphine into the CnF. We suggest that morphine related antinociceptive effect elicited from the CnF is mediated, in part, by NMDA receptor at the level of the NRM whereas kainite/AMPA receptor has a net inhibitory influence at the same pathway. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Nucleus cuneiformis; Nucleus raphe magnus; Morphine; NMDA receptor; Kainate/AMPA receptor; Pain modulation

The midbrain periaqueductal gray (PAG) and adjacent nucleus cuneiformis (CnF) and their major caudal projection target, the rostral ventromedial medulla (RVM) are important components of a descending pain modulatory circuit [2,7,23,35]. These nuclei play a well-established role in the control of nociceptive processing. The CnF plays an important role in sensory/motor integration relevant to pain transmission [35]. Previous studies have shown that morphine when microinjected into the CnF [13], the nucleus raphe magnus (NRM) [8,9] and the PAG [21,22] produce powerful analgesia. The rostral ventromedial medulla is a critical relay for midbrain

∗ Corresponding author at: Neuroscience Research Center, Shaheed Beheshti Medical University, P.O. Box 19615-1178, Tehran, Iran. Tel.: +98 21 22431624; fax: +98 21 22431624. E-mail address: [email protected] (A. Haghparast).

0304-3940/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2007.09.003

regions, including the PAG and CnF that control nociception at the spinal cord [24]. Accordingly, NRM is a key medullary relay for descending pain modulation and is critically involved in opioid-induced analgesia [10] and receives direct connection from mesencephalic tectum neurons in which it was found serotonin-containing neuronal cells [7] involved in noxious stimulus processing [28]. While neuroanatomical [5,34], physiological [3] and pharmacological [12] evidence concerning link between the PAG and NRM has widely been investigated and even detailed connections of projecting neurons from the PAG to on- and off-cells of the RVM have been studied [32], anatomical research focusing on the CnF-NRM link is limited although some anatomical, immunohistochemical and physiological studies have revealed that the CnF provides a major source of afferents to the NRM [4,6,35]. On- and off-cells were found throughout the region from which tail-flick suppression was observed but were most concentrated in the areas of CnF

A. Haghparast et al. / Neuroscience Letters 427 (2007) 44–49

and nucleus parabrachialis (NPB). The presence of on- and offcells in the CnF/NPB region as well as the RVM suggests that the input to the RVM from the CnF/NPB may be important in descending nociceptive modulation [14]. On the other hand, many studies have shown the existence of excitatory amino acid (EAA) receptors in the NRM [1,27,33,34] and their role in morphine-induced antinociception [1,33]. Glutamate-containing neurons projecting to the NRM are originated mainly from the PAG [1,5,34] and the CnF [26]. Several lines of studies that have investigated the involvement of glutamate receptors in the RVM in descending pain modulatory system, attributed the stimulation of these receptors to glutamatergic projecting neurons from the PAG; in these studies morphine microinjected into the different areas of the PAG to elicit analgesia and a variety of glutamate-receptor antagonists (NMDA and non-NMDA) administered into the RVM to reveal if they antagonize the morphine-induced antinociception [15,29]. Regarding above mentioned studies and our laboratory finding [13] that morphine microinjection into the CnF elicits analgesia, this study was undertaken to test whether morphineinduced antinociceptive response elicited from the CnF is mediated by glutamate receptors that are widely distributed within the NRM [1,33,34]. This study investigated the role of microinjection of the non-competitive NMDA antagonist, MK-801 and the kainate/AMPA antagonist, DNQX into the NRM with/without concurrent morphine microinjection into the CnF. Hundred and five male Wistar rats (250–300 g) were housed three per cage at a room controlled temperature (22 ± 2 ◦ C) and maintained on a 12-h light/dark cycle with free access to the standard rat breeding diet and tap water. Following anesthesia with sodium thiopental (45–60 mg/kg) intraperitoneally (ip), two stainless steel guide cannulae (23 gauge) were placed stereotaxically (Stoelting, USA) into the CnF and NRM of rats. All Experiments were done in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 80-23, revised 1996). According to the atlas of Paxinos and Watson [25] the stereotaxic coordinates were 7.5–8.5 mm caudal to bregma, 1.6–1.9 mm lateral to the midline and 5.5–6.5 mm from the skull for the CnF, and 10.7–11.3 mm caudal to bregma, 0.2–0.3 mm lateral to the midline and 8.5–9.5 mm from the skull for the NRM. The guide cannulae, 1 mm above the appropriate nucleus, were affixed to the skull with stainless steel screws and secured to anchor screws with dental acrylic. All animals were allowed 1-week recovery before beginning experimental sessions. All microinfusions were administered in 0.5 ␮l volumes of saline at a rate of 0.1 ␮l/10 s through a stainless steel internal cannula (30 gauge). The cannula was connected to a 1-␮l Hamilton microsyringe by polyethylene tubing (PE-20). The tubing was pre-filled with drug solution. Morphine sulfate (Temad Co., Iran), MK-801 (Sigma), and DNQX (Sigma) were dissolved in physiological saline and then microinjected slowly into either the CnF and/or NRM. The injector was left in situ for 30 s after drug administration and followed by replacement of the occluding stylette. Naloxone hydrochloride (Sigma) was administered

