- Email: [email protected]
The 5-HT3 receptor facilitates at-level mechanical allodynia following spinal cord injury
Pain 110 (2004) 259–268 www.elsevier.com/locate/pain
The 5-HT3 receptor facilitates at-level mechanical allodynia following spinal cord injury Mark A. Oatwaya,b, Yuhua Chena, Lynne C. Weavera,b,* a
The Spinal Cord Injury Team, Biotherapeutics Research Group, Robarts Research Institute, P.O. Box 5015, 100 Perth Drive, London, Ont., Canada N6A 5K8 b The Graduate Program in Neuroscience, The University of Western Ontario, London, Ont., Canada N6A 5K8 Received 23 January 2004; received in revised form 15 March 2004; accepted 29 March 2004
Abstract Spinal cord injury (SCI) results in the development of mechanical allodynia immediately rostral to the lesion site, within the dermatome border of normal sensation and sensory loss (at-level mechanical allodynia). We propose that an observed threefold increase in serotonergic fibre immunoreactivity within spinal segments corresponding to these allodynic dermatomes facilitates the maintenance of chronic neuropathic pain via activation of the 5-HT3 receptor (5-HT3-R). Serotonin (5-HT), the non-selective 5-HT1/5-HT2 receptor antagonist, methysergide, the 5-HT3-R agonist, m-chlorophenylbiguanide (m-CPBG) or the 5-HT3-R antagonist, ondansetron were intrathecally administered five weeks following SCI in rats. Ondansetron produced a robust, long-term reduction of at-level mechanical allodynia, while m-CPBG exacerbated allodynia. Exogenous 5-HT transiently reduced at-level mechanical allodynia. This effect was opposed by methysergide, which enhanced mechanical allodynia. Co-administration of 5-HT and ondansetron produced a short-lasting partial summation of effects, further decreasing mechanical allodynia while co-administration of methysergide attenuated the anti-allodynic effect of ondansetron. Depletion of spinal 5-HT via 5,7-dihydroxytryptamine (5,7-DHT) resulted in decreased at-level mechanical allodynia. The reduction of allodynia by ondansetron was lost following 5,7-DHT administration, suggesting that reduced allodynia following intrathecal ondansetron is via blockade of 5-HT-induced excitation of the 5-HT3-R. These results suggest that increased 5-HT fibre density immediately rostral to the SCI lesion site could have transient effects to reduce mechanical allodynia via actions at 5-HT1 and/or 5-HT2 receptors. However, the more long-lasting effects of this enhanced serotonergic input may facilitate chronic, at-level allodynia via the 5-HT3-R. q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: Spinal cord injury; Serotinin; Neuropathic pain; 5-HT3 receptor
1. Introduction The development and maintenance of chronic pain following spinal cord injury (SCI) is a significant problem that is often refractory to a majority of clinical interventions (Yezierski, 1996; Warms et al., 2002). Neuropathic pain varies in quality and intensity amongst patients and is classified according to anatomical location (Vierck, Jr. et al., 2000). At-level neuropathic pain occurs at the sensory transitional zone, within a band of two to four spinal segments rostral to the lesion level (Siddall and Loeser, 2001). Animal models of spinal ischemia and injury have * Corresponding author. Biotherapeutics Research Group, Robarts Research Institute, P.O. Box 5015, 100 Perth Dr., London, Ont., Canada N6A 5K8. Tel.: þ 519-663-3776; fax: þ519-663-3789. E-mail address: [email protected] (L.C. Weaver).
been successful in reproducing at-level allodynia (Hao et al., 1991; Yezierski, 2000). Although such models have suggested that chronic alterations in spinal and supraspinal endogenous opioid systems may contribute to post-SCI allodynia (Weisenfeld-Hallin et al., 1997), SCI patients with neuropathic pain generally respond poorly to opioids (Parisod et al., 2003). A further understanding of the morphological alterations above the lesion level following SCI, primarily within descending modulatory pain pathways, is required to characterize alternative pharmacological targets. The loss of descending serotonergic innervation caudal to the SCI lesion has been causally linked to the development of allodynia below the lesion level (Hains et al., 2002). Promotion of serotonergic actions via transplantation of 5-HT-secreting cells (Hains et al., 2001), or administration of a selective serotonin reuptake
0304-3959/$20.00 q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2004.03.040
260
M.A. Oatway et al. / Pain 110 (2004) 259–268
inhibitor (Horiuchi et al., 2002) transiently decreases belowlevel neuropathic pain. However, previous work has shown a significant increase in serotonin (5-HT) fibre density in spinal segments immediately rostral to the lesion level that corresponds with the dermatomes from which at-level mechanical allodynia is elicited (Bruce et al., 2002; Inman and Steward, 2003). The maintenance of neuropathic pain has been attributed to descending facilitation resulting from plastic changes within the rostral ventromedial medulla (RVM) (Ossipov et al., 2001). Chronic noxious stimuli activate RVM serotonergic cells, causing increases in spinal 5-HT turnover and contributing to descending facilitation (Zhuo and Gebhart, 1997; Mason, 1999). The influence of descending 5-HT on pain transmission within the dorsal horn is dependent on the class of 5-HT receptor subtype activated (Millan, 2002). Spinal 5-HT receptors biphasically inhibit or facilitate ascending pain transmission at the second-order neuron. Whereas the 5-HT1 and 5-HT2 receptor subtypes bear the highest affinity for the endogenous ligand, and are believed to exert an overall antinociceptive action (Hamon and Bourgoin, 1999), models of persistent pain have suggested a role for 5-HT3 receptor (5-HT3-R) activation in the maintenance of pain (Green et al., 2000; Zeitz et al., 2002). Furthermore, the spinal 5-HT3-R mediate a part of the excitatory projections within facilitatory spinal-brainstem-spinal loops which are enhanced during chronic pain conditions (Suzuki et al., 2002). In the present study, we have tested the hypothesis that, following SCI, the persistence of at-level mechanical allodynia is maintained via excitation of spinal 5-HT3-R corresponding to the increased serotonergic fibre density immediately rostral to the lesion site. Using the clinicallyrelevant clip-compression model of SCI (Rivlin and Tator, 1977), we determined the effects of an intrathecallyadministered 5-HT3-R antagonist on at-level mechanical allodynia five weeks post-SCI when given alone, in combination with non-selective 5-HT receptor agonists and antagonists and following neurotoxin-induced spinal 5-HT depletion.
2. Materials and methods 2.1. Animals and surgical procedures Male Wistar rats (Charles River, Ontario, Canada) weighing 220 –350 g were housed individually with free access to food and water. All experimental procedures were approved by the University of Western Ontario Animal Care Committee in accordance with the Canadian Guide to Care and Use of Experimental Animals. Experiments also adhered to the guidelines of the Committee for Research and Ethical Issues of the International Association for the Study of Pain (Zimmermann, 1983). Prior to surgery, animals were medicated with diazepam (3.5 mg/kg, ip;
Sabex 2002 Inc., Boucherville, PQ) and atropine (0.05 mg/ kg, sc; Sigma-Aldrich, St Louis, MO) as described previously (Weaver et al., 2001) and anesthetised with 2.0% halothane in oxygen. A dorsal laminectomy was performed, removing the dorsal processes of the 10th and 11th thoracic vertebrae and exposing the 12th – 13th thoracic segmental junction. Using a modified aneurysm clip calibrated to generate a force of 35 g, the cord was compressed dorsoventrally for 60s producing a moderate, incomplete SCI. Care was taken to place the clip around the cord without disrupting the dura mater or adjacent dorsal roots. The muscle and skin were closed in layers and animals recovered under a heat lamp. Post-operative care was provided using antibiotics (Baytril 50 mg/ml, sc; Bayer Inc., Toronto, ON) and subcutaneous hydration with 0.9% normal saline daily for 3 days. Animals were housed on extra bedding to prevent the development of pressure and friction sores. Urinary bladders were manually emptied by abdominal compression twice daily until spontaneous voiding returned at 7– 10 days. 2.2. Behavioural procedures 2.2.1. Assessment of at-level mechanical allodynia The development of mechanical allodynia immediately rostral to and at the level of the SCI lesion was tested on the dorsal surface of the trunk. Using a Modified SemmesWeinstein monofilament (Stoelting Co., Wood Dale, IL), calibrated to generate a force of 20 mN, the dermatome areas corresponding to the 9th – 11th thoracic spinal segments were stimulated. Prior to testing, rats were acclimated to a plexiglass box similar to their home cage for 20 min in a designated testing room. A testing session consisted of ten stimulations to random points within the dorsal trunk area. Each stimulus lasted 3s and was separated by a 5s interim period. The number of avoidance responses elicited from ten stimulations was then tabulated. An avoidance response was defined by one of the following behaviours: trunk shakes, jumping or escaping to another area of the cage, vocalization or abnormal aggressive behaviour such as biting or swatting the filament. Rats were determined to exhibit allodynia if they displayed an increased number of avoidance behaviours to the innocuous stimulus post-injury when compared to pre-injury behaviour. A pre-injury baseline response to the filament was established by determining the mean number of avoidance responses during the week prior to injury. Testing resumed seven days following injury, with two testing sessions per week for the subsequent four post-injury weeks. 2.2.2. Intrathecal injections At post-SCI day 28, rats were fixed in a stereotaxic frame under general anaesthesia with 2.0% halothane in oxygen via a facemask. Intrathecal (i.t.) catheters consisting of 8.0 cm of Micro-Renathane tubing (0.025 in. O.D. £ 0.012 in. I.D.; Braintree Scientific Inc., Braintree, MA) fixed to
M.A. Oatway et al. / Pain 110 (2004) 259–268
5.3 cm of 32-gauge tubing (Recathco Inc, Allison Park, PA) were inserted through the atlanto-occipital membrane. The tip of the 32-gauge tubing was threaded along the spinal cord to terminate at the 8th –10th thoracic segment. The Micro-Renathane tubing connection exited the skin and was fixed to the back of the head. Following insertion, catheters were flushed with 20 ml of 0.9% saline using an insulin syringe (Becton Dickinson and Co., Franklin Lakes, NJ) and flushed each subsequent day until behavioural testing resumed. Rats were allowed to recover for three days following catheter insertion before testing resumed. Catheter positions were verified following perfusion by performing a full cervical and thoracic laminectomy. In some rats, i.t. administration of luxol fast blue dye immediately prior to perfusion verified the dispersion of fluids through the catheter and along the spinal cord. In most cases, the solutions reached the thoracic spinal cord rostral, but not into or caudal to the lesion site. 2.2.3. Treatment protocols Drug interventions commenced during the 5th post-SCI week, when at-level allodynia has reliably developed (Bruce et al., 2002). Prior to drug administration, rats were acclimated to the testing cages for 20 min. Each rat was administered an i.t. injection of drug compound dissolved in 0.9% saline or saline alone (10 ml/rat) followed by a saline flush (20 ml/rat). Rats ðn ¼ 5/group) received either serotonin hydrochloride (1, 10, 20 mg, Sigma-Aldrich), the 5-HT3-R agonist, m-chlorophenylbiguanide (m-CPBG; 1, 10, 20 mg, Sigma-Aldrich), the 5-HT3-R antagonist, ondansetron (1, 10, 20 mg, GlaxoSmithKline, Research Triangle Park, NC), the 5-HT1/5-HT2 receptor antagonist, methysergide maleate salt (30 mg, Research Biochemicals International, Natick, MA) or saline. Mechanical allodynia was assessed at 5, 15, 30, 45 and 60 min following injections. At each time point, rats received 10 stimulations to the dorsal trunk as described above. Co-administrative paradigms were designed to characterize the involvement of multiple serotonergic receptor subtypes. In each case, one drug was administered at time 0 min followed by the assessment of mechanical allodynia 5 min later. The second drug was then administered at 10 min and allodynia testing resumed at 15, 30, 45 and 60 min. 2.2.4. 5,7-Dihydroxytryptamine administration Endogenous spinal 5-HT was depleted with the neurotoxin, 5,7-dihydroxy-tryptamine creatinine sulphate salt (5,7-DHT; Sigma-Aldrich) administered intrathecally. Three days following i.t. catheter insertion, rats were pretreated with desipramine hydrochloride (25 mg/kg, i.p.; Sigma-Aldrich) to prevent uptake of the neurotoxin into noradrenergic neurons. After 45 min, rats received either 0.9% saline ðn ¼ 6Þ or 60 mg of 5,7-DHT ðn ¼ 7Þ in a volume of 20 ml followed by a 20 ml saline flush. This dose of 5,7-DHT has been reported to deplete endogenous spinal
261
5-HT significantly (Sawynok and Reid, 1994). All drug assessments were completed prior to day nine following 5,7-DHT administration to avoid the complication of supersensitivity development after this time (Howe and Yaksh, 1982). 2.3. Tissue preparation and processing Seven days following i.t. 5,7-DHT administration, rats were deeply anaesthetized with urethane (2.5 g/kg, i.p; Sigma Chemical Co.) and perfused as described previously (Bruce et al., 2002). To assess the density and distribution of spinal 5-HT fibres immediately rostral to lesion site following 5,7-DHT administration, spinal cord tissue was processed for 5-HT immunoreactivity. The spinal cord (9th– 11th thoracic segments) was cut with a cryostat into 30 mm transverse sections and floated in 0.1 M phosphate buffer. Floating sections were blocked for 60 min in TPBS-X containing 10% normal goat serum (NGS) and incubated in a rabbit anti-5-HT antibody (1:20,000; Immunostar, Hudson, WI) in 1% NGS for 72 h. The tissue was incubated in biotin-conjugated goat anti-rabbit antibody (1:250; Jackson Immunoresearch, Westgrove, PA) for 12 h and incubated in extravidin-peroxidase (Sigma-Aldrich) for 4 h. Immunoreactivity was revealed using a nickelenhanced process and diaminobenzidine as the chromogen. Several sections were processed without the primary antibody incubation step to verify a lack of non-specific staining. 2.4. Data analysis In assessing the development of mechanical allodynia, values are represented as mean number of avoidance responses to ten stimulations ^ standard error mean (SEM). Dose-response results were subjected to a repeated measures analysis of variance and a Fischer’s protected t-test. All drug treatments were compared against the salinetreated group unless otherwise noted. Differences were considered significant at P , 0:05:
3. Results 3.1. Development of at-level mechanical allodynia following SCI At-level mechanical allodynia was assessed during the second post-SCI week (day 7) until the middle of the fourth post-SCI week (day 26). During this time, the presence of mechanical allodynia was tested on the dorsal trunk immediately rostral and proximal to the lesion site. The mechanical stimulus was determined to be innocuous, as uninjured rats did not respond to stimulation of the dorsal trunk one-week prior to injury (Fig. 1). During the second post-injury week, SCI rats responded significantly more
262
M.A. Oatway et al. / Pain 110 (2004) 259–268
Fig. 1. The development of mechanical allodynia immediately rostral to the SCI lesion site. Testing sessions consisted of ten stimulations with an innocuous, monofilament one week prior to injury (baseline) and at the second, third and fourth post-SCI weeks. Each data point represents the mean ^ SEM number of avoidance responses made to 10 stimulations. The number of avoidance responses made to dorsal trunk stimulation significantly increased over time following SCI ðn ¼ 6Þ suggesting the development of at-level mechanical allodynia. *, P , 0:05 compared to pre-SCI mean responses; þ, P , 0:05 compared to avoidance responses made during the 2nd post-SCI week.
