- Email: [email protected]
The hypotension evoked by visceral nociception is mediated by delta opioid receptors in the periaqueductal gray
Brain Research 1019 (2004) 237 – 245 www.elsevier.com/locate/brainres
Research report
The hypotension evoked by visceral nociception is mediated by delta opioid receptors in the periaqueductal gray Sinan Cavun a,b, Gokhan Goktalay a,b, William R. Millington a,* a
Department of Basic and Pharmaceutical Sciences, Albany College of Pharmacy, Union University, 106 New Scotland Ave, Albany, NY 12208-3492, USA b Department of Pharmacology and Clinical Pharmacology, Uludag University School of Medicine, Bursa, Turkey Accepted 4 June 2004 Available online 10 July 2004
Abstract This study tested the hypothesis that the ventrolateral column of the midbrain periaqueductal gray (vlPAG) region mediates the hypotension and bradycardia evoked by visceral nociception. To test this, the local anesthetic lidocaine (2%; 0.5 Al) was microinjected into the vlPAG of halothane-anesthetized rats bilaterally and visceral nociception was induced 2 min later by injecting 5% acetic acid (0.5 ml) intraperitoneally. Acetic acid injection caused an abrupt fall in arterial pressure ( 12.2 F 2.1 mm Hg) and heart rate ( 37 F 93 bpm) lasting approximately 15 min. Lidocaine injection into the vlPAG prevented the fall in arterial pressure and heart rate completely. Cobalt chloride (5 mM; 0.2 or 0.5 Al) injection into the vlPAG also prevented nociceptive hypotension but it did not affect the fall in heart rate significantly. Lidocaine pretreatment also inhibited the depressor response caused by intramuscular formalin (5%; 0.2 ml) administration, a model of deep somatic nociception, although it did not prevent the response completely. To determine if opioid receptors mediate the response, selective mu, delta or kappa opioid receptor antagonists were microinjected into the vlPAG 5 min before intraperitoneal (ip) acetic acid administration. Naltrindole, a delta receptor antagonist, inhibited the response significantly but mu and kappa antagonists were completely ineffective. Lidocaine and naltrindole had no effect when injected into the dorsolateral PAG and did not influence cardiovascular function when injected into the vlPAG of saline treated control animals. These data support the hypothesis that the vlPAG mediates the depressor response evoked by visceral nociception and indicate that delta opioid receptors participate in the response. D 2004 Elsevier B.V. All rights reserved. Theme: Endocrine and autonomic regulation Topic: Cardiovascular regulation Keywords: Visceral nociception; Pain; Central cardiovascular regulation; Periaqueductal gray; Opioid receptor; Delta receptor
1. Introduction Pain produces quite different autonomic and behavioral reactions depending upon its location. Visceral and deep somatic pain cause arterial pressure and heart rate to fall and produce a behavioral response characterized by quiescence, immobility and inattention to environmental stimuli [3,22,29,31]. Cutaneous and superficial somatic pain generally produce a completely different reaction, characterized by hypertension, tachycardia and behavioral activation, * Corresponding author. Tel.: +1-518-445-7242; fax: +1-518-4457202. E-mail address: [email protected] (W.R. Millington). 0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.06.003
the classic fight or flight response [3,49,55]. The central mechanisms that coordinate the autonomic and behavioral reactions to visceral and somatic pain are not fully understood, and particularly so for visceral pain. The midbrain periaqueductal gray (PAG) region is thought to play a pivotal role in the coordinate regulation of cardiovascular function and behavior [3,6,29]. The PAG is a highly organized structure that is segregated into functionally distinct longitudinal columns [3]. Excitatory amino acid injection into the ventrolateral column of the PAG (vlPAG) evokes a prompt depressor response in anesthetized rats [9,24,29] and decerebrate cats [6,7] and causes quiescence, inattention to the environment and reduced social interaction in conscious animals, a response
238
S. Cavun et al. / Brain Research 1019 (2004) 237–245
much like that caused by visceral pain [17,23,37]. In contrast, excitatory amino acids evoke a pressor response and either behavioral activation or reactive immobility when injected into the lateral (lPAG) or dorsolateral PAG (dlPAG) [3,29], similar to the effects of somatic pain. These findings suggest that the vlPAG mediates the depressor effect of visceral pain whereas the lPAG and dlPAG contribute to the effects of somatic pain. The hypothesis that PAG neurons participate in the cardiovascular response to visceral and somatic pain finds additional support in studies using the immediate early gene c-fos as a marker of neuronal activation. Intraperitoneal (ip) administration of noxious chemicals to anesthetized rats, a standard method for provoking inescapable visceral nociception [39], induces c-fos expression in the vlPAG but not, to an equivalent extent, in other PAG subregions [11,23,42,46]. By contrast, cutaneous nociception induces c-fos expression primarily in the lPAG and dlPAG [21,42]. These findings are consistent with the idea that visceral pain lowers arterial pressure by activating vlPAG neurons although they by no means prove such a connection exists. Visceral pain may very well induce cfos in vlPAG neurons that influence pain perception, motor activity or other complex behaviors rather than cardiovascular regulation, for example [3]. Hence, c-fos expression does not provide direct evidence that visceral nociception lowers arterial pressure by activating vlPAG neurons. The first objective of this study was to test the hypothesis that the vlPAG mediates the depressor response to visceral nociception by determining if the local anesthetic lidocaine or the synaptic inhibitor cobalt chloride would prevent the response [32]. We found that bilateral injection of lidocaine or cobalt chloride into the vlPAG of halothane-anesthetized rats inhibited the fall in arterial pressure and heart rate caused by visceral nociception significantly, consistent with the hypothesis that the vlPAG plays a critical role in the response. The second objective was to determine if opioid receptors in the vlPAG contribute to the cardiovascular effects of visceral nociception. Opioid receptors have been localized in the PAG and their role in pain perception has been studied extensively [54] but there is also evidence that they influence cardiovascular function. Microinjection of a selective delta opioid receptor agonist [24] or the endogenous opioid peptide h-endorphin [9] into the vlPAG lowers arterial pressure and heart rate in halothane-anesthetized rats. Mu receptor activation, on the other hand, evokes a pressor response and kappa receptor agonist injection produces an equivocal effect [24]. Here we report that blockade of delta receptors in the vlPAG effectively inhibits the depressor effect of visceral nociception. These data support the hypothesis that the vlPAG plays a critical role in the cardiovascular effects of visceral pain and indicate that delta receptors participate in the response.
2. Materials and methods 2.1. Animals and surgery One hundred five male Sprague – Dawley rats (250 – 350 g; Charles River Laboratories, Wilmington, MA) were housed at a constant temperature (21 F 1 jC) under a 12 h light/dark schedule. Food and water were freely available. The rats were anesthetized with 4% halothane and maintained with 1.5% halothane in 100% O2. The left carotid artery was cannulated with PE-50 tubing filled with heparinized saline (100 U/ml), exteriorized at the nape of the neck and sealed until use [8]. At the beginning of each experiment, the arterial catheter was attached to a volumetric pressure transducer and arterial pressure and heart rate were monitored continuously and recorded at 1 min intervals using a MicroMed BPA-200 blood pressure analyzer (Micro-Med, Louisville, KY). The animal protocols were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee. 2.2. Lidocaine, cobalt chloride and opioid receptor antagonist injections Two 26 gauge guide cannulae were implanted bilaterally above the caudal vlPAG or dlPAG at a 27j rostro-caudal angle through burr holes drilled through the skull [8]. For caudal vlPAG injections, the tips of the cannulae were positioned 0.8 mm lateral and 8.3 mm posterior to bregma and 6.5 mm below the skull surface according to the atlas of Paxinos and Watson [44]. For caudal dlPAG injections, the guide cannulae were implanted bilaterally 0.8 mm lateral and 8.3 mm posterior to bregma and 4.6 mm below the skull surface. After obtaining stable baseline arterial pressure and heart rate measurements, two 33 gauge injection cannulae were inserted through the guide cannulae to a depth of 1.0 mm below the tip of the guide cannula and lidocaine (2%), cobalt chloride (5 mM), naltrindole HCl (1 nmol/cannula), D-Phe-Cys-Tyr-D-Trp-Orn-Thr-Pen-Thramide (CTOP; 5 nmol/cannula), nor-binaltorphimine dihydrochloride (nor-BNI; 3 nmol/cannula; Sigma, St. Louis, MO) or saline was injected bilaterally. The lidocaine and cobalt chloride doses were based on Malpeli [32] and Cavun and Millington [8], the naltrindole dose was selected from previous dose – response studies [9] and the CTOP and nor-BNI doses were based on earlier reports [18,33,34,53]. Saline-treated controls were interspersed with drug treated animals to ensure against the possibility that different groups of animals or inadvertent changes in technique might influence the results. Drugs were dissolved and injected in 0.2 or 0.5 Al saline (pH 7.2– 7.4) and were delivered at a constant rate over a 1-min period. The injection volume was monitored by observing the movement of an air bubble placed in the tubing.
S. Cavun et al. / Brain Research 1019 (2004) 237–245
239
2.3. Noxious stimuli Two or five minutes after PAG drug administration, animals received a single ip injection (0.5 ml) of 5% acetic acid to induce visceral nociception [11,12,23]. Deep somatic nociception was caused by injecting 5% formalin (0.2 ml; Sigma Chemical Co.; contains 10 –15% methanol) into the gastrocnemius muscle bilaterally [11,26]. Blood pressure and heart rate were recorded for 15 min after ip acetic acid injection or 40 min after im formalin administration. Each rat was used for only one experiment. 2.4. Histology At the end of each experiment, the cannulae placements were verified by injecting 0.2 or 0.5 Al India ink. The animal was then sacrificed, the brain was removed, immersed in 10% paraformaldehyde, imbedded with paraffin, sectioned (50 Am) with a sliding microtome and stained with eosin. 2.5. Statistical analysis Results are expressed as mean F S.E.M. Data were analyzed by repeated measures two-way analysis of variance (ANOVA) with repeated measures followed by Dunnett’s multiple comparison test using SigmaStat version 3.0 (SPSS, Chicago, IL). The criterion for statistical significance was p < 0.05.