45

(2 mg/kg, ip) following morphine-induced analgesia in tail-flick test. The tail-flick test was used for evaluating the antinociceptive properties of morphine. The latency to withdraw the tail from a feedback-controlled projector lamp focused on the blackened ventral surface of tail was used as a measure of nociceptive responsiveness. The heat was applied in succession to a spot 3, 5 or 7 cm from the caudal tip of tail. The beam intensity (50%) was adjusted so that the average control withdrawal latency was 3–4 s. The tail-flick latency (TFL) more than 10 s was considered as a cut-off point to avoid any tissue damage. TFLs (s) are expressed either as raw data or as percentage of maximal possible effect (%MPE) [13]. Tail-flick trials were initiated at 5-min intervals throughout the protocol. All microinjections into the NRM preceded microinjections into the CnF by 1 min. The effects of MK-801 and DNQX infused into the NRM on TFLs were determined with/without subsequent microinjection of morphine into the CnF. Following determination of baseline TFL, MK-801 (10 ␮g) [30] and DNQX (0.5 ␮g) [29] in 0.5 ␮l saline were infused into the NRM over a period of approximately 1 min with a subsequent microinjection of either physiological saline or morphine (20 ␮g in 0.5 ␮l saline) into the CnF. This dose is a representative of ED50% of antinociceptive effect of morphine elicited from the CnF. Two minutes later, TFLs were measured for the period from 2 to 27 min (2, 7, 12, 17, 22 and 27 min; 6 trials) after second drug microinjection. Naloxone (2 mg/kg, ip) was given for groups in which morphine microinjected into the CnF and the protocol continued for two extra trials (5th and 10th minutes after naloxone injection). In addition, we tested a group of intact rats (without any manipulation) for determining baseline TFL and a group of sham-operated rats in which TFLs were determined after recovery from surgery but without any drug microinjection. The results obtained are expressed as mean ± S.E.M. (standard error of mean). The average of TFLs were compared by student’s unpaired t-test and repeated measures analysis of variance (ANOVA) followed by protected Tukey’s or Dunnett’s test for multiple comparison. Furthermore, data in all groups (intact, sham-operated, saline and experimental groups) were subjected to one-way and/or two-way ANOVA followed by post hoc analysis, as needed. P-values less than 0.05 were considered to be statistically significant. After completion of the experiments, rats were killed using overdose of sodium thiopental and the cannulae placements were examined visually. The brains were removed and placed in a 10% formalin solution for at least 3 days. Injection sites were subsequently examined in coronal sections (50 ␮m) stained with Cresyl violet. Infusion sites were histologically verified and plotted on standardized sections derived from the atlas [25]. The data reported here are only from animals in which cannulae were histologically verified to have been placed in the right nucleus (Fig. 1A and B). Average baseline TF latency in these experiments was 4.05 ± 0.12 s. There were no significant differences in TFLs among between the intact (n = 11) and sham-operated (n = 10) groups and a saline control group (saline delivered into the NRM