often (2.5 ^ 0.4 out of 10 stimulations) when compared to uninjured rats ðP , 0:05Þ: Previous results from our laboratory showed that sham-injured rats do not exhibit signs of mechanical allodynia simply due to laminectomyinduced inflammation during the second post-injury week (Bruce et al., 2002). The number of avoidance responses increased further during the third week (5.8 ^ 0.8) and remain stable during the fourth week post-SCI (6.4 ^ 0.7). At both of these time points, SCI rats responded significantly more to the innocuous stimulus when compared to the number of responses made prior to SCI and to the number during the second post-injury week ðP , 0:05Þ: In a follow-up study, we observed that the number of avoidance responses at four weeks remains constant for twelve weeks following SCI (data not shown). We therefore used the fourth week value as a baseline comparison for subsequent drug treatments. 3.2. A 5-HT3-R antagonist reduces at-level mechanical allodynia following SCI Intrathecal administration of various serotonergic agents was performed during the 5th post-injury week, a time when all rats had developed at-level mechanical allodynia. Baseline values on all figures represent the number of avoidance responses made during the 4th post-SCI week prior to i.t. catheter insertion. Placement of the catheter did not alter avoidance response frequencies as all animals responded at three days post-catheter placement in a manner similar to their responses prior to this surgery (data not shown).
Fig. 2. Effects of acute 5-HT3-R stimulation (m-CPBG) or blockade (ondansetron) on at-level mechanical allodynia five weeks post-SCI. Baseline represents the number of avoidance responses made during the fourth post-SCI week, prior to intrathecal catheter insertion. (A) m-CPBG (1, 10 and 20 mg), (B) ondansetron (1, 10 and 20 mg) or saline was administered at 0 min and mechanical allodynia was assessed for the subsequent 60 min. (C) The effect of alternating the order of injection of m-CPBG and ondansetron on at-level mechanical allodynia. m-CPBG (20 mg) or ondansetron (20 mg) was administered at 0 min and mechanical allodynia was assessed at 5 min. The second agent was then administered at 10 min (indicated by 2nd injection). The values shown represent mean number of avoidance responses to ten stimulations ^ SEM (n ¼ 5/group). *, P , 0:05 compared to saline-treated controls.
Saline-injected rats reacted to the stimulation consistently with an average of 6.2 ^ 0.04 responses over the 1 h testing period (Fig. 2). Administration of the 5-HT3-R agonist, m-CPBG, at doses of 10 and 20 mg significantly increased the number of avoidance responses for 30 min following injection (Fig. 2A). In most cases, avoidance responses
M.A. Oatway et al. / Pain 110 (2004) 259–268
following m-CPBG administration were characterized by an increase in aggressive behaviours and vocalization. m-CPBG (20 mg) treatment significantly increased the number of avoidance responses at 5 min (8.6 ^ 0.4), 15 min (8.4 ^ 0.3) and 30 min (8.8 ^ 0.4) ðP , 0:05Þ: On the other hand, delivery of the 5-HT3-R antagonist, ondansetron, resulted in a dose-dependent reduction in avoidance responses at the three doses used (1, 10 and 20 mg) (Fig. 2B). The maximal decrease in avoidance responses by the three doses of ondansetron was stable for the initial 30 min and remained significantly decreased at 60 min post-injection. The greatest effect was observed with the 20 mg dose of ondansetron, that significantly decreased avoidance responses compared to the number after saline treatment at all times tested ðP , 0:05Þ: The reduction in avoidance responses produced by ondansetron was reversed when m-CPBG was administered 10 min later (Fig. 2C). The number of avoidance responses following this m-CPBG administration did not differ from those observed in saline-treated rats. Conversely, the increased number of avoidance responses produced by m-CPBG injection was reversed by injection of ondansetron at 10 min (Fig. 2C). This reversal was most evident at the 30 min time point when avoidance responses were significantly less (3.4 ^ 0.6) than those in saline-treated rats (6.4 ^ 0.4) at the same time point ðP , 0:05Þ: 3.3. Exogenous 5-HT transiently reduces at-level mechanical allodynia following SCI At the three doses used (1, 10 and 20 mg), intrathecallyadministered 5-HT transiently reduced the number of avoidance responses ðP , 0:05Þ (Fig. 3A). The 10 mg dose of 5-HT had the greatest effect in reducing allodynic behaviour, decreasing avoidance responses to 3.8 ^ 1.3 at 5 min post-injection. The effect of all three 5-HT doses was present only for the initial 5 min, and avoidance responses throughout the remaining time course were not different from those in saline-treated rats. 3.4. The reduction of at-level mechanical allodynia is greater when a 5-HT3-R antagonist is combined with exogenous 5-HT To determine more accurately the role of spinal 5-HT receptors in reducing or facilitating mechanical allodynia, combinations of different serotonergic receptor agonists and antagonists were administered intrathecally. The increase in avoidance responses following m-CPBG (7.2 ^ 0.4) was reduced to 5.3 ^ 0.5 responses by a subsequent administration of 5-HT (10 mg) at 10 min (Fig. 3B). Similar to the effect of exogenous 5-HT alone, the reversal was transient, present only 5 min post-injection. Administration of 5-HT (10 mg) following pre-treatment with ondansetron resulted in a greater decrease in avoidance responses, reducing the number observed following ondansetron alone (2.4 ^ 0.3)
263
Fig. 3. Effect of acute (A) exogenous 5-HT alone or (B) following treatment with m-CPBG or ondansetron on at-level mechanical allodynia five weeks post-SCI. Baseline represents the number of avoidance responses made during the fourth post-SCI week, prior to intrathecal catheter insertion. (A) 5-HT (1, 10 and 20 mg) or saline was administered at 0 min and mechanical allodynia was assessed for the subsequent 60 min. (B) m-CPBG (20 mg) or ondansetron (20 mg) was administered at 0 min and mechanical allodynia was assessed at 5 min. 5-HT (10 or 20 mg) or saline was then administered at 10 min (indicated by 2nd injection). The values shown represent mean number of avoidance responses out of ten stimulations ^ SEM (n ¼ 5/group). *, P , 0:05 compared to saline-treated controls.
to 1.2 ^ 0.4 after the combined treatment (Fig. 3B). This partial summation of effects was also transient. Methysergide was used to block 5-HT1/5-HT2 receptors non-selectively following delivery of 5-HT3-R agonists or antagonists. Methysergide alone significantly increased the number of avoidance responses (8.2 ^ 0.4; P , 0:05) to a similar level to that observed with selective 5-HT3-R activation (Fig. 4). This increase in responses was observed only at 20 min post-injection. Administration of methysergide 10 min following pre-treatment with m-CPBG further increased the number of avoidance responses (9.2 ^ 0.4). The reduction in mechanical allodynia following ondansetron administration was reversed 20 min after methysergide injection, increasing avoidance responses from 2.6 ^ 0.5 following ondansetron alone to 5.8 ^ 0.6 following the combination treatment. 3.5. Neurotoxin-induced depletion of spinal 5-HT eliminates the effect of the 5-HT3 -R antagonist To determine whether ondansetron acted to block the action of endogenous 5-HT at the 5-HT3-R, spinal 5-HT was
264
M.A. Oatway et al. / Pain 110 (2004) 259–268
Fig. 4. Effect of post-treatment of methysergide on at-level mechanical allodynia following treatment with m-CPBG or ondansetron five weeks post-SCI. m-CPBG (20 mg), ondansetron (20 mg) or saline was administered at 0 min and mechanical allodynia was assessed at 5 min. Methysergide (30 mg) was then administered at 10 min (indicated by 2nd injection). The values shown represent mean number of avoidance responses to ten stimulations ^ SEM ðn ¼ 5/group). *, P , 0:05 compared to saline-treated controls.
depleted 4 weeks post-SCI by i.t. treatment with the neurotoxin, 5,7-DHT. At four weeks following SCI, animals received either one i.t. dose of 5,7-DHT or saline. In animals receiving saline treatment, 5-HT-immunoreactive fibres were densely packed within the superficial laminae and randomly distributed throughout laminae III and IV in the lower thoracic cord (T9-T11) immediately rostral to the lesion site (Fig. 5A). In contrast, administration of 5,7-DHT resulted in decreased 5-HT throughout the spinal cord, particularly within the at-level area of interest (T9-T11) where the dorsal horn was void of serotonergic immunoreactivity with the exception of sparsely distributed varicose fibres within the superficial laminae (Fig. 5B). Three days following pre-treatment with either 5,7-DHT or saline, toxin-treated rats responded significantly less to
the innocuous stimulus following an acute saline injection than did rats pretreated with saline (Fig. 6A). Pre-treatment with 5,7-DHT did not appear to cause adverse motor or behavioural effects during testing sessions. Acute administration of the 5-HT3-R agonist, m-CPBG, increased avoidance responses in both 5,7-DHT and saline pre-treated groups (Fig. 6B). The time courses for changes in response behaviour were identical for both groups with significant increases in avoidance responses occurring for the initial 30 min ðP , 0:05Þ: Conversely, animals with depleted spinal 5-HT had approximately the same number of avoidance responses following ondansetron treatment as those receiving only saline, for the entire 60 min time course (Fig. 6C). Animals without the neurotoxin treatment had the expected response to ondansetron, responding significantly less than did 5,7-DHT pre-treated animals ðP , 0:05Þ:
4. Discussion Our study investigated the involvement of the altered spinal serotonergic system in maintaining mechanical allodynia at the level of SCI and generated four major findings. First, selective blockade of spinal 5-HT3-R with the antagonist, ondansetron, produced a sustained reduction of at-level mechanical allodynia, whereas selective activation of 5-HT3-R with m-CPBG enhanced allodynia. Second, non-selective excitation of spinal 5-HT receptors with exogenous 5-HT transiently reduced allodynia, whereas non-selective blockade of 5-HT1/5-HT2 receptors with methysergide caused a transient increase in allodynia. Third, selective blockade of 5-HT3-R, combined with a nonselective excitation of the 5-HT1/5-HT2 receptors, further reduced at-level mechanical allodynia. Fourth, neurotoxininduced depletion of spinal 5-HT eliminated the effect of ondansetron, suggesting that the reduction in at-level mechanical allodynia is due to the blockade of serotonergic activation of the 5-HT3-R.