3. Results 3.1. Visceral nociception To test the hypothesis that the vlPAG mediates the depressor effect of visceral nociception, we inhibited neuronal activity in the vlPAG with lidocaine before inducing visceral nociception with ip acetic acid. As shown in Fig. 1, ip injection of 5% acetic acid (0.5 ml) evoked a prompt fall in arterial pressure ( 12.3 F 2.1 mm Hg) and heart rate ( 27 F 9 bpm), consistent with data published previously [11]. The depressor response was transient and both arterial pressure and heart rate returned to baseline values within 15 min (Fig. 1). Bilateral lidocaine injection (0.5 Al of a 2% solution) into the caudal vlPAG prevented the hypotension [ F(15,180) = 27.2, P < 0.001] and bradycardia [ F(15,180) = 10.24, P < 0.001] evoked by ip acetic acid completely (Fig. 1). Lidocaine did not affect arterial pressure or heart rate when microinjected into the caudal vlPAG of control animals that received ip saline in lieu of acetic acid (Fig. 1). To verify the specificity of the vlPAG injections, we tested whether lidocaine administration into the dlPAG would influence the depressor effect of ip acetic acid
Fig. 1. Lidocaine injection into the vlPAG inhibits the fall in arterial pressure and heart rate caused by visceral nociception. Lidocaine (2%; 0.5 Al) or saline was injected bilaterally into the caudal vlPAG of halothaneanesthetized rats and, 2 min later, acetic acid (5%, 0.5 ml) was administered ip. Mean arterial pressure (MAP; top panel) and heart rate (bottom panel) were recorded at 1 min intervals for 15 min. Data are presented as mean F S.E.M. change in mean arterial pressure or heart rate from baseline values. The numbers in parentheses indicate the number of animals in each group. Baseline mean arterial pressure at the 2 min time point was 99.5 F 2.4 mm Hg for the saline + acetic acid group, 99.4 F 1.9 mm Hg for the lidocaine + acetic acid group and 98.7 F 1.9 mm Hg for the lidocaine + saline controls. Baseline heart rate was 346 F 13 bpm for the saline + acetic acid group, 348 F 9 bpm for the lidocaine + acetic acid group and 360 F 24 bpm for the lidocaine + saline controls. *P < 0.05, **P < 0.01 compared to the corresponding time point for saline-treated control animals.
injection. Previous investigations showed that excitatory amino acid injection into the dlPAG or lPAG produces hypertension and tachycardia [3,29], which indicates that the dlPAG is unlikely to participate in the depressor response evoked by noxious stimuli. As expected, lidocaine injection into the caudal dlPAG bilaterally failed to prevent the fall in arterial pressure or heart rate evoked by subsequent administration of 5% acetic acid by ip injection (Fig. 2). The depressor effect of visceral nociception is thus prevented by neuronal inactivation in the vlPAG, but not the dlPAG.
240
S. Cavun et al. / Brain Research 1019 (2004) 237–245
imately 15 min, and the total duration of the response was longer, at least 40 min, than that of visceral nociception, however. Furthermore, although arterial pressure returned toward baseline values toward the end of the 40 min experimental period, heart rate remained depressed (Fig. 3). Lidocaine pretreatment attenuated the fall in arterial pressure caused by im formalin injection significantly [ F(20,240) = 2.6, P < 0.001] although it did not prevent the response completely (Fig. 3). Heart rate was not influenced significantly by lidocaine pretreatment. 3.3. Cobalt chloride injection To determine whether synaptic activity is required for the response, we tested whether the vasodepression
Fig. 2. Lidocaine injection into the dlPAG does not prevent the hypotension and bradycardia evoked by ip acetic acid administration. Lidocaine (2%, 0.5 Al) or saline was injected bilaterally into the caudal dlPAG of halothane-anesthetized rats 2 min before injection of 5% acetic acid (0.5 ml, ip) and mean arterial pressure (MAP; top panel) and heart rate (bottom panel) were recorded for 15 min. Baseline mean arterial pressure was 102.4 F 2.1 mm Hg for the saline + acetic acid group, 101.3 F 3.8 mm Hg for the lidocaine + acetic acid group and 97.6 F 1.1 mm Hg for the lidocaine + saline group. Baseline heart rate was 336 F 11 bpm for the saline + acetic acid group, 358 F 20 bpm for the lidocaine + acetic acid group and 344 F 17 bpm for the lidocaine + saline group.
3.2. Deep somatic nociception To investigate whether the vlPAG mediates the depressor effect of deep somatic nociception, we tested whether lidocaine microinjection into the caudal vlPAG prevents the hypotension and bradycardia evoked by im formalin administration to halothane-anesthetized rats [11]. Formalin (5%; 0.2 ml) was injected into both gastrocnemius muscles 2 min after lidocaine (2%; 0.5 Al) or saline was microinjected into the vlPAG bilaterally. In control animals, 5% formalin injection lowered arterial pressure 14.8 F 2.2 mm Hg and reduced heart rate 68 F 5 bpm (Fig. 3), comparable to the hypotension and bradycardia caused by ip acetic acid injection. The maximal response took longer to develop, approx-
Fig. 3. Lidocaine injection into the vlPAG attenuates the fall in arterial pressure and heart rate induced by intramuscular formalin injection. Lidocaine (2%, 0.5 Al) or saline was injected bilaterally into the caudal vlPAG and, after a 2-min delay, formalin (0.2 ml) was injected into both gastrocnemius muscles of halothane-anesthetized rats. Baseline mean arterial pressure at the 2 min time point was 99.3 F 2.2 mm Hg for the saline + formalin group and 100.7 F 0.9 mm Hg for the lidocaine + formalin group. Baseline heart rate was 382 F 7 bpm for the saline + formalin group and 356 F 8 bpm for the lidocaine + formalin group. *P < 0.05 compared to saline-treated control animals.
S. Cavun et al. / Brain Research 1019 (2004) 237–245
241
caine, cobalt chloride injection into the vlPAG inhibits the hypotension evoked by ip lidocaine without influencing arterial pressure or heart rate in normotensive animals. 3.4. Opioid receptor antagonists To test whether delta opioid receptors in the vlPAG contribute to the depressor effect of visceral nociception, we injected naltrindole (1 nmol/cannula) bilaterally into the caudal vlPAG 5 min before 5% acetic acid was administered ip to halothane-anesthetized rats. Naltrindole pretreatment inhibited the fall in arterial pressure
Fig. 4. Cobalt chloride injection into the vlPAG prevents the fall in arterial pressure and heart rate evoked by ip acetic acid. Cobalt chloride (5 mM; 0.2 or 0.5 Al) or saline was injected bilaterally into the vlPAG 5 min before acetic acid (5%, 0.5 ml) was injected ip. Baseline mean arterial pressure was 97.0 F 3.8 mm Hg for the saline + acetic acid group, 101.2 F 4.3 mm Hg for the 0.2 Al cobalt chloride + acetic acid group and 94.3 F 2.3 mm Hg for the 0.5 Al cobalt chloride + acetic acid group. Baseline heart rate was 354 F 8 bpm for the saline + acetic acid group, 343 F 8 bpm for the 0.2 Al cobalt chloride + acetic acid group and 356 F 13 bpm for the 0.5 Al cobalt chloride + acetic acid group. *P < 0.05, **P < 0.01 compared to salinetreated control animals.
caused by ip acetic acid is attenuated by cobalt chloride, which inhibits synaptic transmission but not axonal conductance [24]. Cobalt chloride (5 mM; 0.2 or 0.5 Al) was injected into the vlPAG of halothane-anesthetized rats bilaterally followed, 5 min later, by ip acetic acid. Cobalt chloride pretreatment inhibited the fall in arterial pressure caused by ip acetic acid administration significantly [ F(30,240) = 3.8, P < 0.001] although it did not affect the fall in heart rate significantly (Fig. 4). Interestingly, 0.2 Al injections of 5 mM cobalt chloride inhibited nociceptive hypotension to approximately the same extent as 0.5 Al (Fig. 4). Previously, we showed that cobalt chloride (5 mM; 0.5 Al) injection into the vlPAG of normotensive control animals had no effect on arterial pressure or heart rate [8]. Hence, like lido-
Fig. 5. Naltrindole injection into the caudal vlPAG inhibits the hypotension and bradycardia caused by visceral nociception. The delta opioid receptor antagonist naltrindole (1 nmol/cannula) or saline (0.5 Al) was injected bilaterally into the caudal vlPAG and, after a 5-min delay, acetic acid (5%, 0.5 ml) was administered ip to halothane-anesthetized rats. Data are presented as the mean F S.E.M. change in mean arterial pressure (top panel) and heart rate (bottom panel) from baseline values. Baseline mean arterial pressure at the 5 min time point was 104.0 F 2.1 mm Hg for the saline + acetic acid group, 100.6 F 1.6 mm Hg for the naltrindole + acetic acid group and 102.9 F 1.5 mm Hg for naltrindole + saline controls. Baseline heart rate was 357 F 8 bpm for the saline + acetic acid group, 326 F 4 bpm for the naltrindole + acetic acid group and 342 F 7 bpm for the naltrindole + saline group. *P < 0.05, **P < 0.01 compared to the corresponding time point for saline-treated controls.
242
S. Cavun et al. / Brain Research 1019 (2004) 237–245
prevent the hypotension or bradycardia caused by ip acetic acid. Fig. 6 shows that CTOP and nor-BNI failed to significantly attenuate the fall in arterial pressure or heart rate ip acetic acid causes. Together, these data support the conclusion that activation of delta, but not mu or kappa, receptors in the vlPAG contributes to the depressor effect of visceral nociception. 3.5. Location of injection sites Histological analysis confirmed that lidocaine, cobalt chloride, naltrindole, CTOP and nor-BNI injections were located in or immediately adjacent to the vlPAG or dlPAG (Fig. 7A). Some injection sites were centered outside, but immediately adjacent to, the vlPAG. Likewise, some injections targeted on the dlPAG were centered outside, but immediately adjacent to, the dlPAG or lPAG. Fig. 7B illustrates the area of the midbrain stained by 0.5 Al dye that identifies the site of a lidocaine injection targeted on the vlPAG. The injection covered an area approximately 0.7 mm in diameter in both the rostrocaudal and dorsoventral planes and spread outside the boundaries of the vlPAG ventrolaterally into the adjacent
Fig. 6. CTOP and nor-BNI fail to prevent the fall in arterial pressure (top panel) and heart rate (bottom panel) caused by visceral nociception. The mu receptor antagonist CTOP (5 nmol/cannula), the kappa receptor antagonist nor-BNI (3 nmol/cannula) or saline (0.5 Al) was injected bilaterally into the caudal vlPAG of halothane-anesthetized rats; after a 5min delay, acetic acid (5%, 0.5 ml) was injected ip. Baseline mean arterial pressure at the 5 min time point was 99.0 F 3.3 mm Hg for the saline + acetic acid group, 100.7 F 3.4 mm Hg for the CTOP + acetic acid group and 99.1 F 2.5 mm Hg for the nor-BNI + acetic acid group. Baseline heart rate was 370 F 6 bpm for the saline + acetic acid group, 376 F 15 bpm for the CTOP + acetic acid group and 359 F 8 bpm for the nor-BNI + acetic acid group.