46

A. Haghparast et al. / Neuroscience Letters 427 (2007) 44–49

Fig. 1. Histology. A and B, Photomicrographs scan of a coronal section (50 ␮m) showing the guide cannula track and microinjection site (asterisk) in the CnF (top) and NRM (bottom). Microinjection sites locations are summarized on three representative coronal sections for C, the CnF and D, the NRM (open circles, cannula placement controls; filled circles, drug microinjection sites). 7, Facial nucleus; 4n, Trochlear nerve; 4V, 4th Ventricle; Aq, Aqueduct; CnF, Cuneiform nucleus; dmPAG, Dorsomedial periaqueductal gray; Gi, Gigantocellular reticular nucleus; GiA, Gigantocellular reticular nucleus (alpha part); LPAG, Lateral periaqueductal gray; LPGiA, Lateral paragigantocellular nucleus (alpha part); NRM, Nucleus raphe magnus; py, Pyramidal tract; Sp5O, Spinal trigeminal nucleus oralis; vlPAG, Ventrolateral periaqueductal gray; Scale Bar = 500 ␮m.

and CnF in a volume of 0.5 ␮l; n = 10). So, all experimental groups were compared with saline group as a control and TFL results of the saline control group considered as baseline in all 5-min intervals. Additionally, one-way ANOVA revealed that there were not any significant differences in the average of TF latencies among the saline control group and cannula placement controls for the CnF and NRM (Fig. 1C and D) after drug administration [F(4,33) = 1.25, P = 0.31, N.S.]. Although CnF-cannula placement controls showed increase in the average of TFLs (4.43 ± 0.49 s) after solely morphine administration, this was not significant. Microinjections of MK-801 (10 ␮g) or DNQX (0.5 ␮g) in 0.5 ␮l saline adjacent to the NRM (Fig. 1D) were also ineffective. Moreover, the results of two-way ANOVA showed that there were no significant differences in baseline TF latencies among the experimental groups at any time points prior to the drug microinjection into these nuclei. Microinjection of morphine (10, 20 and 40 ␮g; 10 animals/group) into the CnF showed antinociceptive response in a dose-dependent manner. One-way ANOVA followed by Dunnett’s multiple comparisons test showed that the average of TFLs in morphine treated rats was significantly increased in comparison to saline treated group [F(3,39) = 26.48; P < 0.0001]. The average of %MPE was also dose-dependent (29.6 ± 6.3%, 58.9 ± 7.7% and 84.8 ± 5.8% for 10, 20 and 40 ␮g morphine, respectively). However, in order to examine the relationship between EAA receptors located in the NRM and morphineinduced antinociception in the CnF, the dose of 20 ␮g morphine near to ED50% was microinjected into the CnF because of its significant, but not maximal analgesic effect (58.9 ± 7.7%). Microinjection of the NMDA receptor antagonist MK-801 in the NRM along with saline injection into the CnF resulted in sig-

nificant decrease (2.96 ± 0.2 s in comparison with 4.05 ± 0.12 s of saline group; P < 0.05) in the average of TFLs. However, this significant decrease was seen only at 17 and 27 min after morphine microinjection (Fig. 2A). Administration of MK-801 into the NRM 1 min before morphine injection (20 ␮g) into the CnF significantly attenuated the antinociceptive effect of morphine (n = 12; Fig. 2A). Two-way ANOVA followed by Bonferroni post-test revealed that the TF latencies in MK801 + morphine treated group were significantly lower than those in the CnF-microinjected morphine treated group in trials 3 through 6 [treatment effect: F(1,108) = 19.85, P < 0.0001; time effect: F(5,108) = 2.55, P = 0.032; interaction: F(5,108) = 1.699, P = 0.14]. The onset of this effect was 12 min after morphine injection. On the other hand, the two-way ANOVA revealed that the magnitude of difference between TFLs of MK801 + morphine and morphine (alone) groups was significantly higher than the difference between TFLs of MK-801 treated animals and saline group [treatment effect: F(1,100) = 46.86, P < 0.0001; time effect: F(5,100) = 1.883, P = 0.104; interaction: F(5,100) = 0.513, P = 0.766]. In all morphine treated animals in these experiments, naloxone (2 mg/kg; ip) completely blocked morphine-induced antinociception (Fig. 2A). Two-tailed student’s unpaired t-test showed that there was no significant difference in the average of TFLs between NRMmicroinjected saline (4.05 ± 0.12 s) and DNQX (4.81 ± 0.16 s) treated animals [t(18) = 1.33, P = 0.199]. However, there was a significant difference in TFL at the first trial (5.28 ± 0.22; P < 0.05; Fig. 2B). Microinjection of DNQX into the NRM 1 min before morphine injection (20 ␮g) into the CnF significantly potentiated the antinociceptive effect of morphine (n = 12; Fig. 2B) that the average of TF latencies increased