Fig. 5. Photomicrographs of 5-HT immunoreactivity in dorsal horn transverse sections from the 9th thoracic spinal segment of a (A) saline-treated and (B) 5,7-DHT treated rat five weeks post-SCI. Animals received either intrathecal saline or 5,7-DHT, 7 days prior to perfusion. Following 5,7-DHT, the 5-HT immunoreactive fibres were only sparsely distributed within the laminae I-IV when compared to the robust distribution of fibres in the saline-treated, injured animal. Bar ¼ 100 mm.
M.A. Oatway et al. / Pain 110 (2004) 259–268
265
pathology for the acute and chronic phases of post-SCI allodynia. Most notably, short-lasting allodynia is attributed to glutamate-induced excitation of NMDA and AMPA receptors (Xu et al., 1993; Bennett et al., 2000), accompanied by a loss of GABAergic inhibition (Weisenfeld-Hallin et al., 1997). The chronic-phase of allodynia, occurring days to months post-injury, involves neuronal hyperexcitability (Finnerup et al., 2003; Hoheisel et al., 2003) and upregulation of the anti-opioid polypeptide, cholecystokinin (Xu et al., 1994). Differentiating the pathophysiology underlying at-level versus below-level neuropathic pain is more difficult. A recent sampling of patients within the first six months after SCI reported a higher prevalence of neuropathic at-level pain than below-level pain, particularly in patients with incomplete injuries (Siddall et al., 1999). In humans, at-level neuropathic pain begins within days to weeks after SCI while below-level pain occurs within months to years post-injury (Beric et al., 1988). Our study in rats revealed a significant increase in the development of atlevel mechanical allodynia occurring at the second post-SCI week; below-level allodynia in our model develops about one week later (Bruce et al., 2002). The differences between post-SCI at-level and below-level neuropathic pain may, therefore, be associated with the differential pathophysiological substrates related to complex changes in descending pathways, particularly, in the role of serotonergic inputs. 4.2. Effects of 5-HT3-R blockade on at-level mechanical allodynia following SCI
Fig. 6. Effect of 5,7-DHT pre-treatment on at-level mechanical allodynia five weeks post-SCI. 5,7-DHT (60 mg) or saline was intrathecally administered at 28 days post-SCI and mechanical allodynia was assessed 3-7 days later following acute injection of (A) saline (B) m-CPBG or (C) ondansetron. The values shown represent the mean number of avoidance responses to ten stimulations ^ SEM ðn ¼ 6 in saline pre-treated groups; n ¼ 7 in 5,7-DHT pre-treated groups). *, P , 0:05 compared to salinetreated controls.
4.1. Mechanical allodynia following SCI The mechanisms underlying post-SCI mechanical allodynia remain controversial. The robust cascade of acute pathophysiological events following SCI, which include excitotoxicity, inflammation and cell death, results in chronic secondary maladaptive plasticity, and these may all contribute to the development of neuropathic pain (Krenz and Weaver, 1998; Eide, 1998). Evidence from ischemic and spinal injury models has revealed different
The present study demonstrates that selective blockade of spinal 5-HT 3-R using intrathecal ondansetron significantly attenuates at-level mechanical allodynia. The 5-HT3-R is an excitatory ligand-gated ion channel localized in high density within the superficial laminae of the dorsal horn (Kia et al., 1995; Morales et al., 1996). The exact neuronal populations expressing the 5-HT3-R remain controversial. 5-HT3-R immunoreactivity is not found in glutamic acid decarboxylase-expressing interneurons, but is present in primary afferent fibres. A small proportion of these fibres contain calcitonin gene-related peptide (CGRP) (Maxwell et al., 2003), and the others are proposed to be an uncharacterized subset of myelinated and unmyelinated nociceptors (Zeitz et al., 2002). Exogenous 5-HT potentiates the evoked release of substance P, CGRP and neurokinin-A from the central terminals of primary afferent fibres via 5-HT3-R activation (Saria et al., 1990; Inoue et al., 1997). Hence, although the development of mechanical allodynia after SCI is not due to sprouting of the central arbour of primary afferent fibres (Bruce et al., 2002), long-term upregulation of serotonergic fibres may maintain neuropathic pain through evoked release of pro-nociceptive peptides from primary afferents via 5-HT3-R activation. The role of the 5-HT3-R in nociception is uncertain due to the extensive range of experimental paradigms used in
266
M.A. Oatway et al. / Pain 110 (2004) 259–268
different studies. Antinociception resulting from 5-HT3-R excitation has been reported, primarily using acute pain models (Glaum et al., 1990; Alhaider et al., 1991; Paul et al., 2001). Data from models of long-term pain, most notably the formalin test, have shown that intrathecal blockade of 5-HT3-R reduces second phase, but not first phase formalin behaviours (Oyama et al., 1996; Zeitz et al., 2002), suggesting that the 5-HT3-R contributes to persistent rather than acute nociception. These findings have recently translated into the clinic where ondansetron significantly reduces pain scores in chronic neuropathic pain patients (McCleane et al., 2003). Blockade of spinal 5-HT3-R mimics the decreased mechanical and thermal evoked responses of deep dorsal horn neurons as well as the reduced formalin response observed following selective ablation of laminae I/III NK-1 receptor-expressing neurons (Suzuki et al., 2002). This result, in parallel with ours, demonstrates mechanistic similarities across various pain models suggesting that the 5-HT3-R provides a spinal target for excitatory projections from the brainstem, thus driving a facilitatory spinalbrainstem-spinal loop pathway which results in a shift towards increased spinal excitability. Descending facilitation arising from the RVM is involved in the maintenance, but not the initiation, of neuropathic pain (Burgess et al., 2002). Similarly, low-intensity electrical or chemical stimulation of cells within the nucleus raphe magnus produces a descending facilitatory influence via the ventrolateral funiculus while descending inhibition is observed following high-intensity stimulation mediated by descending serotonergic fibres in the dorsolateral funiculus (Zhuo and Gebhart, 1994; Zhuo and Gebhart, 1997). Combined with the evidence of the current study, the magnitude and time course of supraspinal serotonergic release are crucial in establishing a descending facilitatory influence. These characteristics will be important in differentiating acute and chronic pain conditions. In addition, the induction of abnormal spinal plasticity following SCI appears to upregulate spinal 5-HT pathways which may drive the maintenance of neuropathic pain via enhanced 5-HT turnover from the RVM. Ensuing excitation of spinal 5-HT3-R may enhance neurotransmitter release from primary afferent neurons, maintaining an ascending, excitatory drive to brainstem regions, thus completing the facilitatory loop. 4.3. Effects of exogenous 5-HT on at-level mechanical allodynia following SCI Although our results show that selective blockade of spinal 5-HT3-R significantly reduces at-level mechanical allodynia, we also observed a transient reduction following intrathecal exogenous 5-HT. This observation contributes to evidence that 5-HT biphasically modulates nociceptive processing through activation of individual classes of spinal receptor subtypes. Others have shown that intrathecal 5-HT
has a dose-dependent, dual effect on pain transmission with low doses reducing, and high doses increasing, response duration in the formalin test (Oyama et al., 1996). 5-HT binds with modest affinity to the spinal 5-HT3 -R (Ki ¼ 500 nM), and binds with highest affinity at spinal 5-HT1 and 5-HT2 receptor sites (approximate Ki , 10 nM) (Glennon and Dukat, 1995). The spinal 5-HT1A receptors have been proposed to be the high-affinity receptors mediating descending inhibition and the 5-HT3-R to be low affinity receptors mediating descending facilitation of nociceptive transmission (Oyama et al., 1996). A majority of evidence supports this antinociceptive (Millan, 1994; Obata et al., 2001) and anti-allodynic action (Sasaki et al., 2001; Colpaert et al., 2002) of 5-HT1A and 5-HT2A/2C receptor agonists. Our results suggest that the delivery of exogenous 5-HT may transiently reduce at-level mechanical allodynia via acute activation of 5-HT1 and 5-HT2 receptors. Similar transient antihyperalgesic effects of intrathecal 5-HT have been reported in a model of mononeuropathy (Bardin et al., 2000). Our results demonstrated that the anti-allodynic effect of 5-HT3-R blockade was enhanced transiently by exogenous 5-HT. Thus, a combined inhibition of 5-HT3-R and excitation of 5-HT1/5-HT2 receptors greatly reduced at-level mechanical allodynia. Furthermore, nonselective blockade of 5-HT1/5-HT2 receptors with methysergide exacerbated the mechanical allodynia, in a transient manner, to levels similar to those observed after m-CPBG delivery. In this condition, only the 5-HT3-R would be available for endogenous 5-HT activation. These observations could be explained by a model in which long-term alterations in the endogenous spinal 5-HT system follow SCI. Increased levels of ligand may then result in a greater participation of receptors, such as the 5-HT3-R, that normally have only modest affinity for the ligand. 4.4. Effects of the depletion of spinal 5-HT on at-level mechanical allodynia following SCI Depletion of spinal cord 5-HT levels via intrathecal administration of the neurotoxin, 5,7-DHT has been previously reported as a method for characterizing the role of 5-HT in nociceptive modulation, (Duan and Sawynok, 1987; Liu et al., 1988). We observed that depletion of spinal 5-HT by 5,7-DHT eliminated the anti-allodynic activity of ondansetron, verifying that the reduction in mechanical allodynia requires endogenous 5-HT and is due specifically to the blockade of 5-HT3-R activation by 5-HT. Therefore, although ondansetron has local anaesthetic properties (Ye et al., 1997), the reduction in mechanical allodynia by ondansetron appears to be the result of a selective blockade of spinal serotonergic activity. The depletion of spinal 5-HT levels immediately above the lesion site decreased at-level mechanical allodynia. Similarly, the development of autotomy following sciatic nerve transection is suppressed
M.A. Oatway et al. / Pain 110 (2004) 259–268
by selective neurotoxic depletion of lumbosacral 5-HT (Feria et al., 1992). To our knowledge, ours is the first study to investigate the role of spinal serotonergic depletion on SCI-induced mechanical allodynia and our data provide further evidence that the increased serotonergic fibre density immediately rostral to the lesion site is involved in the maintenance of at-level mechanical allodynia. However, the residual at-level allodynia following spinal 5-HT depletion suggests that, in the absence of this mediator, alternative mechanisms may develop.
5. Summary Abnormal neuroplasticity resulting from SCI results in changes in serotonergic fibre density and distribution that directly affect the activation of post-synaptic spinal 5-HT receptors. The present study shows that the increased density of serotonergic fibres immediately rostral to the lesion site contributes to the maintenance of at-level mechanical allodynia via activation of the 5-HT3-R. Patients reporting the presence of at-level mechanical allodynia are generally refractory to a majority of analgesics, although some investigations suggest benefits of intrathecal baclofen in central, deafferentation pain (Taira et al., 1995; Gatscher et al., 2002). Based on the pronociceptive actions of 5-HT that we have shown, as well as the present clinical availability of 5-HT3-R antagonists, our study suggests a new avenue of focus for pharmacological treatment of at-level neuropathic pain following SCI.
Acknowledgements This research was supported by grants from the Ontario Neurotrauma Foundation and the Canadian Institutes of Health Research. M.A.O. was supported by a studentship award from the Ontario Neurotrauma Foundation. We thank Drs. Arthur Brown, Canio Polosa and Jana Sawynok for their constructive criticisms of this manuscript.
