- Email: [email protected]
Riluzole Normalizes Early-Life Stress-Induced Visceral Hypersensitivity in Rats: Role of Spinal Glutamate Reuptake Mechanisms
GASTROENTEROLOGY 2010;138:2418 –2425
Riluzole Normalizes Early-Life Stress-Induced Visceral Hypersensitivity in Rats: Role of Spinal Glutamate Reuptake Mechanisms ROMAIN–DANIEL GOSSELIN,* RICHARD M. O’CONNOR,* MONICA TRAMULLAS,* MARCELA JULIO–PIEPER,* TIMOTHY G. DINAN,*,‡ and JOHN F. CRYAN*,§,储 *Laboratory of Neurogastroenterology, Alimentary Pharmabiotic Centre; ‡Departments of Psychiatry and §Pharmacology and Therapeutics, and 储School of Pharmacy, University College Cork, Cork, Ireland
BASIC– ALIMENTARY TRACT
BACKGROUND & AIMS: The molecular basis underlying visceral hypersensitivity in functional irritable bowel syndrome remains elusive, resulting in poor treatment effectiveness. Because alterations in spinal non-neuronal (astrocytic) glutamate reuptake are suspected to participate in chronic pain, we asked whether such processes occur in visceral hypersensitivity. METHODS: Visceral hypersensitivity was induced in Sprague–Dawley rats by maternal separation. Separated adults were given a systemic administration of riluzole (5 mg/kg), an approved neuroprotective agent activating glutamate reuptake. Visceral hypersensitivity was assessed using colorectal distension (40 mm Hg). Somatic nociception was quantified using Hot Plate, Randall–Sellito, and Hargreaves tests. Spinal proteins were quantified using immunofluorescence and Western blot. The dependence of visceral sensory function upon spinal glutamate transport was evaluated by intrathecal injection of glutamate transport antagonist DL-threo--benzyloxyaspartate (TBOA). For in vitro testing of riluzole and TBOA, primary cultures of astrocytes were used. RESULTS: We show that riluzole counteracts stress-induced visceral hypersensitivity without affecting visceral response in nonseparated rats or altering nociceptive responses to somatic pain stimulation. In addition, maternal separation produces a reduction in glial excitatory amino acid transporter (EAAT)-1 with no change in EAAT-2 or ␥-amino butyric acid transporters. Stress was not associated with changes in glial fibrillary acidic protein or astrocytic morphology per se. Furthermore, visceral normosensitivity relies on spinal EAAT, as intrathecal TBOA is sufficient to induce hypersensitivity in normal rats. CONCLUSIONS: We identify spinal EAAT as a therapeutic target, and establish riluzole as a candidate to counteract gastrointestinal hypersensitivity in disorders such as irritable bowel syndrome. Keywords: Irritable Bowel Syndrome; Maternal Separation; Astrocytes; Spinal Cord.
V
isceral hypersensitivity is a core symptom of functional gastrointestinal disorders such as irritable bowel syndrome (IBS) and is a pressing clinical issue that remains an ongoing challenge for the pharmaceutical industry.1,2 The ineffectiveness of current medications against visceral hypersensitivity is attributable at least in
part to the lack of a unifying hypothesis regarding the chain of events generating pain. Despite this absence of consensus, stress occurring in early life is known to have a key detrimental upstream influence on nociceptive pathways and in the manifestation of IBS symptoms.3–5 Hence, the enhanced visceromotor response following pup isolation from dams (maternal separation) is viewed as an IBS-relevant model in animals.6 In addition, central sensitization (ie, an increased activation of spinal nociceptive neurons to peripheral stimuli) may be an end mechanism accounting for visceral hyperalgesia in IBS,7 presumably due to a sustained glutamatergic sensory neurotransmission. However, glutamate receptor antagonists show a constellation of side effects strongly restricting their utilization in patients. Interestingly, beyond glutamate receptor targeting, the efficacy of glutamatergic neurotransmission may be modulated by acting on glutamate reuptake systems. Indeed, the proper and regulated action of glutamate relies on its fast elimination from extracellular synaptic milieu as a result of the scavenging machinery, namely the excitatory amino-acid transporters (EAAT)-1 and EAAT-2.8 These transporters are expressed by non-neuronal glial cells, primarily astrocytes, whose endfeet cover virtually all synapses in the central nervous system. Remarkably, the importance of spinal EAAT in nociception has been emphasized