A. Haghparast et al. / Neuroscience Letters 427 (2007) 44–49

Fig. 2. Alterations in tail-flick latencies in rats microinjected with saline, morphine (A) non-competitive NMDA antagonist (MK-801; 10 ␮g/0.5 ␮l saline) or (B) kainate/AMPA antagonist (DNQX; 0.5 ␮g/0.5 ␮l saline), alone and in combination with morphine (20 ␮g/0.5 ␮l saline). Morphine effects were reversed by naloxone (2 mg/kg; ip) in morphine treated rats. Mean TFLs in intact rats (without any manipulation) have been shown as one point (n = 11). Values were expressed as mean ± S.E.M. (n = 7–12 rats at each time point). *P < 0.05; **P < 0.01; ***P < 0.001 compared with saline group. †P < 0.05; ††P < 0.01 compared with morphine treated group.

to 8.9 ± 0.19 s. Two-way ANOVA followed by Bonferroni post-test revealed that the TF latencies in DNQX + morphine treated group were significantly higher than those in the CnF-microinjected morphine treated group [treatment effect: F(1,101) = 23.62, P < 0.0001; time effect: F(5,101) = 0.992, P = 0.427; interaction: F(5,101) = 0.277, P = 0.925]. The onset of this effect was 7 min after morphine injection. On the other hand, the two-way ANOVA revealed that the magnitude of difference between TFLs of DNQX + morphine and morphine (alone)

47

groups was significantly higher than the difference between TFLs of DNQX treated animals and saline group [treatment effect: F(1,92) = 40.2, P < 0.0001; time effect: F(5,92) = 0.678, P = 0.641; interaction: F(5,92) = 0.07, P = 0.996]. The present study found that NRM administration of non-competitive NMDA receptor antagonist, MK-801 but not kainate/AMPA antagonist, DNQX significantly inhibited morphine-induced analgesia elicited from the CnF. This effect was highly selective since MK-801 microinjected into the placement control cannulae failed to alter morphine-induced analgesia following microinjection into the CnF suggesting that MK-801 exerts its inhibitory effect on antinociception largely at the NRM but not through diffusion. In addition, MK-801 microinjection into the NRM without concurrent CnF administration of morphine caused a small decrease in TFLs which can account for the removal of a baseline facilitatory input in modulatory pathway either by the direct blockade of postsynaptic NMDA receptors on the NRM off-cells and/or by the decrease in the net activity of endogenous opioids in this region. Likewise, Spinella et al. [29] showed that MK-801 produced non-significant and small reduction in tail-flick latency. In contrast, NRM administration of DNQX not only did not reduce morphine-induced analgesia, but also somewhat potentiated morphine-induced antinociception. Accordingly, DNQX without concurrent morphine injection into the CnF showed a small and non-significant increase in TFLs. Similarly, Spinella et al. [29] results show that CNQX (kainate/AMPA antagonist) itself produced a small and non-significant increase in TFL. Several lines of studies have shown the involvement of NMDA receptors in descending pain inhibitory system at the level of RVM [16,17,19,26,27,29]. Spinella et al. [29] previously demonstrated that RVM administration of either the non-competitive (MK-801) or competitive (AP7) NMDA antagonist significantly inhibited morphine analgesia elicited from the PAG. Alternatively, CnF-induced activation of the NRM may apparently be mediated, in part, by an excitatory glutamatergic projection [26]. Thus, we antagonized NMDA receptor at the NRM using MK-801 with concurrent CnF-administered morphine to elucidate whether the morphine-induced analgesia is mediated by NMDA receptor at the NRM. This study is in agreement with other findings that supraspinal NMDA receptors may be involved in potentiating the antinociceptive effects of morphine [1,13,16–18,20,29]. PAG-microinjected highly mu-selective agonist Tyr-d-Ala-Gly-MePhe-Gly-ol-enkephalin (DAGO) inhibits spontaneous and noxious-evoked on-cell firing (attenuating the characteristic on-cell burst), and excites spontaneous off-cell firing, preventing the characteristic off-cell pause at doses which suppress the tail flick. Therefore, the presence of on- and off-cells in the CnF/NPB region [14] as well as the similarity between the CnF and PAG at the both ultrastructural [11] and functional [13] level suggest that there are neurons in the CnF that project to the RVM and have excitatory effect on the RVM off-cells. Given these findings and present study, we propose that morphine related antinociceptive effects elicited from the CnF is mediated, in part, by NMDA receptors at the level of NRM. However, it is likely that other neurochemical substrates may have a role in mediating antinociception elicited from the