References Alhaider AA, Lei SZ, Wilcox GL. Spinal 5-HT3 receptor-mediated antinociception: possible release of GABA. J Neurosci 1991;11: 1881–8. Bardin L, Schmidt J, Alloui A, Eschalier A. Effect of intrathecal administration of serotonin in chronic pain models in rats. Eur J Pharmacol 2000;409:37– 43. Bennett AD, Everhart AW, Hulsebosch CE. Intrathecal administration of an NMDA or a non-NMDA receptor antagonist reduces mechanical but not thermal allodynia in a rodent model of chronic central pain after spinal cord injury. Brain Res 2000;859:72 –82. Beric A, Dimitrijevic MR, Lindblom U. Central dysesthesia syndrome in spinal cord injury patients. Pain 1988;34:109 –16. Bruce JC, Oatway MA, Weaver LC. Chronic pain after clip compression injury of the rat spinal cord. Exp Neurol 2002;178:33 –48.
267
Burgess SE, Gardell LR, Ossipov MH, Malan TP, Jr., Vanderah TW, Lai J, Porreca F. Time-dependent descending facilitation from the rostral ventro medial medulla maintains but does not initiate, neuropathic pain. J Neurosci 2002;22:5129–36. Colpaert FC, Tarayre JP, Koek W, Pauwels PJ, Bardin L, Xu XJ, Wiesenfeld-Hallin Z, Cosi C, Carilla-Durand E, Assie MB, Vacher B. Large-amplitude 5-HT1A receptor activation: a new mechanism of profound, central analgesia. Neuropharmacology 2002;43:945–58. Duan J, Sawynok J. Enhancement of clonidine-induced analgesia by lesions induced with spinal and intracerebroventricular administration of 5,7dihydroxytryptamine. Neuropharmacology 1987;26:323– 9. Eide PK. Pathophysiological mechanisms of central neuropathic pain after spinal cord injury. Spinal Cord 1998;36:601–12. Feria M, Sanchez A, Abad F, Abreu P. Effects of selective neurotoxic lesion of lumbosacral serotonergic and noradrenergic systems on autotomy behaviour in rats. Pain 1992;51:101–9. Finnerup NB, Johannesen IL, Fuglsang-Frederiksen A, Bach FW, Jensen TS. Sensory function in spinal cord injury patients with and without central pain. Brain 2003;126:57–70. Gatscher S, Becker R, Uhle E, Bertalanffy H. Combined intrathecal baclofen and morphine infusion for the treatment of spasticity related pain and central deafferentiation pain. Acta Neurochir Suppl 2002;79: 75– 6. Glaum SR, Proudfit HK, Anderson EG. 5-HT3 receptors modulate spinal nociceptive reflexes. Brain Res 1990;510:12–16. Glennon RA, Dukat M. Serotonin receptor subtypes. In: Bloom FE, Kupfer DJ, editors. Psychopharmacology: The fourth generation of progress. New York: Raven Press; 1995. p. 415–29. Green GM, Scarth J, Dickenson A. An excitatory role for 5-HT in spinal inflammatory nociceptive transmission; state-dependent actions via dorsal horn 5-HT(3) receptors in the anaesthetized rat. Pain 2000;89: 81– 8. Hains BC, Johnson KM, McAdoo DJ, Eaton MJ, Hulsebosch CE. Engraftment of serotonergic precursors enhances locomotor function and attenuates chronic central pain behavior following spinal hemisection injury in the rat. Exp Neurol 2001;171:361–78. Hains BC, Everhart AW, Fullwood SD, Hulsebosch CE. Changes in serotonin, serotonin transporter expression and serotonin denervation supersensitivity: Involvement in chronic central pain after spinal hemisection in the rat. Exp Neurol 2002;175:347– 62. Hamon M, Bourgoin S. Serotonin and Its Receptors in Pain Control. In: Sawynok J, Cowan A, editors. Novel Aspects of Pain Management: Opioids and Beyond. Wiley; 1999. p. 203 –28. Hao JX, Xu XJ, Aldskogius H, Seiger A, Wiesenfeld-Hallin Z. Allodynialike effects in rat after ischaemic spinal cord injury photochemically induced by laser irradiation. Pain 1991;45:175–85. Hoheisel U, Scheifer C, Trudrung P, Unger T, Mense S. Pathophysiological activity in rat dorsal horn neurones in segments rostral to a chronic spinal cord injury. Brain Res 2003;974:134–45. Horiuchi H, Ogata T, Morino T, Takeba J, Yamamoto H. Serotonergic signaling inhibits hyperalgesia induced by spinal cord damage. Brain Res 2002;963:312– 20. Howe JR, Yaksh TL. Changes in sensitivity to intrathecal norepinephrine and serotonin after 6-hydroxydopamine (6-OHDA) 5,6-dihydroxytryptamine (5,6-DHT) or repeated monoamine administration. J Pharmacol Exp Ther 1982;220:311– 21. Inman DM, Steward O. Ascending sensory, but not other long-tract axons, regenerate into the connective tissue matrix that forms at the site of a spinal cord injury in mice. J Comp Neurol 2003;462:431–49. Inoue A, Hashimoto T, Hide I, Nishio H, Nakata Y. 5-Hydroxytryptaminefacilitated release of substance P from rat spinal cord slices is mediated by nitric oxide and cyclic GMP. J Neurochem 1997;68:128–33. Kia HK, Miquel MC, McKernan RM, Laporte AM, Lombard MC, Bourgoin S, Hamon M, Verge D. Localization of 5-HT3 receptors in the rat spinal cord: immunohistochemistry and in situ hybridization. Neuroreport 1995;6:257–61.
268
M.A. Oatway et al. / Pain 110 (2004) 259–268
Krenz NR, Weaver LC. Sprouting of primary afferent fibers after spinal cord transection in the rat. Neurosci 1998;85:443 –58. Liu MY, Su CF, Lin MT. The antinociceptive role of a bulbospinal serotonergic pathway in the rat brain. Pain 1988;33:123–9. Mason P. Central mechanisms of pain modulation. Curr Opin Neurobiol 1999;9:436–41. Maxwell DJ, Kerr R, Rashid S, Anderson E. Characterisation of axon terminals in the rat dorsal horn that are immunoreactive for serotonin 5-HT3A receptor subunits. Exp Brain Res 2003;149:114–24. McCleane GJ, Suzuki R, Dickenson AH. Does a single intravenous injection of the 5HT3 receptor antagonist ondansetron have an analgesic effect in neuropathic pain? A double-blinded, placebocontrolled cross-over study. Anesth Analg 2003;97:1474–8. Millan MJ. Serotonin and pain: evidence that activation of 5-HT1A receptors does not elicit antinociception against noxious thermal, mechanical and chemical stimuli in mice. Pain 1994;58:45–61. Millan MJ. Descending control of pain. Prog Neurobiol 2002;66:355–474. Morales M, Battenberg E, de Lecea L, Sanna PP, Bloom FE. Cellular and subcellular immunolocalization of the type 3 serotonin receptor in the rat central nervous system. Brain Res Mol Brain Res 1996;36:251– 60. Obata H, Saito S, Sasaki M, Ishizaki K, Goto F. Antiallodynic effect of intrathecally administered 5-HT(2) agonists in rats with nerve ligation. Pain 2001;90:173 –9. Ossipov M, Lai J, Malan JrT, Vanderah T, Porreca F. Tonic descending facilitation as a mechanism of neuropathic pain. Neuropathic Pain: pathophysiology and treatment. Seattle: International Association for the Study of Pain; 2001. p. 107–24. Oyama T, Ueda M, Kuraishi Y, Akaike A, Satoh M. Dual effect of serotonin on formalin-induced nociception in the rat spinal cord. Neurosci Res 1996;25:129–35. Parisod E, Siddall PJ, Viney M, McClelland JM, Cousins MJ. Allodynia after acute intrathecal morphine administration in a patient with neuropathic pain after spinal cord injury. Anesth Analg 2003;97:183– 6. Paul D, Yao D, Zhu P, Minor LD, Garcia MM. 5-hydroxytryptamine3 (5-HT3) receptors mediate spinal 5-HT antinociception: an antisense approach. J Pharmacol Exp Ther 2001;298:674–8. Rivlin AS, Tator CH. Objective clinical assessment of motor function after experimental spinal cord injury in the rat. J Neurosurg 1977;47:577– 81. Saria A, Javorsky F, Humpel C, Gamse R. 5-HT3 receptor antagonists inhibit sensory neuropeptide release from the rat spinal cord. Neuroreport 1990;1:104–6. Sasaki M, Ishizaki K, Obata H, Goto F. Effects of 5-HT2 and 5-HT3 receptors on the modulation of nociceptive transmission in rat spinal cord according to the formalin test. Eur J Pharmacol 2001;424:45– 52. Sawynok J, Reid A. Spinal supersensitivity to 5-HT1, 5-HT2 and 5-HT3 receptor agonists following 5,7-dihydroxytryptamine. Eur J Pharmacol 1994;264:249–57. Siddall PJ, Loeser JD. Pain following spinal cord injury. Spinal Cord 2001; 39:63–73. Siddall PJ, Xu CL, Floyd N, Keay KA. C-fos expression in the spinal cord of rats exhibiting allodynia following contusive spinal cord injury. Brain Res 1999;851:281–6.
Suzuki R, Morcuende S, Webber M, Hunt SP, Dickenson AH. Superficial NK1-expressing neurons control spinal excitability through activation of descending pathways. Nat Neurosci 2002;5:1319–26. Taira T, Kawamura H, Tanikawa T, Kawabatake H, Iseki H, Ueda A, Takakura K. A new approach to the control of central deafferentation pain–spinal intrathecal baclofen. Acta Neurochir Suppl (Wien) 1995; 64:136–8. Vierck Jr CJ, Siddall P, Yezierski RP. Pain following spinal cord injury: animal models and mechanistic studies. Pain 2000;89:1–5. Warms CA, Turner JA, Marshall HM, Cardenas DD. Treatments for chronic pain associated with spinal cord injuries: many are tried, few are helpful. Clin J Pain 2002;18:154–63. Weaver LC, Verghese P, Bruce JC, Fehlings MG, Krenz NR, Marsh DR. Autonomic dysreflexia and primary afferent sprouting after clipcompression injury of the rat spinal cord. J Neurotrauma 2001;18: 1107–19. Weisenfeld-Hallin Z, Aldskogius H, Grant G, Hao JX, Hokfelt T, Xu XJ. Central inhibitory dysfunctions: mechanisms and clinical implications. Behav Brain Sci 1997;20(3):420– 5. ˚ , Wiesenfeld-Hallin Z. Systemic excitatory amino Xu X-J, Hao JX, Seiger A acid receptor antagonists of the (-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor and of the N-methyl-Daspartate (NMDA) receptor relieve mechanical hypersensitivity after transient spinal cord ischemia in rats. J Pharm Exper Therap 1993; 267(1):140–4. ˚ , Hughes J, Ho¨kfelt T, Wiesenfeld-Hallin Z. Xu X-J, Hao JX, Seiger A Chronic pain-related behaviors in spinally injured rats: evidence for functional alterations of the endogenous cholecystokinin and opioid systems. Pain 1994;56:271–7. Ye JH, Mui WC, Ren J, Hunt TE, Wu WH, Zbuzek VK. Ondansetron exhibits the properties of a local anesthetic. Anesth Analg 1997;85: 1116–21. Yezierski RP. Pain following spinal cord injury: the clinical problem and experimental studies. Pain 1996;68:185–94. Yezierski RP. Pain following spinal cord injury: pathophysiology and central mechanisms. Prog Brain Res 2000;129(429-49): 429 – 49. Zeitz KP, Guy N, Malmberg AB, Dirajlal S, Martin WJ, Sun L, Bonhaus DW, Stucky CL, Julius D, Basbaum AI. The 5-HT3 subtype of serotonin receptor contributes to nociceptive processing via a novel subset of myelinated and unmyelinated nociceptors. J Neurosci 2002; 22:1010– 9. Zhuo M, Gebhart GF. Spinal serotonin receptors mediate descending facilitation of a nociceptive reflex from the nuclei reticularis gigantocellularis and gigantocellularis pars alpha in the rat. Brain Res 1994;550:35–48. Zhuo M, Gebhart GF. Biphasic modulation of spinal nociceptive transmission from the medullary raphe nuclei in the rat. J Neurophysiol 1997;78:746–58. Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;16:109 –10.