by recent evidence showing that selective intrathecal inhibition of glial glutamate transport in naïve rats results in pain behaviors.9 Moreover, a reduction in EAAT expression occurs in chronic pain states10 –12 as a component of the profound phenotypic alterations taking place in spinal astrocytes, which encompasses an increased synthesis of glial fibrillary acidic protein (GFAP) and a release in sensitizing soluble factors.13 These alterations are consistently observed in a variety of pain models, giving astrocytes an emerging key role in the maintenance of chronic pain. However, there is a paucity of data regarding the Abbreviations used in this paper: EAAT, excitatory amino-acid transporter; GAT, ␥-amino butyric acid transporter; GFAP, glial fibrillary acidic protein; IBS, irritable bowel syndrome; PBS, phosphate-buffered saline; TBOA, DL-threo--benzyloxyaspartate. © 2010 by the AGA Institute 0016-5085/$36.00 doi:10.1053/j.gastro.2010.03.003
involvement of astrocytes in visceral pain, especially with regard to the possible role of glutamate uptake system. Riluzole (2-amino-6-trifluoromethoxybenzothiazole) is a neuroprotective drug already approved for amyotrophic lateral sclerosis.14 Riluzole is a potent activator of glutamate reuptake both in vitro and in vivo. In addition, pain-modulatory properties of riluzole have been reported,11,15 but its influence on visceral pain remains unknown. Therefore, in the present study, we assessed whether riluzole could reduce visceral hypersensitivity induced by maternal separation and how this could shed some insight on underlying mechanisms in stress-induced visceral pain. We found that riluzole given intraperitoneally normalizes visceral hypersensitivity in maternally separated rats. Moreover, it failed to affect visceral responses in nonseparated animals and had no effects on somatic nociception; maternal separation produces a reduction in EAAT-1 in the spinal cord; stress did not result in changes in EAAT-2, ␥-amino butyric acid transporter (GAT), and ␥-amino butyric acid expression or in astrocytic morphological alterations; direct intrathecal inhibition of EAAT using DL-threo--benzyloxyaspartate (TBOA) is sufficient to produce a visceral hypersensitivity in naïve rats. Taken together, the data presented here add another dimension to the use of riluzole in therapeutics, establishing its relevance to counteract visceral hypersensitivity, and point to spinal glial glutamate reuptake as a pharmacological target for pain in functional bowel disorders.
Materials and Methods Animals Male Sprague–Dawley rats weighing 280 –300 g were housed and bred in the local animal facility with food and water ad libidum, on a 12:12-hour reversed dark–light cycle with temperature at 20°C ⫾ 1°C. Animals were group-housed by 4 to 5 per cage in usual plastic cages with sawdust bedding in an enriched environment with shredded paper and a cardboard roll. All experiments were in full accordance with the European Community Council Directive (86/609/EEC). Behavioral assessments were always carried out randomly by an experimenter blind to treatment groups.
Drug Administration Riluzole and TBOA were purchased from Tocris (Bristol, UK). For intrathecal injection, rats were sedated by light isofurane anesthesia and vehicle (phosphatebuffered saline [PBS], 10 L) or TBOA (1.2 g or 12 g in 10 L vehicle) was administrated. Correct positioning was assessed by a tail flick upon needle insertion. Balloons were immediately inserted in preparation for colorectal distension. The absence of spinal damage was assessed prior to distension, and animals presenting any sign of paralysis or spontaneous pain behaviors were immediately culled. Colorectal distensions were carried out 8 –10 minutes after intrathecal injections. This tim-
RILUZOLE AND VISCERAL HYPERSENSITIVITY
2419
ing was selected from previous studies that showed that it was at the longer range used to see maximal effects of TBOA delivered intrathecally.9 For riluzole studies, rats were weighed and a slight volume correction was applied to the stock solution (1.5 mg/mL, in PBS) for the intraperitoneal injection to achieve a final administration of 5 mg/kg riluzole. Control rats were injected with vehicle (1 mL PBS). Colorectal distension was carried out 30 minutes later. For nociceptive testing (Hot Plate, Randall– Sellito, and Hargreaves tests), riluzole was injected 30 minutes prior to testing.