48

A. Haghparast et al. / Neuroscience Letters 427 (2007) 44–49

CnF since it has been cleared that they are involved in analgesia elicited from the PAG [29]. On the other hand, we showed that NRM administration of the selective kainate/AMPA antagonist, DNQX enhanced morphine analgesia elicited from the CnF. This is in contrast with some previous suggestions regarding the role of the kainate/AMPA receptor in pain modulation. vanPraag and Frenk [33] studies showed that NRM microinjection of excitatory amino acid antagonists 1-(p-chlorobenzoyl)-piperazine-2,3-dicarboxylate (PCB, 3.25 ␮mol) or DL-2-amino-5-phosphono-valerate (APV, 25.38 ␮mol) significantly reduced PAG morphine analgesia. Spinella et al. [29] explained these findings: since PCB was more potent on molar basis than APV in exerting these effects, and since PCB, but not APV blocked RVM glutamate analgesia, it appeared that the kainate/quisqualate receptor subtype was primarily responsible for EAA modulation of analgesic responses in the RVM. They came to a different conclusion based upon the potent effectiveness of either AP7 or MK-801 in the RVM to block PAG morphine analgesia as compared to RVM microinjections of CNQX [29]. Nevertheless, there is a considerable difference between their studies and ours to take into account that is, they both studied the mesencephalic analgesia elicited from the PAG whereas we dealt with the mesencephalic antinociception elicited from the CnF. Also, this is consistent with previous finding of our laboratory [13] where CnF microinjection of DNQX increased the TFLs resulted from the CnF-microinjected morphine analgesia. Furthermore, a slight increase in baseline TFLs by DNQX without concurrent morphine administration suggests that kainate/AMPA subtype of glutamate receptor is physiologically involved in pain perception either by facilitatory influence on nociception or inhibitory influence on pain modulation. Consistently, Suh et al. [31] suggested that the blockade of kainate/AMPA receptors located in the spinal cord appears to be involved in enhancing the inhibition of tail-flick response induced by stimulation of spinal mu-, delta-, and kappa-opioid receptors. The same suggestion could well apply to our findings in the NRM and interaction between morphine and DNQX. However, further behavioral and electrophysiological investigations are needed to elucidate the actual role of non-NMDA receptor. These results demands designing elegant experiments to further investigate on the complexity of circuitry among brainstem structures with established role in the top–down pain modulation.

References [1] L.D. Aimone, G.F. Gebhart, Stimulation-produced spinal inhibition from the midbrain in the rat is mediated by an excitatory amino acid neurotransmitter in the medial medulla, J. Neurosci. 6 (1986) 1803–1813. [2] A.I. Basbaum, H.L. Fields, Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry, Ann. Rev. Neurosci. 7 (1984) 309–338. [3] M.M. Behbehani, H.L. Fields, Evidence that an excitatory connection between the periaqueductal gray and nucleus raphe magnus mediates stimulation produced analgesia, Brain Res. 170 (1979) 85–93. [4] M.M. Behbehani, F.P. Zemlan, Response of the nucleus raphe magnus neurons to electrical stimulation of nucleus cuneiformis: role of acetylcholine, Brain Res. 369 (1986) 110–118.