The 5-HT3 receptor facilitates at-level mechanical allodynia following spinal cord injury Mark A. Oatwaya,b, Yuhua Chena, Lynne C. Weavera,b,* a
The Spinal Cord Injury Team, Biotherapeutics Research Group, Robarts Research Institute, P.O. Box 5015, 100 Perth Drive, London, Ont., Canada N6A 5K8 b The Graduate Program in Neuroscience, The University of Western Ontario, London, Ont., Canada N6A 5K8 Received 23 January 2004; received in revised form 15 March 2004; accepted 29 March 2004
Abstract Spinal cord injury (SCI) results in the development of mechanical allodynia immediately rostral to the lesion site, within the dermatome border of normal sensation and sensory loss (at-level mechanical allodynia). We propose that an observed threefold increase in serotonergic fibre immunoreactivity within spinal segments corresponding to these allodynic dermatomes facilitates the maintenance of chronic neuropathic pain via activation of the 5-HT3 receptor (5-HT3-R). Serotonin (5-HT), the non-selective 5-HT1/5-HT2 receptor antagonist, methysergide, the 5-HT3-R agonist, m-chlorophenylbiguanide (m-CPBG) or the 5-HT3-R antagonist, ondansetron were intrathecally administered five weeks following SCI in rats. Ondansetron produced a robust, long-term reduction of at-level mechanical allodynia, while m-CPBG exacerbated allodynia. Exogenous 5-HT transiently reduced at-level mechanical allodynia. This effect was opposed by methysergide, which enhanced mechanical allodynia. Co-administration of 5-HT and ondansetron produced a short-lasting partial summation of effects, further decreasing mechanical allodynia while co-administration of methysergide attenuated the anti-allodynic effect of ondansetron. Depletion of spinal 5-HT via 5,7-dihydroxytryptamine (5,7-DHT) resulted in decreased at-level mechanical allodynia. The reduction of allodynia by ondansetron was lost following 5,7-DHT administration, suggesting that reduced allodynia following intrathecal ondansetron is via blockade of 5-HT-induced excitation of the 5-HT3-R. These results suggest that increased 5-HT fibre density immediately rostral to the SCI lesion site could have transient effects to reduce mechanical allodynia via actions at 5-HT1 and/or 5-HT2 receptors. However, the more long-lasting effects of this enhanced serotonergic input may facilitate chronic, at-level allodynia via the 5-HT3-R. q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: Spinal cord injury; Serotinin; Neuropathic pain; 5-HT3 receptor
1. Introduction The development and maintenance of chronic pain following spinal cord injury (SCI) is a significant problem that is often refractory to a majority of clinical interventions (Yezierski, 1996; Warms et al., 2002). Neuropathic pain varies in quality and intensity amongst patients and is classified according to anatomical location (Vierck, Jr. et al., 2000). At-level neuropathic pain occurs at the sensory transitional zone, within a band of two to four spinal segments rostral to the lesion level (Siddall and Loeser, 2001). Animal models of spinal ischemia and injury have * Corresponding author. Biotherapeutics Research Group, Robarts Research Institute, P.O. Box 5015, 100 Perth Dr., London, Ont., Canada N6A 5K8. Tel.: þ 519-663-3776; fax: þ519-663-3789. E-mail address: [email protected] (L.C. Weaver).
been successful in reproducing at-level allodynia (Hao et al., 1991; Yezierski, 2000). Although such models have suggested that chronic alterations in spinal and supraspinal endogenous opioid systems may contribute to post-SCI allodynia (Weisenfeld-Hallin et al., 1997), SCI patients with neuropathic pain generally respond poorly to opioids (Parisod et al., 2003). A further understanding of the morphological alterations above the lesion level following SCI, primarily within descending modulatory pain pathways, is required to characterize alternative pharmacological targets. The loss of descending serotonergic innervation caudal to the SCI lesion has been causally linked to the development of allodynia below the lesion level (Hains et al., 2002). Promotion of serotonergic actions via transplantation of 5-HT-secreting cells (Hains et al., 2001), or administration of a selective serotonin reuptake
0304-3959/$20.00 q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2004.03.040
260
M.A. Oatway et al. / Pain 110 (2004) 259–268
inhibitor (Horiuchi et al., 2002) transiently decreases belowlevel neuropathic pain. However, previous work has shown a significant increase in serotonin (5-HT) fibre density in spinal segments immediately rostral to the lesion level that corresponds with the dermatomes from which at-level mechanical allodynia is elicited (Bruce et al., 2002; Inman and Steward, 2003). The maintenance of neuropathic pain has been attributed to descending facilitation resulting from plastic changes within the rostral ventromedial medulla (RVM) (Ossipov et al., 2001). Chronic noxious stimuli activate RVM serotonergic cells, causing increases in spinal 5-HT turnover and contributing to descending facilitation (Zhuo and Gebhart, 1997; Mason, 1999). The influence of descending 5-HT on pain transmission within the dorsal horn is dependent on the class of 5-HT receptor subtype activated (Millan, 2002). Spinal 5-HT receptors biphasically inhibit or facilitate ascending pain transmission at the second-order neuron. Whereas the 5-HT1 and 5-HT2 receptor subtypes bear the highest affinity for the endogenous ligand, and are believed to exert an overall antinociceptive action (Hamon and Bourgoin, 1999), models of persistent pain have suggested a role for 5-HT3 receptor (5-HT3-R) activation in the maintenance of pain (Green et al., 2000; Zeitz et al., 2002). Furthermore, the spinal 5-HT3-R mediate a part of the excitatory projections within facilitatory spinal-brainstem-spinal loops which are enhanced during chronic pain conditions (Suzuki et al., 2002). In the present study, we have tested the hypothesis that, following SCI, the persistence of at-level mechanical allodynia is maintained via excitation of spinal 5-HT3-R corresponding to the increased serotonergic fibre density immediately rostral to the lesion site. Using the clinicallyrelevant clip-compression model of SCI (Rivlin and Tator, 1977), we determined the effects of an intrathecallyadministered 5-HT3-R antagonist on at-level mechanical allodynia five weeks post-SCI when given alone, in combination with non-selective 5-HT receptor agonists and antagonists and following neurotoxin-induced spinal 5-HT depletion.
2. Materials and methods 2.1. Animals and surgical procedures Male Wistar rats (Charles River, Ontario, Canada) weighing 220 –350 g were housed individually with free access to food and water. All experimental procedures were approved by the University of Western Ontario Animal Care Committee in accordance with the Canadian Guide to Care and Use of Experimental Animals. Experiments also adhered to the guidelines of the Committee for Research and Ethical Issues of the International Association for the Study of Pain (Zimmermann, 1983). Prior to surgery, animals were medicated with diazepam (3.5 mg/kg, ip;
Sabex 2002 Inc., Boucherville, PQ) and atropine (0.05 mg/ kg, sc; Sigma-Aldrich, St Louis, MO) as described previously (Weaver et al., 2001) and anesthetised with 2.0% halothane in oxygen. A dorsal laminectomy was performed, removing the dorsal processes of the 10th and 11th thoracic vertebrae and exposing the 12th – 13th thoracic segmental junction. Using a modified aneurysm clip calibrated to generate a force of 35 g, the cord was compressed dorsoventrally for 60s producing a moderate, incomplete SCI. Care was taken to place the clip around the cord without disrupting the dura mater or adjacent dorsal roots. The muscle and skin were closed in layers and animals recovered under a heat lamp. Post-operative care was provided using antibiotics (Baytril 50 mg/ml, sc; Bayer Inc., Toronto, ON) and subcutaneous hydration with 0.9% normal saline daily for 3 days. Animals were housed on extra bedding to prevent the development of pressure and friction sores. Urinary bladders were manually emptied by abdominal compression twice daily until spontaneous voiding returned at 7– 10 days. 2.2. Behavioural procedures 2.2.1. Assessment of at-level mechanical allodynia The development of mechanical allodynia immediately rostral to and at the level of the SCI lesion was tested on the dorsal surface of the trunk. Using a Modified SemmesWeinstein monofilament (Stoelting Co., Wood Dale, IL), calibrated to generate a force of 20 mN, the dermatome areas corresponding to the 9th – 11th thoracic spinal segments were stimulated. Prior to testing, rats were acclimated to a plexiglass box similar to their home cage for 20 min in a designated testing room. A testing session consisted of ten stimulations to random points within the dorsal trunk area. Each stimulus lasted 3s and was separated by a 5s interim period. The number of avoidance responses elicited from ten stimulations was then tabulated. An avoidance response was defined by one of the following behaviours: trunk shakes, jumping or escaping to another area of the cage, vocalization or abnormal aggressive behaviour such as biting or swatting the filament. Rats were determined to exhibit allodynia if they displayed an increased number of avoidance behaviours to the innocuous stimulus post-injury when compared to pre-injury behaviour. A pre-injury baseline response to the filament was established by determining the mean number of avoidance responses during the week prior to injury. Testing resumed seven days following injury, with two testing sessions per week for the subsequent four post-injury weeks. 2.2.2. Intrathecal injections At post-SCI day 28, rats were fixed in a stereotaxic frame under general anaesthesia with 2.0% halothane in oxygen via a facemask. Intrathecal (i.t.) catheters consisting of 8.0 cm of Micro-Renathane tubing (0.025 in. O.D. £ 0.012 in. I.D.; Braintree Scientific Inc., Braintree, MA) fixed to
M.A. Oatway et al. / Pain 110 (2004) 259–268
5.3 cm of 32-gauge tubing (Recathco Inc, Allison Park, PA) were inserted through the atlanto-occipital membrane. The tip of the 32-gauge tubing was threaded along the spinal cord to terminate at the 8th –10th thoracic segment. The Micro-Renathane tubing connection exited the skin and was fixed to the back of the head. Following insertion, catheters were flushed with 20 ml of 0.9% saline using an insulin syringe (Becton Dickinson and Co., Franklin Lakes, NJ) and flushed each subsequent day until behavioural testing resumed. Rats were allowed to recover for three days following catheter insertion before testing resumed. Catheter positions were verified following perfusion by performing a full cervical and thoracic laminectomy. In some rats, i.t. administration of luxol fast blue dye immediately prior to perfusion verified the dispersion of fluids through the catheter and along the spinal cord. In most cases, the solutions reached the thoracic spinal cord rostral, but not into or caudal to the lesion site. 2.2.3. Treatment protocols Drug interventions commenced during the 5th post-SCI week, when at-level allodynia has reliably developed (Bruce et al., 2002). Prior to drug administration, rats were acclimated to the testing cages for 20 min. Each rat was administered an i.t. injection of drug compound dissolved in 0.9% saline or saline alone (10 ml/rat) followed by a saline flush (20 ml/rat). Rats ðn ¼ 5/group) received either serotonin hydrochloride (1, 10, 20 mg, Sigma-Aldrich), the 5-HT3-R agonist, m-chlorophenylbiguanide (m-CPBG; 1, 10, 20 mg, Sigma-Aldrich), the 5-HT3-R antagonist, ondansetron (1, 10, 20 mg, GlaxoSmithKline, Research Triangle Park, NC), the 5-HT1/5-HT2 receptor antagonist, methysergide maleate salt (30 mg, Research Biochemicals International, Natick, MA) or saline. Mechanical allodynia was assessed at 5, 15, 30, 45 and 60 min following injections. At each time point, rats received 10 stimulations to the dorsal trunk as described above. Co-administrative paradigms were designed to characterize the involvement of multiple serotonergic receptor subtypes. In each case, one drug was administered at time 0 min followed by the assessment of mechanical allodynia 5 min later. The second drug was then administered at 10 min and allodynia testing resumed at 15, 30, 45 and 60 min. 2.2.4. 5,7-Dihydroxytryptamine administration Endogenous spinal 5-HT was depleted with the neurotoxin, 5,7-dihydroxy-tryptamine creatinine sulphate salt (5,7-DHT; Sigma-Aldrich) administered intrathecally. Three days following i.t. catheter insertion, rats were pretreated with desipramine hydrochloride (25 mg/kg, i.p.; Sigma-Aldrich) to prevent uptake of the neurotoxin into noradrenergic neurons. After 45 min, rats received either 0.9% saline ðn ¼ 6Þ or 60 mg of 5,7-DHT ðn ¼ 7Þ in a volume of 20 ml followed by a 20 ml saline flush. This dose of 5,7-DHT has been reported to deplete endogenous spinal
261
5-HT significantly (Sawynok and Reid, 1994). All drug assessments were completed prior to day nine following 5,7-DHT administration to avoid the complication of supersensitivity development after this time (Howe and Yaksh, 1982). 2.3. Tissue preparation and processing Seven days following i.t. 5,7-DHT administration, rats were deeply anaesthetized with urethane (2.5 g/kg, i.p; Sigma Chemical Co.) and perfused as described previously (Bruce et al., 2002). To assess the density and distribution of spinal 5-HT fibres immediately rostral to lesion site following 5,7-DHT administration, spinal cord tissue was processed for 5-HT immunoreactivity. The spinal cord (9th– 11th thoracic segments) was cut with a cryostat into 30 mm transverse sections and floated in 0.1 M phosphate buffer. Floating sections were blocked for 60 min in TPBS-X containing 10% normal goat serum (NGS) and incubated in a rabbit anti-5-HT antibody (1:20,000; Immunostar, Hudson, WI) in 1% NGS for 72 h. The tissue was incubated in biotin-conjugated goat anti-rabbit antibody (1:250; Jackson Immunoresearch, Westgrove, PA) for 12 h and incubated in extravidin-peroxidase (Sigma-Aldrich) for 4 h. Immunoreactivity was revealed using a nickelenhanced process and diaminobenzidine as the chromogen. Several sections were processed without the primary antibody incubation step to verify a lack of non-specific staining. 2.4. Data analysis In assessing the development of mechanical allodynia, values are represented as mean number of avoidance responses to ten stimulations ^ standard error mean (SEM). Dose-response results were subjected to a repeated measures analysis of variance and a Fischer’s protected t-test. All drug treatments were compared against the salinetreated group unless otherwise noted. Differences were considered significant at P , 0:05:
3. Results 3.1. Development of at-level mechanical allodynia following SCI At-level mechanical allodynia was assessed during the second post-SCI week (day 7) until the middle of the fourth post-SCI week (day 26). During this time, the presence of mechanical allodynia was tested on the dorsal trunk immediately rostral and proximal to the lesion site. The mechanical stimulus was determined to be innocuous, as uninjured rats did not respond to stimulation of the dorsal trunk one-week prior to injury (Fig. 1). During the second post-injury week, SCI rats responded significantly more
262
M.A. Oatway et al. / Pain 110 (2004) 259–268
Fig. 1. The development of mechanical allodynia immediately rostral to the SCI lesion site. Testing sessions consisted of ten stimulations with an innocuous, monofilament one week prior to injury (baseline) and at the second, third and fourth post-SCI weeks. Each data point represents the mean ^ SEM number of avoidance responses made to 10 stimulations. The number of avoidance responses made to dorsal trunk stimulation significantly increased over time following SCI ðn ¼ 6Þ suggesting the development of at-level mechanical allodynia. *, P , 0:05 compared to pre-SCI mean responses; þ, P , 0:05 compared to avoidance responses made during the 2nd post-SCI week.