[ F(30,270) = 4.9, P < 0.001] and heart rate [ F(30,225) = 5.9, P < 0.001] significantly (Fig. 5) although it did not influence arterial pressure or heart rate when injected into the vlPAG of control rats that were given saline instead of 5% acetic acid ip (Fig. 5). As shown previously for lidocaine, bilateral naltrindole (2 nmol; 0.5 Al) injections targeted on the dlPAG failed to influence arterial pressure (saline = 13.8 F 1.9 mm Hg; naltrindole = 12.9 F 2.6 mm Hg) or heart rate (saline = 26 F 6 bpm; naltrindole = 25 F 2 bpm) 2 min after ip acetic acid injection. To verify the receptor specificity of the response, we tested whether bilateral injection of the mu receptor antagonist CTOP (5 nmol/cannula) or the kappa receptor antagonist nor-BNI (3 nmol/cannula) into the vlPAG would
Fig. 7. Location of injection sites in the PAG. (Panel A) Schematic representation of coronal sections through the caudal PAG at 8.0 mm (left), 8.3 mm (center) and 8.7 mm (right) from bregma illustrating the location of lidocaine, cobalt chloride, naltrindole, CTOP and nor-BNI injection sites. (Panel B) Photomicrograph of a section through the caudal PAG (8.3 mm from bregma) illustrating the location of 0.5 Al India ink marking the site of a lidocaine injection. The boundaries of the PAG are from Paxinos and Watson [44]. Scale bar = 1.0 mm. (Panel C) Schematic representation of a 0.5 Al India ink injection illustrated in panel B. Aq = Sylvian aqueduct.
S. Cavun et al. / Brain Research 1019 (2004) 237–245
tegmentum (Fig. 7B and C). A smaller injection volume, 0.2 Al, perfused a proportionately smaller area, approximately 0.4 mm in diameter (data not shown).
4. Discussion These data show that lidocaine injection into the vlPAG prevents the hypotension and bradycardia evoked by visceral nociception in halothane-anesthetized rats. Nociceptive hypotension was also blocked by cobalt chloride, which indicates that synaptic transmission within the vlPAG is necessary for visceral nociception to evoke a depressor response. Lidocaine inactivation of the dlPAG was ineffective, consistent with evidence that visceral nociception induces c-fos expression in the vlPAG to a substantially greater extent than in the dlPAG [11]. Ventrolateral PAG inactivation also attenuated the hypotension caused by deep somatic nociception although it did not eliminate the response completely. Nevertheless, the vlPAG apparently plays a significant role in the cardiovascular effects of both visceral and deep somatic pain. These findings are consistent with reports that visceral and deep somatic nociception cause hypotension and bradycardia whereas superficial somatic pain produces hypertension and tachycardia [11,26]. It is an oversimplification to assume that all types of visceral pain produce the same effect on cardiovascular function, however. Mechanical distension of the stomach [28], colon [27] or bladder [41] raises arterial pressure and heart rate, for example, whereas distension of the small intestine [10,38] or renal pelvis [4] causes arterial pressure to fall. It is important to note these studies were conducted in anesthetized animals, however, and anesthesia can influence cardiovascular reflexes. Colorectal distension evokes a depressor response in pentobarbital-anesthetized rats [27,40], for example, but raises arterial pressure and heart rate in conscious animals [15,40]. The specific type of anesthetic is also an important factor. Colorectal distension also lowers arterial pressure in rats anesthetized with a-chloralose or urethane but causes hypertension under halothane or ketamine anesthesia as it does in conscious animals [39,40]. Ideally, this type of investigation should be conducted in conscious animals but this raises ethical concerns for models of prolonged or inescapable pain. For this reason, the present experiments were performed in anesthetized animals but this raises the possibility that the noxious stimuli we used may evoke a depressor response only under halothane-anesthesia. To test this possibility, Keay et al. [26] administered formalin im to rats under brief (1– 2 min) carbon dioxide narcosis. Carbon dioxide caused a transient pressor response but arterial pressure subsequently fell 15 – 20 mm Hg and remained below baseline levels for the rest of the 20 min experiment. The magnitude and duration of the response was comparable to that evoked in halothane-
243
anesthetized rats [11] and is consistent with the present data. Intramuscular formalin administration thus appears to lower arterial pressure to a comparable extent in halothane-anesthetized and unanesthetized rats. Nevertheless, it is important to emphasize that, without conducting parallel experiments in conscious animals, we cannot rule out the possibility that halothane-anesthesia may have influenced the results reported here. We also found the vasodepression caused by visceral nociception to be inhibited by naltrindole, a delta opioid receptor antagonist, but not by CTOP or nor-BNI, which block mu and kappa receptors. The ineffectiveness of CTOP and nor-BNI is unlikely to be due to an inadequate drug dose because Kd values for [3H]-CTOP (0.16 nM) [18] and [3H]nor-BNI (0.10 nM) [33] binding to rat brain membranes are orders of magnitude lower than the concentrations of CTOP (10 mM) and nor-BNI (6 mM) microinjected into the vlPAG in the present study. CTOP and nor-BNI also prevent stressinduced analgesia at doses that are substantially lower than used in this study [34,53]. The Kd for [3H]-naltrindole binding to delta opioid receptors (0.08 nM) [14] is also considerably lower than the naltrindole concentration used here (2 mM) and previous studies have shown that naltrindole injection into the vlPAG inhibits analgesia evoked from the amygdala at doses (0.3 –8 nmol) [43,48] similar to the dose used in the present study. These considerations support the conclusion that delta, but not mu or kappa, receptors mediate the fall in arterial pressure and heart rate evoked by visceral nociception. Histological localization of injection sites revealed that many of the lidocaine, cobalt chloride and naltrindole injections spread beyond the borders of the vlPAG into the adjacent midbrain tegmentum. Earlier mapping studies also showed that excitatory amino acids lower arterial pressure when injected into the midbrain tegmentum ventrolateral to the vlPAG of halothane-anesthetized rats [3,24] and precollicular decerebrated cats [6]. Excitatory amino acid injection into this region also produces the same behavioral response, quiescence and hyporeactivity, as injections restricted to the vlPAG and both visceral and deep somatic nociception induce c-fos expression in neurons both within and ventrolateral to the vlPAG [3]. Importantly, delta receptor agonist injection evokes a depressor response both within the vlPAG and in the tegmental area ventrolateral to the vlPAG [24]. The midbrain region that mediates the depressor response to visceral nociception thus extends beyond the boundaries of the vlPAG. The present findings extend earlier evidence that lidocaine [8] and naltrindole [9] administration into the vlPAG delays the onset and reduces the magnitude of hemorrhagic hypotension. Earlier investigations had shown the decompensatory phase of hemorrhage to be attenuated by intraventricular administration of naloxone [47] and delta receptor antagonists [30]. We found that naltrindole injection into the vlPAG attenuated the decompensatory phase of hemorrhage at doses ranging from 0.2 to 20 nmol; the dose used in the
244
S. Cavun et al. / Brain Research 1019 (2004) 237–245
present study, 2 nmol, was approximately the ED50 [9]. Mu and kappa receptor antagonists were ineffective. These findings suggest that visceral nociception and severe blood loss lower arterial pressure through the same, or a similar, anatomical pathway through the vlPAG in which delta opioid receptors are an important component. Several lines of evidence suggest that nociceptive and hemorrhagic hypotension may be mediated by a common anatomical pathway similar to the descending pathway that modulates pain perception [35]. Ventrolateral PAG neurons innervate the nucleus raphe magnus, an important component of the descending pain control pathway [35], and a more caudal midline medullary region encompassing portions of the raphe obscurus and raphe pallidus nuclei [19]. Activation of neurons in the caudal midline medulla with excitatory amino acids evokes a fall in arterial pressure and heart rate [13,19,51]. Conversely, inactivation of this region with lidocaine or cobalt chloride prevents the fall in arterial pressure caused by hemorrhage [20] without affecting resting blood pressure or baroreceptor reflexes [20,45]. Serotonergic neurons in the nucleus raphe magnus, raphe pallidus and raphe obscurus directly innervate the sympathoexcitatory region of the rostral ventrolateral medulla (RVLM) [2,56] and blockade of serotonin-1A receptors in the RVLM prevents the hypotension caused by vlPAG stimulation [1] and severe hemorrhage [16]. These data suggest that vlPAG activation lowers arterial pressure through a descending pathway that includes the rostral [52] or caudal [13,16,20,50] midline medulla and RVLM. Alternatively, vlPAG neurons innervate the RVLM directly [2,5,7,50] and there is evidence that deep somatic nociception evokes c-fos expression in vlPAG neurons that innervate the RVLM independent of the medullary depressor region [25]. In summary, these data support the hypothesis that visceral nociception and severe blood loss cause hypotension and bradycardia by activating a descending pathway through the vlPAG in which delta opioid receptors play a significant role. In some respects, this pathway is similar to the descending pain control pathway although pain control is mediated primarily by mu, rather than delta, opioid receptors in the vlPAG [54]. Hypotension and bradycardia appear to be part of a broader, integrated response to inescapable pain and severe blood loss that includes behavioral quiescence, hyporeactivity to environmental stimuli and opioid mediated analgesia [3,36,53]. Presumably, the hypotension, behavioral inactivity and profound analgesia that develops when the vlPAG is activated serves to reduce blood loss during severe hemorrhage and prevent exacerbation of deep tissue injury.
Acknowledgements This research was supported by a grant from the Office of Naval Research (N00014-98-1-02-0249).