Colorectal Distension Animals were sedated using isofurane, and 7-cm–long deflated balloons attached to polyethylene tubing were inserted intrarectally through the anus. Rats were left to recover for 10 minutes and the tubing was connected to a barostat machine. The protocol consisted in a tonic distension at a constant pressure of 40 mm Hg during 10 minutes. Visceromotor response to distension was quantified as the number of abdominal contractions resulting in animal arching during the procedure. At the end of the distension, balloons were deflated and the animals were culled.
Hot-Plate Test Rats were placed with all 4 paws on the hot plate (Incremental Hot Plate; Stoelting, Dublin, Ireland) heated at 55°C and the latency to hindpaw lick or shake was measured. The cutoff time was set at 20 seconds to avoid tissue damage. Baseline latencies were recorded 30 minutes prior to the experiment.
Paw Pressure Test, Randall–Sellito Responses to noxious mechanical stimuli were measured with a Digital Randall–Sellito test (Stoelting). Rats were handled and partly restrained. Following forepaw positioning, an increasing pressure was applied to the dorsal surface. Nociceptive threshold was defined as the pressure in grams eliciting the paw withdrawal. Measurements were performed in triplicate. A cutoff weight was set at 400 g. Baseline responses were recorded 1 day before drug administration.
Paw Withdrawal Test, Hargreaves Paw withdrawal response to noxious radiant heat stimuli was assessed using Plantar Test (Plantar Test Analgesia Meter; Stoelting). Rats were placed in individual Plexiglas chambers on a glass platform heated at 37°C and habituated for 5 minutes. A mobile radiant heat source was located under the platform and focused onto the mid-plantar surface of each hindpaw sequentially. The paw withdrawal latency was defined as the time from radiant heat onset to paw withdrawal. The apparatus was calibrated to give a paw withdrawal latency of 7–9 seconds prior to drug injection. A total of 6 paw withdrawal latencies (in triplicate, each separated by at
BASIC– ALIMENTARY TRACT
June 2010
2420
GOSSELIN ET AL
least 5 minutes) were measured. A cutoff time of 20 seconds was used to avoid tissue damage. Baseline responses were taken the day before drug administration.
Rotarod Testing of Motor Coordination
BASIC– ALIMENTARY TRACT
All animals were trained prior to accessing performance on the rotarod (HARotaRod System). Rotating speed was fixed at 12 rpm. Training comprised of 2 days: on day 1, animals were placed on the rotarod (perpendicular to the axis of rotation, facing away from the direction of rotation so that the animal had to walk forward to stay on the bar) until they completed 1 trial of 180 seconds; on day 2, animals completed 3 sessions of 180 seconds within 1 hour. After the final trial on day 2, animals were placed on the rotarod multiple times for habituation, until the rats stepped voluntarily onto the rotating bar. On the test day (day 3), 3 trials were conducted, all trials having a cutoff point of 180 seconds. Trial 1 was conducted 3 minutes prior to treatment; any animals that did not successfully complete trial 1 were omitted from the study. Trial 2 was conducted 30 minutes after treatment and trial 3 was conducted 180 minutes after treatment. For riluzole treatment, final doses of 2.5 mg/kg, 5.0 mg/kg, and 10 mg/kg were used. Control animals were injected with 1 mL vehicle alone (PBS). All trials were scored manually by the researcher.
Maternal Separation Maternal separation was performed as described previously.16 Briefly, rat pups were separated from their dams as a whole litter during 180 minutes every day over a period of 10 days between postnatal days 2 and 12. Separations were conducted between 9 AM and 12 AM in plastic cages placed on top of heater pads (30°C–33°C) and placed in a separated room. Control rats consisted of nonhandled pups, left untouched with their mothers. After postnatal day 12, rats were left undisturbed except for routine cage cleaning every two days until they were 8 –10 weeks old.