often (2.5 ^ 0.4 out of 10 stimulations) when compared to uninjured rats ðP , 0:05Þ: Previous results from our laboratory showed that sham-injured rats do not exhibit signs of mechanical allodynia simply due to laminectomyinduced inflammation during the second post-injury week (Bruce et al., 2002). The number of avoidance responses increased further during the third week (5.8 ^ 0.8) and remain stable during the fourth week post-SCI (6.4 ^ 0.7). At both of these time points, SCI rats responded significantly more to the innocuous stimulus when compared to the number of responses made prior to SCI and to the number during the second post-injury week ðP , 0:05Þ: In a follow-up study, we observed that the number of avoidance responses at four weeks remains constant for twelve weeks following SCI (data not shown). We therefore used the fourth week value as a baseline comparison for subsequent drug treatments. 3.2. A 5-HT3-R antagonist reduces at-level mechanical allodynia following SCI Intrathecal administration of various serotonergic agents was performed during the 5th post-injury week, a time when all rats had developed at-level mechanical allodynia. Baseline values on all figures represent the number of avoidance responses made during the 4th post-SCI week prior to i.t. catheter insertion. Placement of the catheter did not alter avoidance response frequencies as all animals responded at three days post-catheter placement in a manner similar to their responses prior to this surgery (data not shown).
Fig. 2. Effects of acute 5-HT3-R stimulation (m-CPBG) or blockade (ondansetron) on at-level mechanical allodynia five weeks post-SCI. Baseline represents the number of avoidance responses made during the fourth post-SCI week, prior to intrathecal catheter insertion. (A) m-CPBG (1, 10 and 20 mg), (B) ondansetron (1, 10 and 20 mg) or saline was administered at 0 min and mechanical allodynia was assessed for the subsequent 60 min. (C) The effect of alternating the order of injection of m-CPBG and ondansetron on at-level mechanical allodynia. m-CPBG (20 mg) or ondansetron (20 mg) was administered at 0 min and mechanical allodynia was assessed at 5 min. The second agent was then administered at 10 min (indicated by 2nd injection). The values shown represent mean number of avoidance responses to ten stimulations ^ SEM (n ¼ 5/group). *, P , 0:05 compared to saline-treated controls.
Saline-injected rats reacted to the stimulation consistently with an average of 6.2 ^ 0.04 responses over the 1 h testing period (Fig. 2). Administration of the 5-HT3-R agonist, m-CPBG, at doses of 10 and 20 mg significantly increased the number of avoidance responses for 30 min following injection (Fig. 2A). In most cases, avoidance responses
M.A. Oatway et al. / Pain 110 (2004) 259–268
following m-CPBG administration were characterized by an increase in aggressive behaviours and vocalization. m-CPBG (20 mg) treatment significantly increased the number of avoidance responses at 5 min (8.6 ^ 0.4), 15 min (8.4 ^ 0.3) and 30 min (8.8 ^ 0.4) ðP , 0:05Þ: On the other hand, delivery of the 5-HT3-R antagonist, ondansetron, resulted in a dose-dependent reduction in avoidance responses at the three doses used (1, 10 and 20 mg) (Fig. 2B). The maximal decrease in avoidance responses by the three doses of ondansetron was stable for the initial 30 min and remained significantly decreased at 60 min post-injection. The greatest effect was observed with the 20 mg dose of ondansetron, that significantly decreased avoidance responses compared to the number after saline treatment at all times tested ðP , 0:05Þ: The reduction in avoidance responses produced by ondansetron was reversed when m-CPBG was administered 10 min later (Fig. 2C). The number of avoidance responses following this m-CPBG administration did not differ from those observed in saline-treated rats. Conversely, the increased number of avoidance responses produced by m-CPBG injection was reversed by injection of ondansetron at 10 min (Fig. 2C). This reversal was most evident at the 30 min time point when avoidance responses were significantly less (3.4 ^ 0.6) than those in saline-treated rats (6.4 ^ 0.4) at the same time point ðP , 0:05Þ: 3.3. Exogenous 5-HT transiently reduces at-level mechanical allodynia following SCI At the three doses used (1, 10 and 20 mg), intrathecallyadministered 5-HT transiently reduced the number of avoidance responses ðP , 0:05Þ (Fig. 3A). The 10 mg dose of 5-HT had the greatest effect in reducing allodynic behaviour, decreasing avoidance responses to 3.8 ^ 1.3 at 5 min post-injection. The effect of all three 5-HT doses was present only for the initial 5 min, and avoidance responses throughout the remaining time course were not different from those in saline-treated rats. 3.4. The reduction of at-level mechanical allodynia is greater when a 5-HT3-R antagonist is combined with exogenous 5-HT To determine more accurately the role of spinal 5-HT receptors in reducing or facilitating mechanical allodynia, combinations of different serotonergic receptor agonists and antagonists were administered intrathecally. The increase in avoidance responses following m-CPBG (7.2 ^ 0.4) was reduced to 5.3 ^ 0.5 responses by a subsequent administration of 5-HT (10 mg) at 10 min (Fig. 3B). Similar to the effect of exogenous 5-HT alone, the reversal was transient, present only 5 min post-injection. Administration of 5-HT (10 mg) following pre-treatment with ondansetron resulted in a greater decrease in avoidance responses, reducing the number observed following ondansetron alone (2.4 ^ 0.3)
263
Fig. 3. Effect of acute (A) exogenous 5-HT alone or (B) following treatment with m-CPBG or ondansetron on at-level mechanical allodynia five weeks post-SCI. Baseline represents the number of avoidance responses made during the fourth post-SCI week, prior to intrathecal catheter insertion. (A) 5-HT (1, 10 and 20 mg) or saline was administered at 0 min and mechanical allodynia was assessed for the subsequent 60 min. (B) m-CPBG (20 mg) or ondansetron (20 mg) was administered at 0 min and mechanical allodynia was assessed at 5 min. 5-HT (10 or 20 mg) or saline was then administered at 10 min (indicated by 2nd injection). The values shown represent mean number of avoidance responses out of ten stimulations ^ SEM (n ¼ 5/group). *, P , 0:05 compared to saline-treated controls.
to 1.2 ^ 0.4 after the combined treatment (Fig. 3B). This partial summation of effects was also transient. Methysergide was used to block 5-HT1/5-HT2 receptors non-selectively following delivery of 5-HT3-R agonists or antagonists. Methysergide alone significantly increased the number of avoidance responses (8.2 ^ 0.4; P , 0:05) to a similar level to that observed with selective 5-HT3-R activation (Fig. 4). This increase in responses was observed only at 20 min post-injection. Administration of methysergide 10 min following pre-treatment with m-CPBG further increased the number of avoidance responses (9.2 ^ 0.4). The reduction in mechanical allodynia following ondansetron administration was reversed 20 min after methysergide injection, increasing avoidance responses from 2.6 ^ 0.5 following ondansetron alone to 5.8 ^ 0.6 following the combination treatment. 3.5. Neurotoxin-induced depletion of spinal 5-HT eliminates the effect of the 5-HT3 -R antagonist To determine whether ondansetron acted to block the action of endogenous 5-HT at the 5-HT3-R, spinal 5-HT was
264
M.A. Oatway et al. / Pain 110 (2004) 259–268
Fig. 4. Effect of post-treatment of methysergide on at-level mechanical allodynia following treatment with m-CPBG or ondansetron five weeks post-SCI. m-CPBG (20 mg), ondansetron (20 mg) or saline was administered at 0 min and mechanical allodynia was assessed at 5 min. Methysergide (30 mg) was then administered at 10 min (indicated by 2nd injection). The values shown represent mean number of avoidance responses to ten stimulations ^ SEM ðn ¼ 5/group). *, P , 0:05 compared to saline-treated controls.
depleted 4 weeks post-SCI by i.t. treatment with the neurotoxin, 5,7-DHT. At four weeks following SCI, animals received either one i.t. dose of 5,7-DHT or saline. In animals receiving saline treatment, 5-HT-immunoreactive fibres were densely packed within the superficial laminae and randomly distributed throughout laminae III and IV in the lower thoracic cord (T9-T11) immediately rostral to the lesion site (Fig. 5A). In contrast, administration of 5,7-DHT resulted in decreased 5-HT throughout the spinal cord, particularly within the at-level area of interest (T9-T11) where the dorsal horn was void of serotonergic immunoreactivity with the exception of sparsely distributed varicose fibres within the superficial laminae (Fig. 5B). Three days following pre-treatment with either 5,7-DHT or saline, toxin-treated rats responded significantly less to
the innocuous stimulus following an acute saline injection than did rats pretreated with saline (Fig. 6A). Pre-treatment with 5,7-DHT did not appear to cause adverse motor or behavioural effects during testing sessions. Acute administration of the 5-HT3-R agonist, m-CPBG, increased avoidance responses in both 5,7-DHT and saline pre-treated groups (Fig. 6B). The time courses for changes in response behaviour were identical for both groups with significant increases in avoidance responses occurring for the initial 30 min ðP , 0:05Þ: Conversely, animals with depleted spinal 5-HT had approximately the same number of avoidance responses following ondansetron treatment as those receiving only saline, for the entire 60 min time course (Fig. 6C). Animals without the neurotoxin treatment had the expected response to ondansetron, responding significantly less than did 5,7-DHT pre-treated animals ðP , 0:05Þ:
4. Discussion Our study investigated the involvement of the altered spinal serotonergic system in maintaining mechanical allodynia at the level of SCI and generated four major findings. First, selective blockade of spinal 5-HT3-R with the antagonist, ondansetron, produced a sustained reduction of at-level mechanical allodynia, whereas selective activation of 5-HT3-R with m-CPBG enhanced allodynia. Second, non-selective excitation of spinal 5-HT receptors with exogenous 5-HT transiently reduced allodynia, whereas non-selective blockade of 5-HT1/5-HT2 receptors with methysergide caused a transient increase in allodynia. Third, selective blockade of 5-HT3-R, combined with a nonselective excitation of the 5-HT1/5-HT2 receptors, further reduced at-level mechanical allodynia. Fourth, neurotoxininduced depletion of spinal 5-HT eliminated the effect of ondansetron, suggesting that the reduction in at-level mechanical allodynia is due to the blockade of serotonergic activation of the 5-HT3-R.