References [1] M. Bago, C. Dean, Sympathoinhibition from ventrolateral periaqueductal gray mediated by 5-HT1A receptors in the RVLM, Am. J. Physiol. 280 (2001) R976 – R984. [2] M. Bago, L. Marson, C. Dean, Serotonergic projections to the rostroventrolateral medulla from midbrain and raphe nuclei, Brain Res. 945 (2002) 249 – 258. [3] R. Bandler, K.A. Keay, Columnar organization in the midbrain periaqueductal gray and the integration of emotional expression, Prog. Brain Res. 107 (1996) 285 – 300. [4] H. Brasch, G. Zetler, Caerulein and morphine in a model of visceral pain. Effects on the hypotensive response to renal pelvis distension in the rat, Naunyn-Schmiedeberg’s Arch. Pharmacol. 319 (1982) 161 – 167. [5] A.A. Cameron, I.A. Khan, K.N. Westlund, W.D. Willis, The efferent projections of the periaqueductal gray in the rat: a Phaseolus vulgarisleucoagglutinin study. II. Descending projections, J. Comp. Neurol. 351 (1995) 585 – 601. [6] P. Carrive, The periaqueductal gray and defensive behavior: functional representation and neuronal organization, Behav. Brain Res. 58 (1993) 27 – 47. [7] P. Carrive, R. Bandler, Viscerotopic organization of neurons subserving hypotensive reactions within the midbrain periaqueductal grey: a correlative functional and anatomical study, Brain Res. 541 (1991) 206 – 215. [8] S. Cavun, W.R. Millington, Evidence that hemorrhagic hypotension is mediated by the ventrolateral periaqueductal gray, Am. J. Physiol. 281 (2001) R747 – R752. [9] S. Cavun, G.E. Resch, M. Rapacon-Baker, A. Evec, W.R. Millington, Blockade of delta opioid receptors in the ventrolateral periaqueductal gray region inhibits the fall in arterial pressure evoked by hemorrhage, J. Pharmacol. Exp. Ther. 297 (2001) 612 – 619. [10] S.J. Clark, R.L. Follenfant, T.W. Smith, Evaluation of opioid-induced antinociceptive effects in anaesthetized and conscious animals, Br. J. Pharmacol. 95 (1988) 275 – 283. [11] C.I. Clement, K.A. Keay, B.K. Owler, R. Bandler, Common patterns of increased and decreased Fos expression in midbrain and pons evoked by noxious deep somatic and noxious visceral manipulations in the rat, J. Comp. Neurol. 366 (1996) 495 – 515. [12] C.I. Clement, K.A. Keay, K. Podzebenko, B.D. Gordon, R. Bandler, Spinal sources of noxious visceral and noxious deep somatic afferent drive onto the ventrolateral periaqueductal gray of the rat, J. Comp. Neurol. 425 (2000) 323 – 344. [13] M.J. Coleman, R.A. Dampney, Powerful depressor and sympathoinhibitory effects evoked from neurons in the caudal raphe pallidus and obscurus, Am. J. Physiol. 268 (1995) R1295 – R1302. [14] P.C. Contreras, L. Tam, E. Drower, M.F. Rafferty, [3H]Naltrindole: a potent and selective ligand for labeling y-opioid receptors, Brain Res. 604 (1993) 160 – 164. [15] R.M. Danzebrink, G.F. Gebhart, Antinociceptive effects of intrathecal adrenoceptor agonists in a rat model of visceral nociception, J. Pharmacol. Exp. Ther. 253 (1990) 698 – 705. [16] C. Dean, M. Bago, Renal sympathoinhibition mediated by 5-HT1A receptors in the RVLM during severe hemorrhage in rats, Am. J. Physiol. 282 (2002) R122 – R130. [17] A. Depaulis, K.A. Keay, R. Bandler, Quiescence and hyporeactivity evoked by activation of cell bodies in the ventrolateral midbrain periaqueductal gray of the rat, Exp. Brain Res. 99 (1994) 75 – 83. [18] K.N. Hawkins, R.J. Knapp, G.K. Lui, K. Gulya, W. Kazmierski, Y.P. Wan, J.T. Pelton, V.J. Hruby, H.I. Yamamura, [3H]-[H-D-Phe-CysTyr-D-Trp-Orn-Thr-Pen-Thr-NH2] ([3H]CTOP), a potent and highly selective peptide for mu opioid receptors in rat brain, J. Pharmacol. Exp. Ther. 248 (1989) 73 – 80. [19] L.A. Henderson, K.A. Keay, R. Bandler, The ventrolateral periaqueductal gray projects to caudal brainstem depressor regions: a func-
S. Cavun et al. / Brain Research 1019 (2004) 237–245
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32] [33]
[34]
[35] [36] [37]
[38]
tional-anatomical and physiological study, Neuroscience 82 (1998) 201 – 221. L.A. Henderson, K.A. Keay, R. Bandler, Caudal midline medulla mediates behaviorally-coupled but not baroreceptor-mediated vasodepression, Neuroscience 98 (2000) 779 – 792. K.A. Keay, R. Bandler, Deep and superficial noxious stimulation increases Fos-like immunoreactivity in different regions of the midbrain periaqueductal grey of the rat, Neurosci. Lett. 154 (1993) 23 – 26. K.A. Keay, R. Bandler, Distinct central representations of inescapable and escapable pain: observations and speculations, Exp. Physiol. 87 (2002) 275 – 279. K.A. Keay, C.I. Clement, B. Owler, A. Depaulis, R. Bandler, Convergence of deep somatic and visceral nociceptive information onto a discrete ventrolateral midbrain periaqueductal gray region, Neuroscience 61 (1994) 727 – 732. K.A. Keay, L.J. Crowfoot, N.S. Floyd, L.A. Henderson, M.J. Christie, R. Bandler, Cardiovascular effects of microinjections of opioid agonists into the ‘depressor region’ of the ventrolateral periaqueductal gray region, Brain Res. 762 (1997) 61 – 71. K.A. Keay, Q.F. Li, R. Bandler, Muscle pain activates a direct projection from ventrolateral periaqueductal gray to rostral ventrolateral medulla in rats, Neurosci. Lett. 290 (2000) 157 – 160. K.A. Keay, C.I. Clement, W.M. Matar, D.J. Heslop, L.A. Henderson, R. Bandler, Noxious activation of spinal or vagal afferents evokes distinct patterns of fos-like immunoreactivity in the ventrolateral periaqueductal gray of unanaesthetized rats, Brain Res. 948 (2002) 122 – 130. C.M. Kozlowski, A. Green, D. Grundy, F.M. Boissonade, C. Bountra, The 5-HT3 receptor antagonist alosetron inhibits the colorectal distention induced depressor response and spinal c-fos expression in the anaesthetised rat, Gut 46 (2000) 474 – 480. P. Li, K. Rowshan, M. Crisostomo, S.C. Tjen-A-Looi, J.C. Longhurst, Effect of electroacupuncture on pressor reflex during gastric distension, Am. J. Physiol. 283 (2002) R1335 – R1345. T.A. Lovick, Integrated activity of cardiovascular and pain regulatory systems: role in adaptive behavioral responses, Prog. Neurobiol. 40 (1993) 631 – 644. J. Ludbrook, S. Ventura, The decompensatory phase of acute hypovolaemia in rabbits involves a central y1-opioid receptor, Eur. J. Pharmacol. 252 (1994) 113 – 116. B.M. Lumb, Inescapable and escapable pain is represented in distinct hypothalamic-midbrain circuits: specific roles for A y- and C-nociceptors, Exp. Physiol. 87 (2002) 281 – 286. J.G. Malpeli, Reversible inactivation of subcortical sites by drug injection, J. Neurosci. Methods 86 (1999) 119 – 128. ´ . Ma´rki, F. O ¨ tvo¨s, G. To´th, S. Hosztafi, A. Borsodi, Tritiated kappa A receptor antagonist norbinaltorphimine: synthesis and in vitro binding in three different tissues, Life Sci. 66 (2000) 43 – 49. L. Menendez, F. Andres-Trelles, A. Hidalgo, A. Baamonde, Involvement of spinal n opioid receptors in a type of footshock induced analgesia in mice, Brain Res. 611 (1993) 264 – 271. M.J. Millan, Descending control of pain, Prog. Neurobiol. 66 (2002) 355 – 474. P.E. Molina, Endogenous opioid analgesia in hemorrhagic shock, J. Trauma 54 (5 Suppl) (2003) S126 – S132. M.M. Morgan, P. Carrive, Activation of the ventrolateral periaqueductal gray reduces locomotion but not mean arterial pressure in awake, freely moving rats, Neuroscience 102 (2001) 905 – 910. H.E. Moss, G.J. Sanger, The effects of granisetron, ICS 205-930 and
[39] [40]
[41]
[42]
[43]
[44] [45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54] [55]
[56]
245
odansetron on the visceral pain reflex induced by duodenal distension, Br. J. Pharmacol. 100 (1990) 497 – 501. T.J. Ness, Models of visceral nociception, ILAR J. 40 (1999) 119 – 128. T.J. Ness, G.F. Gebhart, Colorectal distension as a noxious visceral stimulus: physiologic and pharmacologic characterization of pseudaffective reflexes in the rat, Brain Res. 450 (1988) 153 – 169. T.J. Ness, A. Lewis-Sides, P. Castroman, Characterization of pressor and visceromotor reflex responses to bladder distention in rats: sources of variability and effect of analgesics, J. Urol. 165 (2001) 968 – 974. S. Ohtori, K. Takahashi, T. Chiba, Y. Takahashi, M. Yamagata, H. Sameda, H. Moriya, Fos expression in the rat brain and spinal cord evoked by noxious stimulation to low back muscle and skin, Spine 25 (2000) 2425 – 2430. Z.W. Pavlovic, M.L. Cooper, R.J. Bodnar, Opioid antagonists in the periaqueductal gray inhibit morphine and h-endorphin analgesia elicited from the amygdala of rats, Brain Res. 741 (1996) 13 – 26. G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1998. J.R. Potas, R.A. Dampney, Sympathoinhibitory pathway from caudal midline medulla to RVLM is independent of baroreceptor reflex pathway, Am. J. Physiol. 284 (2003) R1071 – R1078. L. Rodella, R. Rezzani, M. Gioia, G. Tredici, R. Bianchi, Expression of Fos immunoreactivity in the rat supraspinal regions following noxious visceral stimulation, Brain Res. Bull. 47 (1998) 357 – 366. J.C. Schadt, J. Ludbrook, Hemodynamic and neurohumoral responses to acute hypovolemia in conscious mammals, Am. J. Physiol. 260 (1991) H305 – H318. R. Shane, J. Acosta, G.C. Rossi, R.J. Bodnar, Reciprocal interactions between the amygdala and ventrolateral periaqueductal gray in mediating of Q/N1 – 17 -induced analgesia in the rat, Brain Res. 980 (2003), 57 – 70. B.K. Taylor, M.A. Peterson, A.I. Basbaum, Continuous intravenous infusion of naloxone does not change behavioral, cardiovascular, or inflammatory responses to subcutaneous formalin in the rat, Pain 69 (1997) 171 – 177. E.J. Van Bockstaele, G. Aston-Jones, V.A. Pieribone, M. Ennis, M.T. Shipley, Subregions of the periaqueductal gray topographically innervate the rostral ventral medulla in the rat, J. Comp. Neurol. 309 (1991) 305 – 327. A.J. Verberne, D.M. Sartor, A. Berke, Midline medullary depressor responses are mediated by inhibition of RVLM sympathoexcitatory neurons in rats, Am. J. Physiol. 276 (1999) R1054 – R1062. W.H. Wang, T.A. Lovick, The inhibitory effect of the ventrolateral periaqueductal gray matter on neurones in the rostral ventrolateral medulla involves a relay in the medullary raphe nuclei, Exp. Brain Res. 94 (1993) 295 – 300. C.P. Wiedenmayer, G.A. Barr, A Opioid receptors in the ventrolateral periaqueductal gray mediate stress-induced analgesia but not immobility in rat pups, Behav. Neurosci. 114 (2000) 125 – 136. T.L. Yaksh, N.R.F. Al-Rodhan, T.S. Jensen, Sites of action of opiates in production of analgesia, Prog. Brain Res. 77 (1988) 371 – 394. M.H. Yoon, T.L. Yaksh, The effect of intrathecal gabapentin on pain behavior and hemodynamics on the formalin test in the rat, Anesth. Analg. 89 (1999) 434 – 439. A. Zagon, Internal connections in the rostral ventromedial medulla of the rat, J. Auton. Nerv. Syst. 53 (1995) 43 – 56.