[5] A.J. Beitz, Relationship of glutamate and aspartate to the periaqueductal gray-raphe magnus projection: analysis using immunocytochemistry and microdialysis, J. Histochem. Cytochem. 38 (1990) 1755– 1765. [6] A.J. Beitz, The nuclei of origin of brainstem enkephalin and substance P projections to the rodent nucleus raphe magnus, Neuroscience 7 (1982) 2753–2768. [7] N.C. Coimbra, R. De Oliveira, R.L. Freitas, S.J. Ribeiro, K.G. Borelli, R.C. Pacagnella, J.E. Moreira, L.A. da Silva, L.L. Melo, L.O. Lunardi, M.L. Brand˜ao, Neuroanatomical approaches of the tectum-reticular pathways and immunohistochemical evidence for serotonin-positive perikarya on neuronal substrates of the superior colliculus and periaqueductal gray matter involved in the elaboration of the defensive behavior and fear-induced analgesia, Exp. Neurol. 197 (2006) 93–112. [8] A.H. Dickenson, J.L. Oliveras, J.M. Besson, Role of the nucleus raphe magnus in opiate analgesia as studied by the microinjection technique in the rat, Brain Res. 170 (1979) 95–111. [9] H.J. Du, L.M. Kitahata, J.G. Thalhammer, M. Zimmermann, Inhibition of nociceptive neuronal responses in the cat’s spinal dorsal horn by electrical stimulation and morphine microinjection in nucleus raphe magnus, Pain 19 (1984) 249–257. [10] H. Fields, State-dependent opioid control of pain, Nat. Rev. Neurosci. 5 (2004) 565–575. [11] M. Gioia, R. Bianchi, Ultrastructural study of the nucleus cuneiformis in the cat, J. Hirnforsch. 28 (1987) 375–383. [12] A.P. Guimaraes, W.A. Prado, Pharmacological evidence for a periaqueductal gray-nucleus raphe magnus connection mediating the antinociception induced by microinjecting carbachol into the dorsal periaqueductal gray of rats, Brain Res. 827 (1999) 152–159. [13] A. Haghparast, I.P. Gheitasi, R. Lashgari, Involvement of glutamatergic receptors in the nucleus cuneiformis in modulating morphine-induced antinociception in rats, Eur. J. Pain 11 (2007) 855–862. [14] C.M. Haws, A.M. Williamson, H.L. Fields, Putative nociceptive modulatory neurons in the dorsolateral pontomesencephalic reticular formation, Brain Res. 483 (1989) 272–282. [15] E. Hopkins, M. Spinella, Z.W. Pavlovic, R.J. Bodnar, Alterations in swim stress-induced analgesia and hypothermia following serotonergic or NMDA antagonists in the rostral ventromedial medulla of rats, Physiol. Behav. 64 (1998) 219–225. [16] K. Javanmardi, M. Parviz, S.S. Sadr, M. Keshavarz, B. Minaii, A.R. Dehpour, Involvement of N-methyl-d-aspartate receptors and nitric oxide in the rostral ventromedial medulla in modulating morphine pain-inhibitory signals from the periaqueductal grey matter in rats, Clin. Exp. Pharmacol. Physiol. 32 (2005) 585–589. [17] S.L. Jones, G.F. Gebhart, Inhibition of spinal nociceptive transmission from the midbrain, pons and medulla in the rat: activation of descending inhibition by morphine, glutamate and electrical stimulation, Brain Res. 460 (1988) 281–296. [18] E. Kozela, W. Danysz, P. Popik, Uncompetitive NMDA receptor antagonists pontentiate morphine antinociception recorded from the tail not from hind paw in rats, Eur. J. Pharmacol. 423 (2001) 17–26. [19] E. Kozela, P. Popik, A complete analysis of NMDA receptor subunits in periaqueductal grey and ventromedial medulla of morphine tolerant mice, Drug Alcohol Depend. 86 (2007) 290– 293. [20] E. Kozela, P. Popik, The effects of NMDA receptor antagonists on acute morphine antinociception in mice, Amino Acids 23 (2002) 163– 168. [21] J.S. Mohrland, G.F. Gebhart, Effects of focal electrical stimulation and morphine microinjection in the periaqueductal gray of the rat mesencephalon on neuronal activity in the medullary reticular formation, Brain Res. 201 (1980) 23–27. [22] M.M. Morgan, B.W. Tierney, S.L. Ingram, Intermittent dosing prolongs tolerance to the antinociceptive effect of morphine microinjection into the periaqueductal gray, Brain Res. 1059 (2005) 173– 178. [23] M.S. Moss, A.I. Basbaum, The peptidergic organization of the cat periaqueductal gray. II. The distribution of immunoreactive substance P and vasoactive intestinal polypeptide, J. Neurosci. 3 (1983) 1437–1449.