Fig. 5. Photomicrographs of 5-HT immunoreactivity in dorsal horn transverse sections from the 9th thoracic spinal segment of a (A) saline-treated and (B) 5,7-DHT treated rat five weeks post-SCI. Animals received either intrathecal saline or 5,7-DHT, 7 days prior to perfusion. Following 5,7-DHT, the 5-HT immunoreactive fibres were only sparsely distributed within the laminae I-IV when compared to the robust distribution of fibres in the saline-treated, injured animal. Bar ¼ 100 mm.
M.A. Oatway et al. / Pain 110 (2004) 259–268
265
pathology for the acute and chronic phases of post-SCI allodynia. Most notably, short-lasting allodynia is attributed to glutamate-induced excitation of NMDA and AMPA receptors (Xu et al., 1993; Bennett et al., 2000), accompanied by a loss of GABAergic inhibition (Weisenfeld-Hallin et al., 1997). The chronic-phase of allodynia, occurring days to months post-injury, involves neuronal hyperexcitability (Finnerup et al., 2003; Hoheisel et al., 2003) and upregulation of the anti-opioid polypeptide, cholecystokinin (Xu et al., 1994). Differentiating the pathophysiology underlying at-level versus below-level neuropathic pain is more difficult. A recent sampling of patients within the first six months after SCI reported a higher prevalence of neuropathic at-level pain than below-level pain, particularly in patients with incomplete injuries (Siddall et al., 1999). In humans, at-level neuropathic pain begins within days to weeks after SCI while below-level pain occurs within months to years post-injury (Beric et al., 1988). Our study in rats revealed a significant increase in the development of atlevel mechanical allodynia occurring at the second post-SCI week; below-level allodynia in our model develops about one week later (Bruce et al., 2002). The differences between post-SCI at-level and below-level neuropathic pain may, therefore, be associated with the differential pathophysiological substrates related to complex changes in descending pathways, particularly, in the role of serotonergic inputs. 4.2. Effects of 5-HT3-R blockade on at-level mechanical allodynia following SCI
Fig. 6. Effect of 5,7-DHT pre-treatment on at-level mechanical allodynia five weeks post-SCI. 5,7-DHT (60 mg) or saline was intrathecally administered at 28 days post-SCI and mechanical allodynia was assessed 3-7 days later following acute injection of (A) saline (B) m-CPBG or (C) ondansetron. The values shown represent the mean number of avoidance responses to ten stimulations ^ SEM ðn ¼ 6 in saline pre-treated groups; n ¼ 7 in 5,7-DHT pre-treated groups). *, P , 0:05 compared to salinetreated controls.
4.1. Mechanical allodynia following SCI The mechanisms underlying post-SCI mechanical allodynia remain controversial. The robust cascade of acute pathophysiological events following SCI, which include excitotoxicity, inflammation and cell death, results in chronic secondary maladaptive plasticity, and these may all contribute to the development of neuropathic pain (Krenz and Weaver, 1998; Eide, 1998). Evidence from ischemic and spinal injury models has revealed different
The present study demonstrates that selective blockade of spinal 5-HT 3-R using intrathecal ondansetron significantly attenuates at-level mechanical allodynia. The 5-HT3-R is an excitatory ligand-gated ion channel localized in high density within the superficial laminae of the dorsal horn (Kia et al., 1995; Morales et al., 1996). The exact neuronal populations expressing the 5-HT3-R remain controversial. 5-HT3-R immunoreactivity is not found in glutamic acid decarboxylase-expressing interneurons, but is present in primary afferent fibres. A small proportion of these fibres contain calcitonin gene-related peptide (CGRP) (Maxwell et al., 2003), and the others are proposed to be an uncharacterized subset of myelinated and unmyelinated nociceptors (Zeitz et al., 2002). Exogenous 5-HT potentiates the evoked release of substance P, CGRP and neurokinin-A from the central terminals of primary afferent fibres via 5-HT3-R activation (Saria et al., 1990; Inoue et al., 1997). Hence, although the development of mechanical allodynia after SCI is not due to sprouting of the central arbour of primary afferent fibres (Bruce et al., 2002), long-term upregulation of serotonergic fibres may maintain neuropathic pain through evoked release of pro-nociceptive peptides from primary afferents via 5-HT3-R activation. The role of the 5-HT3-R in nociception is uncertain due to the extensive range of experimental paradigms used in
266
M.A. Oatway et al. / Pain 110 (2004) 259–268
different studies. Antinociception resulting from 5-HT3-R excitation has been reported, primarily using acute pain models (Glaum et al., 1990; Alhaider et al., 1991; Paul et al., 2001). Data from models of long-term pain, most notably the formalin test, have shown that intrathecal blockade of 5-HT3-R reduces second phase, but not first phase formalin behaviours (Oyama et al., 1996; Zeitz et al., 2002), suggesting that the 5-HT3-R contributes to persistent rather than acute nociception. These findings have recently translated into the clinic where ondansetron significantly reduces pain scores in chronic neuropathic pain patients (McCleane et al., 2003). Blockade of spinal 5-HT3-R mimics the decreased mechanical and thermal evoked responses of deep dorsal horn neurons as well as the reduced formalin response observed following selective ablation of laminae I/III NK-1 receptor-expressing neurons (Suzuki et al., 2002). This result, in parallel with ours, demonstrates mechanistic similarities across various pain models suggesting that the 5-HT3-R provides a spinal target for excitatory projections from the brainstem, thus driving a facilitatory spinalbrainstem-spinal loop pathway which results in a shift towards increased spinal excitability. Descending facilitation arising from the RVM is involved in the maintenance, but not the initiation, of neuropathic pain (Burgess et al., 2002). Similarly, low-intensity electrical or chemical stimulation of cells within the nucleus raphe magnus produces a descending facilitatory influence via the ventrolateral funiculus while descending inhibition is observed following high-intensity stimulation mediated by descending serotonergic fibres in the dorsolateral funiculus (Zhuo and Gebhart, 1994; Zhuo and Gebhart, 1997). Combined with the evidence of the current study, the magnitude and time course of supraspinal serotonergic release are crucial in establishing a descending facilitatory influence. These characteristics will be important in differentiating acute and chronic pain conditions. In addition, the induction of abnormal spinal plasticity following SCI appears to upregulate spinal 5-HT pathways which may drive the maintenance of neuropathic pain via enhanced 5-HT turnover from the RVM. Ensuing excitation of spinal 5-HT3-R may enhance neurotransmitter release from primary afferent neurons, maintaining an ascending, excitatory drive to brainstem regions, thus completing the facilitatory loop. 4.3. Effects of exogenous 5-HT on at-level mechanical allodynia following SCI Although our results show that selective blockade of spinal 5-HT3-R significantly reduces at-level mechanical allodynia, we also observed a transient reduction following intrathecal exogenous 5-HT. This observation contributes to evidence that 5-HT biphasically modulates nociceptive processing through activation of individual classes of spinal receptor subtypes. Others have shown that intrathecal 5-HT
has a dose-dependent, dual effect on pain transmission with low doses reducing, and high doses increasing, response duration in the formalin test (Oyama et al., 1996). 5-HT binds with modest affinity to the spinal 5-HT3 -R (Ki ¼ 500 nM), and binds with highest affinity at spinal 5-HT1 and 5-HT2 receptor sites (approximate Ki , 10 nM) (Glennon and Dukat, 1995). The spinal 5-HT1A receptors have been proposed to be the high-affinity receptors mediating descending inhibition and the 5-HT3-R to be low affinity receptors mediating descending facilitation of nociceptive transmission (Oyama et al., 1996). A majority of evidence supports this antinociceptive (Millan, 1994; Obata et al., 2001) and anti-allodynic action (Sasaki et al., 2001; Colpaert et al., 2002) of 5-HT1A and 5-HT2A/2C receptor agonists. Our results suggest that the delivery of exogenous 5-HT may transiently reduce at-level mechanical allodynia via acute activation of 5-HT1 and 5-HT2 receptors. Similar transient antihyperalgesic effects of intrathecal 5-HT have been reported in a model of mononeuropathy (Bardin et al., 2000). Our results demonstrated that the anti-allodynic effect of 5-HT3-R blockade was enhanced transiently by exogenous 5-HT. Thus, a combined inhibition of 5-HT3-R and excitation of 5-HT1/5-HT2 receptors greatly reduced at-level mechanical allodynia. Furthermore, nonselective blockade of 5-HT1/5-HT2 receptors with methysergide exacerbated the mechanical allodynia, in a transient manner, to levels similar to those observed after m-CPBG delivery. In this condition, only the 5-HT3-R would be available for endogenous 5-HT activation. These observations could be explained by a model in which long-term alterations in the endogenous spinal 5-HT system follow SCI. Increased levels of ligand may then result in a greater participation of receptors, such as the 5-HT3-R, that normally have only modest affinity for the ligand. 4.4. Effects of the depletion of spinal 5-HT on at-level mechanical allodynia following SCI Depletion of spinal cord 5-HT levels via intrathecal administration of the neurotoxin, 5,7-DHT has been previously reported as a method for characterizing the role of 5-HT in nociceptive modulation, (Duan and Sawynok, 1987; Liu et al., 1988). We observed that depletion of spinal 5-HT by 5,7-DHT eliminated the anti-allodynic activity of ondansetron, verifying that the reduction in mechanical allodynia requires endogenous 5-HT and is due specifically to the blockade of 5-HT3-R activation by 5-HT. Therefore, although ondansetron has local anaesthetic properties (Ye et al., 1997), the reduction in mechanical allodynia by ondansetron appears to be the result of a selective blockade of spinal serotonergic activity. The depletion of spinal 5-HT levels immediately above the lesion site decreased at-level mechanical allodynia. Similarly, the development of autotomy following sciatic nerve transection is suppressed
M.A. Oatway et al. / Pain 110 (2004) 259–268
by selective neurotoxic depletion of lumbosacral 5-HT (Feria et al., 1992). To our knowledge, ours is the first study to investigate the role of spinal serotonergic depletion on SCI-induced mechanical allodynia and our data provide further evidence that the increased serotonergic fibre density immediately rostral to the lesion site is involved in the maintenance of at-level mechanical allodynia. However, the residual at-level allodynia following spinal 5-HT depletion suggests that, in the absence of this mediator, alternative mechanisms may develop.
5. Summary Abnormal neuroplasticity resulting from SCI results in changes in serotonergic fibre density and distribution that directly affect the activation of post-synaptic spinal 5-HT receptors. The present study shows that the increased density of serotonergic fibres immediately rostral to the lesion site contributes to the maintenance of at-level mechanical allodynia via activation of the 5-HT3-R. Patients reporting the presence of at-level mechanical allodynia are generally refractory to a majority of analgesics, although some investigations suggest benefits of intrathecal baclofen in central, deafferentation pain (Taira et al., 1995; Gatscher et al., 2002). Based on the pronociceptive actions of 5-HT that we have shown, as well as the present clinical availability of 5-HT3-R antagonists, our study suggests a new avenue of focus for pharmacological treatment of at-level neuropathic pain following SCI.
Acknowledgements This research was supported by grants from the Ontario Neurotrauma Foundation and the Canadian Institutes of Health Research. M.A.O. was supported by a studentship award from the Ontario Neurotrauma Foundation. We thank Drs. Arthur Brown, Canio Polosa and Jana Sawynok for their constructive criticisms of this manuscript.