Research report
The hypotension evoked by visceral nociception is mediated by delta opioid receptors in the periaqueductal gray Sinan Cavun a,b, Gokhan Goktalay a,b, William R. Millington a,* a
Department of Basic and Pharmaceutical Sciences, Albany College of Pharmacy, Union University, 106 New Scotland Ave, Albany, NY 12208-3492, USA b Department of Pharmacology and Clinical Pharmacology, Uludag University School of Medicine, Bursa, Turkey Accepted 4 June 2004 Available online 10 July 2004
Abstract This study tested the hypothesis that the ventrolateral column of the midbrain periaqueductal gray (vlPAG) region mediates the hypotension and bradycardia evoked by visceral nociception. To test this, the local anesthetic lidocaine (2%; 0.5 Al) was microinjected into the vlPAG of halothane-anesthetized rats bilaterally and visceral nociception was induced 2 min later by injecting 5% acetic acid (0.5 ml) intraperitoneally. Acetic acid injection caused an abrupt fall in arterial pressure ( 12.2 F 2.1 mm Hg) and heart rate ( 37 F 93 bpm) lasting approximately 15 min. Lidocaine injection into the vlPAG prevented the fall in arterial pressure and heart rate completely. Cobalt chloride (5 mM; 0.2 or 0.5 Al) injection into the vlPAG also prevented nociceptive hypotension but it did not affect the fall in heart rate significantly. Lidocaine pretreatment also inhibited the depressor response caused by intramuscular formalin (5%; 0.2 ml) administration, a model of deep somatic nociception, although it did not prevent the response completely. To determine if opioid receptors mediate the response, selective mu, delta or kappa opioid receptor antagonists were microinjected into the vlPAG 5 min before intraperitoneal (ip) acetic acid administration. Naltrindole, a delta receptor antagonist, inhibited the response significantly but mu and kappa antagonists were completely ineffective. Lidocaine and naltrindole had no effect when injected into the dorsolateral PAG and did not influence cardiovascular function when injected into the vlPAG of saline treated control animals. These data support the hypothesis that the vlPAG mediates the depressor response evoked by visceral nociception and indicate that delta opioid receptors participate in the response. D 2004 Elsevier B.V. All rights reserved. Theme: Endocrine and autonomic regulation Topic: Cardiovascular regulation Keywords: Visceral nociception; Pain; Central cardiovascular regulation; Periaqueductal gray; Opioid receptor; Delta receptor
1. Introduction Pain produces quite different autonomic and behavioral reactions depending upon its location. Visceral and deep somatic pain cause arterial pressure and heart rate to fall and produce a behavioral response characterized by quiescence, immobility and inattention to environmental stimuli [3,22,29,31]. Cutaneous and superficial somatic pain generally produce a completely different reaction, characterized by hypertension, tachycardia and behavioral activation, * Corresponding author. Tel.: +1-518-445-7242; fax: +1-518-4457202. E-mail address: [email protected] (W.R. Millington). 0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.06.003
the classic fight or flight response [3,49,55]. The central mechanisms that coordinate the autonomic and behavioral reactions to visceral and somatic pain are not fully understood, and particularly so for visceral pain. The midbrain periaqueductal gray (PAG) region is thought to play a pivotal role in the coordinate regulation of cardiovascular function and behavior [3,6,29]. The PAG is a highly organized structure that is segregated into functionally distinct longitudinal columns [3]. Excitatory amino acid injection into the ventrolateral column of the PAG (vlPAG) evokes a prompt depressor response in anesthetized rats [9,24,29] and decerebrate cats [6,7] and causes quiescence, inattention to the environment and reduced social interaction in conscious animals, a response
238
S. Cavun et al. / Brain Research 1019 (2004) 237–245
much like that caused by visceral pain [17,23,37]. In contrast, excitatory amino acids evoke a pressor response and either behavioral activation or reactive immobility when injected into the lateral (lPAG) or dorsolateral PAG (dlPAG) [3,29], similar to the effects of somatic pain. These findings suggest that the vlPAG mediates the depressor effect of visceral pain whereas the lPAG and dlPAG contribute to the effects of somatic pain. The hypothesis that PAG neurons participate in the cardiovascular response to visceral and somatic pain finds additional support in studies using the immediate early gene c-fos as a marker of neuronal activation. Intraperitoneal (ip) administration of noxious chemicals to anesthetized rats, a standard method for provoking inescapable visceral nociception [39], induces c-fos expression in the vlPAG but not, to an equivalent extent, in other PAG subregions [11,23,42,46]. By contrast, cutaneous nociception induces c-fos expression primarily in the lPAG and dlPAG [21,42]. These findings are consistent with the idea that visceral pain lowers arterial pressure by activating vlPAG neurons although they by no means prove such a connection exists. Visceral pain may very well induce cfos in vlPAG neurons that influence pain perception, motor activity or other complex behaviors rather than cardiovascular regulation, for example [3]. Hence, c-fos expression does not provide direct evidence that visceral nociception lowers arterial pressure by activating vlPAG neurons. The first objective of this study was to test the hypothesis that the vlPAG mediates the depressor response to visceral nociception by determining if the local anesthetic lidocaine or the synaptic inhibitor cobalt chloride would prevent the response [32]. We found that bilateral injection of lidocaine or cobalt chloride into the vlPAG of halothane-anesthetized rats inhibited the fall in arterial pressure and heart rate caused by visceral nociception significantly, consistent with the hypothesis that the vlPAG plays a critical role in the response. The second objective was to determine if opioid receptors in the vlPAG contribute to the cardiovascular effects of visceral nociception. Opioid receptors have been localized in the PAG and their role in pain perception has been studied extensively [54] but there is also evidence that they influence cardiovascular function. Microinjection of a selective delta opioid receptor agonist [24] or the endogenous opioid peptide h-endorphin [9] into the vlPAG lowers arterial pressure and heart rate in halothane-anesthetized rats. Mu receptor activation, on the other hand, evokes a pressor response and kappa receptor agonist injection produces an equivocal effect [24]. Here we report that blockade of delta receptors in the vlPAG effectively inhibits the depressor effect of visceral nociception. These data support the hypothesis that the vlPAG plays a critical role in the cardiovascular effects of visceral pain and indicate that delta receptors participate in the response.
2. Materials and methods 2.1. Animals and surgery One hundred five male Sprague – Dawley rats (250 – 350 g; Charles River Laboratories, Wilmington, MA) were housed at a constant temperature (21 F 1 jC) under a 12 h light/dark schedule. Food and water were freely available. The rats were anesthetized with 4% halothane and maintained with 1.5% halothane in 100% O2. The left carotid artery was cannulated with PE-50 tubing filled with heparinized saline (100 U/ml), exteriorized at the nape of the neck and sealed until use [8]. At the beginning of each experiment, the arterial catheter was attached to a volumetric pressure transducer and arterial pressure and heart rate were monitored continuously and recorded at 1 min intervals using a MicroMed BPA-200 blood pressure analyzer (Micro-Med, Louisville, KY). The animal protocols were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee. 2.2. Lidocaine, cobalt chloride and opioid receptor antagonist injections Two 26 gauge guide cannulae were implanted bilaterally above the caudal vlPAG or dlPAG at a 27j rostro-caudal angle through burr holes drilled through the skull [8]. For caudal vlPAG injections, the tips of the cannulae were positioned 0.8 mm lateral and 8.3 mm posterior to bregma and 6.5 mm below the skull surface according to the atlas of Paxinos and Watson [44]. For caudal dlPAG injections, the guide cannulae were implanted bilaterally 0.8 mm lateral and 8.3 mm posterior to bregma and 4.6 mm below the skull surface. After obtaining stable baseline arterial pressure and heart rate measurements, two 33 gauge injection cannulae were inserted through the guide cannulae to a depth of 1.0 mm below the tip of the guide cannula and lidocaine (2%), cobalt chloride (5 mM), naltrindole HCl (1 nmol/cannula), D-Phe-Cys-Tyr-D-Trp-Orn-Thr-Pen-Thramide (CTOP; 5 nmol/cannula), nor-binaltorphimine dihydrochloride (nor-BNI; 3 nmol/cannula; Sigma, St. Louis, MO) or saline was injected bilaterally. The lidocaine and cobalt chloride doses were based on Malpeli [32] and Cavun and Millington [8], the naltrindole dose was selected from previous dose – response studies [9] and the CTOP and nor-BNI doses were based on earlier reports [18,33,34,53]. Saline-treated controls were interspersed with drug treated animals to ensure against the possibility that different groups of animals or inadvertent changes in technique might influence the results. Drugs were dissolved and injected in 0.2 or 0.5 Al saline (pH 7.2– 7.4) and were delivered at a constant rate over a 1-min period. The injection volume was monitored by observing the movement of an air bubble placed in the tubing.
S. Cavun et al. / Brain Research 1019 (2004) 237–245
239
2.3. Noxious stimuli Two or five minutes after PAG drug administration, animals received a single ip injection (0.5 ml) of 5% acetic acid to induce visceral nociception [11,12,23]. Deep somatic nociception was caused by injecting 5% formalin (0.2 ml; Sigma Chemical Co.; contains 10 –15% methanol) into the gastrocnemius muscle bilaterally [11,26]. Blood pressure and heart rate were recorded for 15 min after ip acetic acid injection or 40 min after im formalin administration. Each rat was used for only one experiment. 2.4. Histology At the end of each experiment, the cannulae placements were verified by injecting 0.2 or 0.5 Al India ink. The animal was then sacrificed, the brain was removed, immersed in 10% paraformaldehyde, imbedded with paraffin, sectioned (50 Am) with a sliding microtome and stained with eosin. 2.5. Statistical analysis Results are expressed as mean F S.E.M. Data were analyzed by repeated measures two-way analysis of variance (ANOVA) with repeated measures followed by Dunnett’s multiple comparison test using SigmaStat version 3.0 (SPSS, Chicago, IL). The criterion for statistical significance was p < 0.05.