A. Haghparast et al. / Neuroscience Letters 427 (2007) 44–49 [24] Z.Z. Pan, H.L. Fields, Endogenous opioid-mediated inhibition of putative pain-modulating neurons in rat rostral ventromedial medulla, Neuroscience 74 (1996) 855–862. [25] G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, 5th ed., Elsevier Academic Press, San Diego, 2005, pp. 98–106, 113–34. [26] R.C. Richter, M.M. Behbehani, Evidence for glutamic acid as a possible neurotransmitter between the mesencephalic nucleus cuneiformis and the medullary nucleus raphe magnus in the lightly anesthetized rat, Brain Res. 544 (1991) 279–286. [27] M. Satoh, R. Oku, A. Akaike, Analgesia produced by microinjection of L-glutamate into the rostral ventromedial bulbar nuclei of the rat and its inhibition by intrathecal alpha-adrenergic blocking agents, Brain Res. 261 (1983) 361–364. [28] S. Sounvoravong, M.N. Nakashima, M. Wada, K. Nakashima, Decrease in serotonin concentration in raphe magnus nucleus and attenuation of morphine analgesia in two mice models of neuropathic pain, Eur. J. Pharmacol. 484 (2004) 217–223. [29] M. Spinella, M.L. Cooper, R.J. Bodnar, Excitatory amino acid antagonists in the rostral ventromedial medulla inhibit mesencephalic morphine analgesia in rats, Pain 64 (1996) 545–552.

49

[30] M. Spinella, V. Znamensky, M. Moroz, A. Ragnauth, R.J. Bodnar, Actions of NMDA and cholinergic receptor antagonists in the rostral ventromedial medulla upon b-endorphin analgesia elicited from the ventrolateral periaqueductal gray, Brain Res. 829 (1999) 151–159. [31] H.W. Suh, D.K. Song, S.O. Huh, Y.H. Kim, Differential potentiative effects of glutamate receptor antagonists in the production of antinociception induced by opioids administered intrathecally in the mouse, Brain Res. Bull. 52 (2000) 43–50. [32] V. Tortorici, M.M. Morgan, Comparison of morphine and kainic acid microinjections into identical PAG sites on the activity of RVM neurons, J. Neurophysiol. 88 (2002) 1707–1715. [33] H. vanPraag, H. Frenk, The role of glutamate in opiate descending inhibition of nociceptive spinal reflexes, Brain Res. 524 (1990) 101–105. [34] L. Wiklund, G. Behzadi, P. Kalen, P.M. Headley, L.S. Nicolopoulos, C.G. Parsons, D.C. West, Autoradiographic and electrophysiological evidence for excitatory amino acid transmission in the periaqueductal gray projection to nucleus raphe magnus in the rat, Neurosci. Lett. 93 (1988) 158–163. [35] F.P. Zemlan, M.M. Behbehani, Nucleus cuneiformis and pain modulation: anatomy and behavioral pharmacology, Brain Res. 453 (1988) 89–102.