References Alhaider AA, Lei SZ, Wilcox GL. Spinal 5-HT3 receptor-mediated antinociception: possible release of GABA. J Neurosci 1991;11: 1881–8. Bardin L, Schmidt J, Alloui A, Eschalier A. Effect of intrathecal administration of serotonin in chronic pain models in rats. Eur J Pharmacol 2000;409:37– 43. Bennett AD, Everhart AW, Hulsebosch CE. Intrathecal administration of an NMDA or a non-NMDA receptor antagonist reduces mechanical but not thermal allodynia in a rodent model of chronic central pain after spinal cord injury. Brain Res 2000;859:72 –82. Beric A, Dimitrijevic MR, Lindblom U. Central dysesthesia syndrome in spinal cord injury patients. Pain 1988;34:109 –16. Bruce JC, Oatway MA, Weaver LC. Chronic pain after clip compression injury of the rat spinal cord. Exp Neurol 2002;178:33 –48.
267
Burgess SE, Gardell LR, Ossipov MH, Malan TP, Jr., Vanderah TW, Lai J, Porreca F. Time-dependent descending facilitation from the rostral ventro medial medulla maintains but does not initiate, neuropathic pain. J Neurosci 2002;22:5129–36. Colpaert FC, Tarayre JP, Koek W, Pauwels PJ, Bardin L, Xu XJ, Wiesenfeld-Hallin Z, Cosi C, Carilla-Durand E, Assie MB, Vacher B. Large-amplitude 5-HT1A receptor activation: a new mechanism of profound, central analgesia. Neuropharmacology 2002;43:945–58. Duan J, Sawynok J. Enhancement of clonidine-induced analgesia by lesions induced with spinal and intracerebroventricular administration of 5,7dihydroxytryptamine. Neuropharmacology 1987;26:323– 9. Eide PK. Pathophysiological mechanisms of central neuropathic pain after spinal cord injury. Spinal Cord 1998;36:601–12. Feria M, Sanchez A, Abad F, Abreu P. Effects of selective neurotoxic lesion of lumbosacral serotonergic and noradrenergic systems on autotomy behaviour in rats. Pain 1992;51:101–9. Finnerup NB, Johannesen IL, Fuglsang-Frederiksen A, Bach FW, Jensen TS. Sensory function in spinal cord injury patients with and without central pain. Brain 2003;126:57–70. Gatscher S, Becker R, Uhle E, Bertalanffy H. Combined intrathecal baclofen and morphine infusion for the treatment of spasticity related pain and central deafferentiation pain. Acta Neurochir Suppl 2002;79: 75– 6. Glaum SR, Proudfit HK, Anderson EG. 5-HT3 receptors modulate spinal nociceptive reflexes. Brain Res 1990;510:12–16. Glennon RA, Dukat M. Serotonin receptor subtypes. In: Bloom FE, Kupfer DJ, editors. Psychopharmacology: The fourth generation of progress. New York: Raven Press; 1995. p. 415–29. Green GM, Scarth J, Dickenson A. An excitatory role for 5-HT in spinal inflammatory nociceptive transmission; state-dependent actions via dorsal horn 5-HT(3) receptors in the anaesthetized rat. Pain 2000;89: 81– 8. Hains BC, Johnson KM, McAdoo DJ, Eaton MJ, Hulsebosch CE. Engraftment of serotonergic precursors enhances locomotor function and attenuates chronic central pain behavior following spinal hemisection injury in the rat. Exp Neurol 2001;171:361–78. Hains BC, Everhart AW, Fullwood SD, Hulsebosch CE. Changes in serotonin, serotonin transporter expression and serotonin denervation supersensitivity: Involvement in chronic central pain after spinal hemisection in the rat. Exp Neurol 2002;175:347– 62. Hamon M, Bourgoin S. Serotonin and Its Receptors in Pain Control. In: Sawynok J, Cowan A, editors. Novel Aspects of Pain Management: Opioids and Beyond. Wiley; 1999. p. 203 –28. Hao JX, Xu XJ, Aldskogius H, Seiger A, Wiesenfeld-Hallin Z. Allodynialike effects in rat after ischaemic spinal cord injury photochemically induced by laser irradiation. Pain 1991;45:175–85. Hoheisel U, Scheifer C, Trudrung P, Unger T, Mense S. Pathophysiological activity in rat dorsal horn neurones in segments rostral to a chronic spinal cord injury. Brain Res 2003;974:134–45. Horiuchi H, Ogata T, Morino T, Takeba J, Yamamoto H. Serotonergic signaling inhibits hyperalgesia induced by spinal cord damage. Brain Res 2002;963:312– 20. Howe JR, Yaksh TL. Changes in sensitivity to intrathecal norepinephrine and serotonin after 6-hydroxydopamine (6-OHDA) 5,6-dihydroxytryptamine (5,6-DHT) or repeated monoamine administration. J Pharmacol Exp Ther 1982;220:311– 21. Inman DM, Steward O. Ascending sensory, but not other long-tract axons, regenerate into the connective tissue matrix that forms at the site of a spinal cord injury in mice. J Comp Neurol 2003;462:431–49. Inoue A, Hashimoto T, Hide I, Nishio H, Nakata Y. 5-Hydroxytryptaminefacilitated release of substance P from rat spinal cord slices is mediated by nitric oxide and cyclic GMP. J Neurochem 1997;68:128–33. Kia HK, Miquel MC, McKernan RM, Laporte AM, Lombard MC, Bourgoin S, Hamon M, Verge D. Localization of 5-HT3 receptors in the rat spinal cord: immunohistochemistry and in situ hybridization. Neuroreport 1995;6:257–61.
268
M.A. Oatway et al. / Pain 110 (2004) 259–268
Krenz NR, Weaver LC. Sprouting of primary afferent fibers after spinal cord transection in the rat. Neurosci 1998;85:443 –58. Liu MY, Su CF, Lin MT. The antinociceptive role of a bulbospinal serotonergic pathway in the rat brain. Pain 1988;33:123–9. Mason P. Central mechanisms of pain modulation. Curr Opin Neurobiol 1999;9:436–41. Maxwell DJ, Kerr R, Rashid S, Anderson E. Characterisation of axon terminals in the rat dorsal horn that are immunoreactive for serotonin 5-HT3A receptor subunits. Exp Brain Res 2003;149:114–24. McCleane GJ, Suzuki R, Dickenson AH. Does a single intravenous injection of the 5HT3 receptor antagonist ondansetron have an analgesic effect in neuropathic pain? A double-blinded, placebocontrolled cross-over study. Anesth Analg 2003;97:1474–8. Millan MJ. Serotonin and pain: evidence that activation of 5-HT1A receptors does not elicit antinociception against noxious thermal, mechanical and chemical stimuli in mice. Pain 1994;58:45–61. Millan MJ. Descending control of pain. Prog Neurobiol 2002;66:355–474. Morales M, Battenberg E, de Lecea L, Sanna PP, Bloom FE. Cellular and subcellular immunolocalization of the type 3 serotonin receptor in the rat central nervous system. Brain Res Mol Brain Res 1996;36:251– 60. Obata H, Saito S, Sasaki M, Ishizaki K, Goto F. Antiallodynic effect of intrathecally administered 5-HT(2) agonists in rats with nerve ligation. Pain 2001;90:173 –9. Ossipov M, Lai J, Malan JrT, Vanderah T, Porreca F. Tonic descending facilitation as a mechanism of neuropathic pain. Neuropathic Pain: pathophysiology and treatment. Seattle: International Association for the Study of Pain; 2001. p. 107–24. Oyama T, Ueda M, Kuraishi Y, Akaike A, Satoh M. Dual effect of serotonin on formalin-induced nociception in the rat spinal cord. Neurosci Res 1996;25:129–35. Parisod E, Siddall PJ, Viney M, McClelland JM, Cousins MJ. Allodynia after acute intrathecal morphine administration in a patient with neuropathic pain after spinal cord injury. Anesth Analg 2003;97:183– 6. Paul D, Yao D, Zhu P, Minor LD, Garcia MM. 5-hydroxytryptamine3 (5-HT3) receptors mediate spinal 5-HT antinociception: an antisense approach. J Pharmacol Exp Ther 2001;298:674–8. Rivlin AS, Tator CH. Objective clinical assessment of motor function after experimental spinal cord injury in the rat. J Neurosurg 1977;47:577– 81. Saria A, Javorsky F, Humpel C, Gamse R. 5-HT3 receptor antagonists inhibit sensory neuropeptide release from the rat spinal cord. Neuroreport 1990;1:104–6. Sasaki M, Ishizaki K, Obata H, Goto F. Effects of 5-HT2 and 5-HT3 receptors on the modulation of nociceptive transmission in rat spinal cord according to the formalin test. Eur J Pharmacol 2001;424:45– 52. Sawynok J, Reid A. Spinal supersensitivity to 5-HT1, 5-HT2 and 5-HT3 receptor agonists following 5,7-dihydroxytryptamine. Eur J Pharmacol 1994;264:249–57. Siddall PJ, Loeser JD. Pain following spinal cord injury. Spinal Cord 2001; 39:63–73. Siddall PJ, Xu CL, Floyd N, Keay KA. C-fos expression in the spinal cord of rats exhibiting allodynia following contusive spinal cord injury. Brain Res 1999;851:281–6.
Suzuki R, Morcuende S, Webber M, Hunt SP, Dickenson AH. Superficial NK1-expressing neurons control spinal excitability through activation of descending pathways. Nat Neurosci 2002;5:1319–26. Taira T, Kawamura H, Tanikawa T, Kawabatake H, Iseki H, Ueda A, Takakura K. A new approach to the control of central deafferentation pain–spinal intrathecal baclofen. Acta Neurochir Suppl (Wien) 1995; 64:136–8. Vierck Jr CJ, Siddall P, Yezierski RP. Pain following spinal cord injury: animal models and mechanistic studies. Pain 2000;89:1–5. Warms CA, Turner JA, Marshall HM, Cardenas DD. Treatments for chronic pain associated with spinal cord injuries: many are tried, few are helpful. Clin J Pain 2002;18:154–63. Weaver LC, Verghese P, Bruce JC, Fehlings MG, Krenz NR, Marsh DR. Autonomic dysreflexia and primary afferent sprouting after clipcompression injury of the rat spinal cord. J Neurotrauma 2001;18: 1107–19. Weisenfeld-Hallin Z, Aldskogius H, Grant G, Hao JX, Hokfelt T, Xu XJ. Central inhibitory dysfunctions: mechanisms and clinical implications. Behav Brain Sci 1997;20(3):420– 5. ˚ , Wiesenfeld-Hallin Z. Systemic excitatory amino Xu X-J, Hao JX, Seiger A acid receptor antagonists of the (-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor and of the N-methyl-Daspartate (NMDA) receptor relieve mechanical hypersensitivity after transient spinal cord ischemia in rats. J Pharm Exper Therap 1993; 267(1):140–4. ˚ , Hughes J, Ho¨kfelt T, Wiesenfeld-Hallin Z. Xu X-J, Hao JX, Seiger A Chronic pain-related behaviors in spinally injured rats: evidence for functional alterations of the endogenous cholecystokinin and opioid systems. Pain 1994;56:271–7. Ye JH, Mui WC, Ren J, Hunt TE, Wu WH, Zbuzek VK. Ondansetron exhibits the properties of a local anesthetic. Anesth Analg 1997;85: 1116–21. Yezierski RP. Pain following spinal cord injury: the clinical problem and experimental studies. Pain 1996;68:185–94. Yezierski RP. Pain following spinal cord injury: pathophysiology and central mechanisms. Prog Brain Res 2000;129(429-49): 429 – 49. Zeitz KP, Guy N, Malmberg AB, Dirajlal S, Martin WJ, Sun L, Bonhaus DW, Stucky CL, Julius D, Basbaum AI. The 5-HT3 subtype of serotonin receptor contributes to nociceptive processing via a novel subset of myelinated and unmyelinated nociceptors. J Neurosci 2002; 22:1010– 9. Zhuo M, Gebhart GF. Spinal serotonin receptors mediate descending facilitation of a nociceptive reflex from the nuclei reticularis gigantocellularis and gigantocellularis pars alpha in the rat. Brain Res 1994;550:35–48. Zhuo M, Gebhart GF. Biphasic modulation of spinal nociceptive transmission from the medullary raphe nuclei in the rat. J Neurophysiol 1997;78:746–58. Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;16:109 –10.