3. Results 3.1. Visceral nociception To test the hypothesis that the vlPAG mediates the depressor effect of visceral nociception, we inhibited neuronal activity in the vlPAG with lidocaine before inducing visceral nociception with ip acetic acid. As shown in Fig. 1, ip injection of 5% acetic acid (0.5 ml) evoked a prompt fall in arterial pressure ( 12.3 F 2.1 mm Hg) and heart rate ( 27 F 9 bpm), consistent with data published previously [11]. The depressor response was transient and both arterial pressure and heart rate returned to baseline values within 15 min (Fig. 1). Bilateral lidocaine injection (0.5 Al of a 2% solution) into the caudal vlPAG prevented the hypotension [ F(15,180) = 27.2, P < 0.001] and bradycardia [ F(15,180) = 10.24, P < 0.001] evoked by ip acetic acid completely (Fig. 1). Lidocaine did not affect arterial pressure or heart rate when microinjected into the caudal vlPAG of control animals that received ip saline in lieu of acetic acid (Fig. 1). To verify the specificity of the vlPAG injections, we tested whether lidocaine administration into the dlPAG would influence the depressor effect of ip acetic acid
Fig. 1. Lidocaine injection into the vlPAG inhibits the fall in arterial pressure and heart rate caused by visceral nociception. Lidocaine (2%; 0.5 Al) or saline was injected bilaterally into the caudal vlPAG of halothaneanesthetized rats and, 2 min later, acetic acid (5%, 0.5 ml) was administered ip. Mean arterial pressure (MAP; top panel) and heart rate (bottom panel) were recorded at 1 min intervals for 15 min. Data are presented as mean F S.E.M. change in mean arterial pressure or heart rate from baseline values. The numbers in parentheses indicate the number of animals in each group. Baseline mean arterial pressure at the 2 min time point was 99.5 F 2.4 mm Hg for the saline + acetic acid group, 99.4 F 1.9 mm Hg for the lidocaine + acetic acid group and 98.7 F 1.9 mm Hg for the lidocaine + saline controls. Baseline heart rate was 346 F 13 bpm for the saline + acetic acid group, 348 F 9 bpm for the lidocaine + acetic acid group and 360 F 24 bpm for the lidocaine + saline controls. *P < 0.05, **P < 0.01 compared to the corresponding time point for saline-treated control animals.
injection. Previous investigations showed that excitatory amino acid injection into the dlPAG or lPAG produces hypertension and tachycardia [3,29], which indicates that the dlPAG is unlikely to participate in the depressor response evoked by noxious stimuli. As expected, lidocaine injection into the caudal dlPAG bilaterally failed to prevent the fall in arterial pressure or heart rate evoked by subsequent administration of 5% acetic acid by ip injection (Fig. 2). The depressor effect of visceral nociception is thus prevented by neuronal inactivation in the vlPAG, but not the dlPAG.
240
S. Cavun et al. / Brain Research 1019 (2004) 237–245
imately 15 min, and the total duration of the response was longer, at least 40 min, than that of visceral nociception, however. Furthermore, although arterial pressure returned toward baseline values toward the end of the 40 min experimental period, heart rate remained depressed (Fig. 3). Lidocaine pretreatment attenuated the fall in arterial pressure caused by im formalin injection significantly [ F(20,240) = 2.6, P < 0.001] although it did not prevent the response completely (Fig. 3). Heart rate was not influenced significantly by lidocaine pretreatment. 3.3. Cobalt chloride injection To determine whether synaptic activity is required for the response, we tested whether the vasodepression
Fig. 2. Lidocaine injection into the dlPAG does not prevent the hypotension and bradycardia evoked by ip acetic acid administration. Lidocaine (2%, 0.5 Al) or saline was injected bilaterally into the caudal dlPAG of halothane-anesthetized rats 2 min before injection of 5% acetic acid (0.5 ml, ip) and mean arterial pressure (MAP; top panel) and heart rate (bottom panel) were recorded for 15 min. Baseline mean arterial pressure was 102.4 F 2.1 mm Hg for the saline + acetic acid group, 101.3 F 3.8 mm Hg for the lidocaine + acetic acid group and 97.6 F 1.1 mm Hg for the lidocaine + saline group. Baseline heart rate was 336 F 11 bpm for the saline + acetic acid group, 358 F 20 bpm for the lidocaine + acetic acid group and 344 F 17 bpm for the lidocaine + saline group.
3.2. Deep somatic nociception To investigate whether the vlPAG mediates the depressor effect of deep somatic nociception, we tested whether lidocaine microinjection into the caudal vlPAG prevents the hypotension and bradycardia evoked by im formalin administration to halothane-anesthetized rats [11]. Formalin (5%; 0.2 ml) was injected into both gastrocnemius muscles 2 min after lidocaine (2%; 0.5 Al) or saline was microinjected into the vlPAG bilaterally. In control animals, 5% formalin injection lowered arterial pressure 14.8 F 2.2 mm Hg and reduced heart rate 68 F 5 bpm (Fig. 3), comparable to the hypotension and bradycardia caused by ip acetic acid injection. The maximal response took longer to develop, approx-
Fig. 3. Lidocaine injection into the vlPAG attenuates the fall in arterial pressure and heart rate induced by intramuscular formalin injection. Lidocaine (2%, 0.5 Al) or saline was injected bilaterally into the caudal vlPAG and, after a 2-min delay, formalin (0.2 ml) was injected into both gastrocnemius muscles of halothane-anesthetized rats. Baseline mean arterial pressure at the 2 min time point was 99.3 F 2.2 mm Hg for the saline + formalin group and 100.7 F 0.9 mm Hg for the lidocaine + formalin group. Baseline heart rate was 382 F 7 bpm for the saline + formalin group and 356 F 8 bpm for the lidocaine + formalin group. *P < 0.05 compared to saline-treated control animals.
S. Cavun et al. / Brain Research 1019 (2004) 237–245
241
caine, cobalt chloride injection into the vlPAG inhibits the hypotension evoked by ip lidocaine without influencing arterial pressure or heart rate in normotensive animals. 3.4. Opioid receptor antagonists To test whether delta opioid receptors in the vlPAG contribute to the depressor effect of visceral nociception, we injected naltrindole (1 nmol/cannula) bilaterally into the caudal vlPAG 5 min before 5% acetic acid was administered ip to halothane-anesthetized rats. Naltrindole pretreatment inhibited the fall in arterial pressure
Fig. 4. Cobalt chloride injection into the vlPAG prevents the fall in arterial pressure and heart rate evoked by ip acetic acid. Cobalt chloride (5 mM; 0.2 or 0.5 Al) or saline was injected bilaterally into the vlPAG 5 min before acetic acid (5%, 0.5 ml) was injected ip. Baseline mean arterial pressure was 97.0 F 3.8 mm Hg for the saline + acetic acid group, 101.2 F 4.3 mm Hg for the 0.2 Al cobalt chloride + acetic acid group and 94.3 F 2.3 mm Hg for the 0.5 Al cobalt chloride + acetic acid group. Baseline heart rate was 354 F 8 bpm for the saline + acetic acid group, 343 F 8 bpm for the 0.2 Al cobalt chloride + acetic acid group and 356 F 13 bpm for the 0.5 Al cobalt chloride + acetic acid group. *P < 0.05, **P < 0.01 compared to salinetreated control animals.
caused by ip acetic acid is attenuated by cobalt chloride, which inhibits synaptic transmission but not axonal conductance [24]. Cobalt chloride (5 mM; 0.2 or 0.5 Al) was injected into the vlPAG of halothane-anesthetized rats bilaterally followed, 5 min later, by ip acetic acid. Cobalt chloride pretreatment inhibited the fall in arterial pressure caused by ip acetic acid administration significantly [ F(30,240) = 3.8, P < 0.001] although it did not affect the fall in heart rate significantly (Fig. 4). Interestingly, 0.2 Al injections of 5 mM cobalt chloride inhibited nociceptive hypotension to approximately the same extent as 0.5 Al (Fig. 4). Previously, we showed that cobalt chloride (5 mM; 0.5 Al) injection into the vlPAG of normotensive control animals had no effect on arterial pressure or heart rate [8]. Hence, like lido-
Fig. 5. Naltrindole injection into the caudal vlPAG inhibits the hypotension and bradycardia caused by visceral nociception. The delta opioid receptor antagonist naltrindole (1 nmol/cannula) or saline (0.5 Al) was injected bilaterally into the caudal vlPAG and, after a 5-min delay, acetic acid (5%, 0.5 ml) was administered ip to halothane-anesthetized rats. Data are presented as the mean F S.E.M. change in mean arterial pressure (top panel) and heart rate (bottom panel) from baseline values. Baseline mean arterial pressure at the 5 min time point was 104.0 F 2.1 mm Hg for the saline + acetic acid group, 100.6 F 1.6 mm Hg for the naltrindole + acetic acid group and 102.9 F 1.5 mm Hg for naltrindole + saline controls. Baseline heart rate was 357 F 8 bpm for the saline + acetic acid group, 326 F 4 bpm for the naltrindole + acetic acid group and 342 F 7 bpm for the naltrindole + saline group. *P < 0.05, **P < 0.01 compared to the corresponding time point for saline-treated controls.
242
S. Cavun et al. / Brain Research 1019 (2004) 237–245
prevent the hypotension or bradycardia caused by ip acetic acid. Fig. 6 shows that CTOP and nor-BNI failed to significantly attenuate the fall in arterial pressure or heart rate ip acetic acid causes. Together, these data support the conclusion that activation of delta, but not mu or kappa, receptors in the vlPAG contributes to the depressor effect of visceral nociception. 3.5. Location of injection sites Histological analysis confirmed that lidocaine, cobalt chloride, naltrindole, CTOP and nor-BNI injections were located in or immediately adjacent to the vlPAG or dlPAG (Fig. 7A). Some injection sites were centered outside, but immediately adjacent to, the vlPAG. Likewise, some injections targeted on the dlPAG were centered outside, but immediately adjacent to, the dlPAG or lPAG. Fig. 7B illustrates the area of the midbrain stained by 0.5 Al dye that identifies the site of a lidocaine injection targeted on the vlPAG. The injection covered an area approximately 0.7 mm in diameter in both the rostrocaudal and dorsoventral planes and spread outside the boundaries of the vlPAG ventrolaterally into the adjacent
Fig. 6. CTOP and nor-BNI fail to prevent the fall in arterial pressure (top panel) and heart rate (bottom panel) caused by visceral nociception. The mu receptor antagonist CTOP (5 nmol/cannula), the kappa receptor antagonist nor-BNI (3 nmol/cannula) or saline (0.5 Al) was injected bilaterally into the caudal vlPAG of halothane-anesthetized rats; after a 5min delay, acetic acid (5%, 0.5 ml) was injected ip. Baseline mean arterial pressure at the 5 min time point was 99.0 F 3.3 mm Hg for the saline + acetic acid group, 100.7 F 3.4 mm Hg for the CTOP + acetic acid group and 99.1 F 2.5 mm Hg for the nor-BNI + acetic acid group. Baseline heart rate was 370 F 6 bpm for the saline + acetic acid group, 376 F 15 bpm for the CTOP + acetic acid group and 359 F 8 bpm for the nor-BNI + acetic acid group.
[ F(30,270) = 4.9, P < 0.001] and heart rate [ F(30,225) = 5.9, P < 0.001] significantly (Fig. 5) although it did not influence arterial pressure or heart rate when injected into the vlPAG of control rats that were given saline instead of 5% acetic acid ip (Fig. 5). As shown previously for lidocaine, bilateral naltrindole (2 nmol; 0.5 Al) injections targeted on the dlPAG failed to influence arterial pressure (saline = 13.8 F 1.9 mm Hg; naltrindole = 12.9 F 2.6 mm Hg) or heart rate (saline = 26 F 6 bpm; naltrindole = 25 F 2 bpm) 2 min after ip acetic acid injection. To verify the receptor specificity of the response, we tested whether bilateral injection of the mu receptor antagonist CTOP (5 nmol/cannula) or the kappa receptor antagonist nor-BNI (3 nmol/cannula) into the vlPAG would
Fig. 7. Location of injection sites in the PAG. (Panel A) Schematic representation of coronal sections through the caudal PAG at 8.0 mm (left), 8.3 mm (center) and 8.7 mm (right) from bregma illustrating the location of lidocaine, cobalt chloride, naltrindole, CTOP and nor-BNI injection sites. (Panel B) Photomicrograph of a section through the caudal PAG (8.3 mm from bregma) illustrating the location of 0.5 Al India ink marking the site of a lidocaine injection. The boundaries of the PAG are from Paxinos and Watson [44]. Scale bar = 1.0 mm. (Panel C) Schematic representation of a 0.5 Al India ink injection illustrated in panel B. Aq = Sylvian aqueduct.
S. Cavun et al. / Brain Research 1019 (2004) 237–245
tegmentum (Fig. 7B and C). A smaller injection volume, 0.2 Al, perfused a proportionately smaller area, approximately 0.4 mm in diameter (data not shown).
4. Discussion These data show that lidocaine injection into the vlPAG prevents the hypotension and bradycardia evoked by visceral nociception in halothane-anesthetized rats. Nociceptive hypotension was also blocked by cobalt chloride, which indicates that synaptic transmission within the vlPAG is necessary for visceral nociception to evoke a depressor response. Lidocaine inactivation of the dlPAG was ineffective, consistent with evidence that visceral nociception induces c-fos expression in the vlPAG to a substantially greater extent than in the dlPAG [11]. Ventrolateral PAG inactivation also attenuated the hypotension caused by deep somatic nociception although it did not eliminate the response completely. Nevertheless, the vlPAG apparently plays a significant role in the cardiovascular effects of both visceral and deep somatic pain. These findings are consistent with reports that visceral and deep somatic nociception cause hypotension and bradycardia whereas superficial somatic pain produces hypertension and tachycardia [11,26]. It is an oversimplification to assume that all types of visceral pain produce the same effect on cardiovascular function, however. Mechanical distension of the stomach [28], colon [27] or bladder [41] raises arterial pressure and heart rate, for example, whereas distension of the small intestine [10,38] or renal pelvis [4] causes arterial pressure to fall. It is important to note these studies were conducted in anesthetized animals, however, and anesthesia can influence cardiovascular reflexes. Colorectal distension evokes a depressor response in pentobarbital-anesthetized rats [27,40], for example, but raises arterial pressure and heart rate in conscious animals [15,40]. The specific type of anesthetic is also an important factor. Colorectal distension also lowers arterial pressure in rats anesthetized with a-chloralose or urethane but causes hypertension under halothane or ketamine anesthesia as it does in conscious animals [39,40]. Ideally, this type of investigation should be conducted in conscious animals but this raises ethical concerns for models of prolonged or inescapable pain. For this reason, the present experiments were performed in anesthetized animals but this raises the possibility that the noxious stimuli we used may evoke a depressor response only under halothane-anesthesia. To test this possibility, Keay et al. [26] administered formalin im to rats under brief (1– 2 min) carbon dioxide narcosis. Carbon dioxide caused a transient pressor response but arterial pressure subsequently fell 15 – 20 mm Hg and remained below baseline levels for the rest of the 20 min experiment. The magnitude and duration of the response was comparable to that evoked in halothane-
243
anesthetized rats [11] and is consistent with the present data. Intramuscular formalin administration thus appears to lower arterial pressure to a comparable extent in halothane-anesthetized and unanesthetized rats. Nevertheless, it is important to emphasize that, without conducting parallel experiments in conscious animals, we cannot rule out the possibility that halothane-anesthesia may have influenced the results reported here. We also found the vasodepression caused by visceral nociception to be inhibited by naltrindole, a delta opioid receptor antagonist, but not by CTOP or nor-BNI, which block mu and kappa receptors. The ineffectiveness of CTOP and nor-BNI is unlikely to be due to an inadequate drug dose because Kd values for [3H]-CTOP (0.16 nM) [18] and [3H]nor-BNI (0.10 nM) [33] binding to rat brain membranes are orders of magnitude lower than the concentrations of CTOP (10 mM) and nor-BNI (6 mM) microinjected into the vlPAG in the present study. CTOP and nor-BNI also prevent stressinduced analgesia at doses that are substantially lower than used in this study [34,53]. The Kd for [3H]-naltrindole binding to delta opioid receptors (0.08 nM) [14] is also considerably lower than the naltrindole concentration used here (2 mM) and previous studies have shown that naltrindole injection into the vlPAG inhibits analgesia evoked from the amygdala at doses (0.3 –8 nmol) [43,48] similar to the dose used in the present study. These considerations support the conclusion that delta, but not mu or kappa, receptors mediate the fall in arterial pressure and heart rate evoked by visceral nociception. Histological localization of injection sites revealed that many of the lidocaine, cobalt chloride and naltrindole injections spread beyond the borders of the vlPAG into the adjacent midbrain tegmentum. Earlier mapping studies also showed that excitatory amino acids lower arterial pressure when injected into the midbrain tegmentum ventrolateral to the vlPAG of halothane-anesthetized rats [3,24] and precollicular decerebrated cats [6]. Excitatory amino acid injection into this region also produces the same behavioral response, quiescence and hyporeactivity, as injections restricted to the vlPAG and both visceral and deep somatic nociception induce c-fos expression in neurons both within and ventrolateral to the vlPAG [3]. Importantly, delta receptor agonist injection evokes a depressor response both within the vlPAG and in the tegmental area ventrolateral to the vlPAG [24]. The midbrain region that mediates the depressor response to visceral nociception thus extends beyond the boundaries of the vlPAG. The present findings extend earlier evidence that lidocaine [8] and naltrindole [9] administration into the vlPAG delays the onset and reduces the magnitude of hemorrhagic hypotension. Earlier investigations had shown the decompensatory phase of hemorrhage to be attenuated by intraventricular administration of naloxone [47] and delta receptor antagonists [30]. We found that naltrindole injection into the vlPAG attenuated the decompensatory phase of hemorrhage at doses ranging from 0.2 to 20 nmol; the dose used in the
244
S. Cavun et al. / Brain Research 1019 (2004) 237–245
present study, 2 nmol, was approximately the ED50 [9]. Mu and kappa receptor antagonists were ineffective. These findings suggest that visceral nociception and severe blood loss lower arterial pressure through the same, or a similar, anatomical pathway through the vlPAG in which delta opioid receptors are an important component. Several lines of evidence suggest that nociceptive and hemorrhagic hypotension may be mediated by a common anatomical pathway similar to the descending pathway that modulates pain perception [35]. Ventrolateral PAG neurons innervate the nucleus raphe magnus, an important component of the descending pain control pathway [35], and a more caudal midline medullary region encompassing portions of the raphe obscurus and raphe pallidus nuclei [19]. Activation of neurons in the caudal midline medulla with excitatory amino acids evokes a fall in arterial pressure and heart rate [13,19,51]. Conversely, inactivation of this region with lidocaine or cobalt chloride prevents the fall in arterial pressure caused by hemorrhage [20] without affecting resting blood pressure or baroreceptor reflexes [20,45]. Serotonergic neurons in the nucleus raphe magnus, raphe pallidus and raphe obscurus directly innervate the sympathoexcitatory region of the rostral ventrolateral medulla (RVLM) [2,56] and blockade of serotonin-1A receptors in the RVLM prevents the hypotension caused by vlPAG stimulation [1] and severe hemorrhage [16]. These data suggest that vlPAG activation lowers arterial pressure through a descending pathway that includes the rostral [52] or caudal [13,16,20,50] midline medulla and RVLM. Alternatively, vlPAG neurons innervate the RVLM directly [2,5,7,50] and there is evidence that deep somatic nociception evokes c-fos expression in vlPAG neurons that innervate the RVLM independent of the medullary depressor region [25]. In summary, these data support the hypothesis that visceral nociception and severe blood loss cause hypotension and bradycardia by activating a descending pathway through the vlPAG in which delta opioid receptors play a significant role. In some respects, this pathway is similar to the descending pain control pathway although pain control is mediated primarily by mu, rather than delta, opioid receptors in the vlPAG [54]. Hypotension and bradycardia appear to be part of a broader, integrated response to inescapable pain and severe blood loss that includes behavioral quiescence, hyporeactivity to environmental stimuli and opioid mediated analgesia [3,36,53]. Presumably, the hypotension, behavioral inactivity and profound analgesia that develops when the vlPAG is activated serves to reduce blood loss during severe hemorrhage and prevent exacerbation of deep tissue injury.
Acknowledgements This research was supported by a grant from the Office of Naval Research (N00014-98-1-02-0249).
References [1] M. Bago, C. Dean, Sympathoinhibition from ventrolateral periaqueductal gray mediated by 5-HT1A receptors in the RVLM, Am. J. Physiol. 280 (2001) R976 – R984. [2] M. Bago, L. Marson, C. Dean, Serotonergic projections to the rostroventrolateral medulla from midbrain and raphe nuclei, Brain Res. 945 (2002) 249 – 258. [3] R. Bandler, K.A. Keay, Columnar organization in the midbrain periaqueductal gray and the integration of emotional expression, Prog. Brain Res. 107 (1996) 285 – 300. [4] H. Brasch, G. Zetler, Caerulein and morphine in a model of visceral pain. Effects on the hypotensive response to renal pelvis distension in the rat, Naunyn-Schmiedeberg’s Arch. Pharmacol. 319 (1982) 161 – 167. [5] A.A. Cameron, I.A. Khan, K.N. Westlund, W.D. Willis, The efferent projections of the periaqueductal gray in the rat: a Phaseolus vulgarisleucoagglutinin study. II. Descending projections, J. Comp. Neurol. 351 (1995) 585 – 601. [6] P. Carrive, The periaqueductal gray and defensive behavior: functional representation and neuronal organization, Behav. Brain Res. 58 (1993) 27 – 47. [7] P. Carrive, R. Bandler, Viscerotopic organization of neurons subserving hypotensive reactions within the midbrain periaqueductal grey: a correlative functional and anatomical study, Brain Res. 541 (1991) 206 – 215. [8] S. Cavun, W.R. Millington, Evidence that hemorrhagic hypotension is mediated by the ventrolateral periaqueductal gray, Am. J. Physiol. 281 (2001) R747 – R752. [9] S. Cavun, G.E. Resch, M. Rapacon-Baker, A. Evec, W.R. Millington, Blockade of delta opioid receptors in the ventrolateral periaqueductal gray region inhibits the fall in arterial pressure evoked by hemorrhage, J. Pharmacol. Exp. Ther. 297 (2001) 612 – 619. [10] S.J. Clark, R.L. Follenfant, T.W. Smith, Evaluation of opioid-induced antinociceptive effects in anaesthetized and conscious animals, Br. J. Pharmacol. 95 (1988) 275 – 283. [11] C.I. Clement, K.A. Keay, B.K. Owler, R. Bandler, Common patterns of increased and decreased Fos expression in midbrain and pons evoked by noxious deep somatic and noxious visceral manipulations in the rat, J. Comp. Neurol. 366 (1996) 495 – 515. [12] C.I. Clement, K.A. Keay, K. Podzebenko, B.D. Gordon, R. Bandler, Spinal sources of noxious visceral and noxious deep somatic afferent drive onto the ventrolateral periaqueductal gray of the rat, J. Comp. Neurol. 425 (2000) 323 – 344. [13] M.J. Coleman, R.A. Dampney, Powerful depressor and sympathoinhibitory effects evoked from neurons in the caudal raphe pallidus and obscurus, Am. J. Physiol. 268 (1995) R1295 – R1302. [14] P.C. Contreras, L. Tam, E. Drower, M.F. Rafferty, [3H]Naltrindole: a potent and selective ligand for labeling y-opioid receptors, Brain Res. 604 (1993) 160 – 164. [15] R.M. Danzebrink, G.F. Gebhart, Antinociceptive effects of intrathecal adrenoceptor agonists in a rat model of visceral nociception, J. Pharmacol. Exp. Ther. 253 (1990) 698 – 705. [16] C. Dean, M. Bago, Renal sympathoinhibition mediated by 5-HT1A receptors in the RVLM during severe hemorrhage in rats, Am. J. Physiol. 282 (2002) R122 – R130. [17] A. Depaulis, K.A. Keay, R. Bandler, Quiescence and hyporeactivity evoked by activation of cell bodies in the ventrolateral midbrain periaqueductal gray of the rat, Exp. Brain Res. 99 (1994) 75 – 83. [18] K.N. Hawkins, R.J. Knapp, G.K. Lui, K. Gulya, W. Kazmierski, Y.P. Wan, J.T. Pelton, V.J. Hruby, H.I. Yamamura, [3H]-[H-D-Phe-CysTyr-D-Trp-Orn-Thr-Pen-Thr-NH2] ([3H]CTOP), a potent and highly selective peptide for mu opioid receptors in rat brain, J. Pharmacol. Exp. Ther. 248 (1989) 73 – 80. [19] L.A. Henderson, K.A. Keay, R. Bandler, The ventrolateral periaqueductal gray projects to caudal brainstem depressor regions: a func-
S. Cavun et al. / Brain Research 1019 (2004) 237–245
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32] [33]
[34]
[35] [36] [37]
[38]
tional-anatomical and physiological study, Neuroscience 82 (1998) 201 – 221. L.A. Henderson, K.A. Keay, R. Bandler, Caudal midline medulla mediates behaviorally-coupled but not baroreceptor-mediated vasodepression, Neuroscience 98 (2000) 779 – 792. K.A. Keay, R. Bandler, Deep and superficial noxious stimulation increases Fos-like immunoreactivity in different regions of the midbrain periaqueductal grey of the rat, Neurosci. Lett. 154 (1993) 23 – 26. K.A. Keay, R. Bandler, Distinct central representations of inescapable and escapable pain: observations and speculations, Exp. Physiol. 87 (2002) 275 – 279. K.A. Keay, C.I. Clement, B. Owler, A. Depaulis, R. Bandler, Convergence of deep somatic and visceral nociceptive information onto a discrete ventrolateral midbrain periaqueductal gray region, Neuroscience 61 (1994) 727 – 732. K.A. Keay, L.J. Crowfoot, N.S. Floyd, L.A. Henderson, M.J. Christie, R. Bandler, Cardiovascular effects of microinjections of opioid agonists into the ‘depressor region’ of the ventrolateral periaqueductal gray region, Brain Res. 762 (1997) 61 – 71. K.A. Keay, Q.F. Li, R. Bandler, Muscle pain activates a direct projection from ventrolateral periaqueductal gray to rostral ventrolateral medulla in rats, Neurosci. Lett. 290 (2000) 157 – 160. K.A. Keay, C.I. Clement, W.M. Matar, D.J. Heslop, L.A. Henderson, R. Bandler, Noxious activation of spinal or vagal afferents evokes distinct patterns of fos-like immunoreactivity in the ventrolateral periaqueductal gray of unanaesthetized rats, Brain Res. 948 (2002) 122 – 130. C.M. Kozlowski, A. Green, D. Grundy, F.M. Boissonade, C. Bountra, The 5-HT3 receptor antagonist alosetron inhibits the colorectal distention induced depressor response and spinal c-fos expression in the anaesthetised rat, Gut 46 (2000) 474 – 480. P. Li, K. Rowshan, M. Crisostomo, S.C. Tjen-A-Looi, J.C. Longhurst, Effect of electroacupuncture on pressor reflex during gastric distension, Am. J. Physiol. 283 (2002) R1335 – R1345. T.A. Lovick, Integrated activity of cardiovascular and pain regulatory systems: role in adaptive behavioral responses, Prog. Neurobiol. 40 (1993) 631 – 644. J. Ludbrook, S. Ventura, The decompensatory phase of acute hypovolaemia in rabbits involves a central y1-opioid receptor, Eur. J. Pharmacol. 252 (1994) 113 – 116. B.M. Lumb, Inescapable and escapable pain is represented in distinct hypothalamic-midbrain circuits: specific roles for A y- and C-nociceptors, Exp. Physiol. 87 (2002) 281 – 286. J.G. Malpeli, Reversible inactivation of subcortical sites by drug injection, J. Neurosci. Methods 86 (1999) 119 – 128. ´ . Ma´rki, F. O ¨ tvo¨s, G. To´th, S. Hosztafi, A. Borsodi, Tritiated kappa A receptor antagonist norbinaltorphimine: synthesis and in vitro binding in three different tissues, Life Sci. 66 (2000) 43 – 49. L. Menendez, F. Andres-Trelles, A. Hidalgo, A. Baamonde, Involvement of spinal n opioid receptors in a type of footshock induced analgesia in mice, Brain Res. 611 (1993) 264 – 271. M.J. Millan, Descending control of pain, Prog. Neurobiol. 66 (2002) 355 – 474. P.E. Molina, Endogenous opioid analgesia in hemorrhagic shock, J. Trauma 54 (5 Suppl) (2003) S126 – S132. M.M. Morgan, P. Carrive, Activation of the ventrolateral periaqueductal gray reduces locomotion but not mean arterial pressure in awake, freely moving rats, Neuroscience 102 (2001) 905 – 910. H.E. Moss, G.J. Sanger, The effects of granisetron, ICS 205-930 and
[39] [40]
[41]
[42]
[43]
[44] [45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54] [55]
[56]
245
odansetron on the visceral pain reflex induced by duodenal distension, Br. J. Pharmacol. 100 (1990) 497 – 501. T.J. Ness, Models of visceral nociception, ILAR J. 40 (1999) 119 – 128. T.J. Ness, G.F. Gebhart, Colorectal distension as a noxious visceral stimulus: physiologic and pharmacologic characterization of pseudaffective reflexes in the rat, Brain Res. 450 (1988) 153 – 169. T.J. Ness, A. Lewis-Sides, P. Castroman, Characterization of pressor and visceromotor reflex responses to bladder distention in rats: sources of variability and effect of analgesics, J. Urol. 165 (2001) 968 – 974. S. Ohtori, K. Takahashi, T. Chiba, Y. Takahashi, M. Yamagata, H. Sameda, H. Moriya, Fos expression in the rat brain and spinal cord evoked by noxious stimulation to low back muscle and skin, Spine 25 (2000) 2425 – 2430. Z.W. Pavlovic, M.L. Cooper, R.J. Bodnar, Opioid antagonists in the periaqueductal gray inhibit morphine and h-endorphin analgesia elicited from the amygdala of rats, Brain Res. 741 (1996) 13 – 26. G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1998. J.R. Potas, R.A. Dampney, Sympathoinhibitory pathway from caudal midline medulla to RVLM is independent of baroreceptor reflex pathway, Am. J. Physiol. 284 (2003) R1071 – R1078. L. Rodella, R. Rezzani, M. Gioia, G. Tredici, R. Bianchi, Expression of Fos immunoreactivity in the rat supraspinal regions following noxious visceral stimulation, Brain Res. Bull. 47 (1998) 357 – 366. J.C. Schadt, J. Ludbrook, Hemodynamic and neurohumoral responses to acute hypovolemia in conscious mammals, Am. J. Physiol. 260 (1991) H305 – H318. R. Shane, J. Acosta, G.C. Rossi, R.J. Bodnar, Reciprocal interactions between the amygdala and ventrolateral periaqueductal gray in mediating of Q/N1 – 17 -induced analgesia in the rat, Brain Res. 980 (2003), 57 – 70. B.K. Taylor, M.A. Peterson, A.I. Basbaum, Continuous intravenous infusion of naloxone does not change behavioral, cardiovascular, or inflammatory responses to subcutaneous formalin in the rat, Pain 69 (1997) 171 – 177. E.J. Van Bockstaele, G. Aston-Jones, V.A. Pieribone, M. Ennis, M.T. Shipley, Subregions of the periaqueductal gray topographically innervate the rostral ventral medulla in the rat, J. Comp. Neurol. 309 (1991) 305 – 327. A.J. Verberne, D.M. Sartor, A. Berke, Midline medullary depressor responses are mediated by inhibition of RVLM sympathoexcitatory neurons in rats, Am. J. Physiol. 276 (1999) R1054 – R1062. W.H. Wang, T.A. Lovick, The inhibitory effect of the ventrolateral periaqueductal gray matter on neurones in the rostral ventrolateral medulla involves a relay in the medullary raphe nuclei, Exp. Brain Res. 94 (1993) 295 – 300. C.P. Wiedenmayer, G.A. Barr, A Opioid receptors in the ventrolateral periaqueductal gray mediate stress-induced analgesia but not immobility in rat pups, Behav. Neurosci. 114 (2000) 125 – 136. T.L. Yaksh, N.R.F. Al-Rodhan, T.S. Jensen, Sites of action of opiates in production of analgesia, Prog. Brain Res. 77 (1988) 371 – 394. M.H. Yoon, T.L. Yaksh, The effect of intrathecal gabapentin on pain behavior and hemodynamics on the formalin test in the rat, Anesth. Analg. 89 (1999) 434 – 439. A. Zagon, Internal connections in the rostral ventromedial medulla of the rat, J. Auton. Nerv. Syst. 53 (1995) 43 – 56.