Immunolabeling Rabbit primary antibodies against EAAT-1 and EAAT-2 were from Abcam (Cambridge, UK); rabbit antibodies against GFAP, GAT-1, and GAT-3 were from Chemicon-Millipore (Tullagreen, Ireland); mouse antiGFAP was from Cell Signaling (Danvers, MA); and mouse anti-actin was purchased from Sigma (Dublin, Ireland). Secondary antibodies were from Jackson Immunoresearch (Newmarket, Suffolk, UK). Immunofluorescence was carried out as described previously17 with minor modifications. Briefly, rats were fixed by transcardial perfusion of 4% paraformaldehyde in PBS under lethal anesthesia. Postfixed and sucrose-cryoprotected samples were frozen in dry ice and cut in 18-m slices using a cryostat (Leica, Ashbourne, Ireland). Mounted slices were blocked using 10% normal goat serum in PBS 0.05% Triton x-100 and incubated overnight at 4°C with the
GASTROENTEROLOGY Vol. 138, No. 7
primary antibody in the same blocking solution (rabbit anti-GFAP: 1/2000; rabbit anti–EAAT-1: 1/1000; mouse anti-GFAP 1/500). After extensive washing, the slides were incubated with fluorescent dye-coupled secondary antibody, washed, and mounted in fluorescence medium (DAKO, Hazel Grove, UK). Images were taken using the Cell-F software (Olympus, Dublin, Ireland) and analyzed using ImageJ software (National Institutes of Health, http://rsb.info.nih.gov/ij/).
Western Blot The Western blot procedure was conducted as described previously.17 Briefly, spinal cords were frozen and stored at ⫺80°C until protein extraction and quantification. Twenty micrograms of proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred onto a polyvinylidene fluoride membrane, and blocked (PBS with 0.1% Tween 20 and 5% Milk). For immunodetection, overnight incubation at 4°C with primary antibodies (1/2000) was performed in blocking solution. After washing, membranes were incubated with horseradish peroxidase– conjugated secondary antibodies (1/10,000) and in electrochemiluminescence reagents (Pierce, IL). Images were obtained using a luminescent image analyzer (LAS-3000; Fugifilm, Ireland). For the subsequent detection of -actin, the membranes were processed in the same manner as in mouse anti-actin antibody (1/5000; Sigma, Dublin, Ireland). Immunoblots were quantified using ImageJ software; 5 rats were used in each group.
Primary Culture of Astrocytes and Aspartate Reuptake Spinal cords were collected in Dulbecco’s modified Eagle’s medium (Sigma, Dublin, Ireland) under aseptic conditions from decapitated newborn rats at postnatal days 3 to 4. Tissues were mechanically dissociated in Dulbecco’s modified Eagle’s medium supplemented with 5% fetal bovine, L-glutamine (2 mM; Sigma), and gentamycin (Sigma). The cell suspension was plated onto polyD-lysine– coated flasks (T-75) using 1 spinal cord per flask and kept in a 95% CO2/5% H2O incubator at 37°C. Medium was changed by half every 3 days, and after 7 or 10 days in vitro, flasks were shaken 3 hours at 200 rpm to detach microglial cells. Adherent astrocytes were trypsinized and seeded at 5.104 cells/mL on poly-Dlysine– coated 6-well microplates. Medium was changed by half every 2 days with complete medium without serum to improve cell differentiation, and astrocytes were used 6 days after their passage. The enrichment in astrocytes was confirmed by a GFAP immunolabeling ⬎95%. Reuptake experiments were conducted as follows. Medium was fully changed, and astrocytes were first equilibrated 10 minutes at 37°C with prewarmed Hank’s Balanced Salt Solution (Sigma, Dublin, Ireland) completed with MgCl2 (1.2 mM) and HEPES (10 mM) and supplemented or not with TBOA or riluzole. Radiolabeled aspartate (50 nM final, [3H]D-aspartate, 12.2 Ci/mmol;
June 2010
RILUZOLE AND VISCERAL HYPERSENSITIVITY
2421
Amersham, Buckinghamshire, UK) was added simultaneously with nonradioactive aspartate (5 M final) from prewarmed 10⫻ solutions in Hank’s Balanced Salt Solution. After 7 minutes, the transport was stopped by 2 rinses using ice-cold Hank’s Balanced Salt Solution and cells were lysed with 1 mL 0.5 M NaOH ⫹ 0.05% sodium dodecyl sulfate in water. Protein content was evaluated using Qubit technology (Invitrogen, Carlsbad, CA), and reuptake activity was expressed in cpm/mg of proteins.
Statistical Analysis
Results Riluzole Counteracts Stress-Induced Visceral Hypersensitivity We explored the possibility that increasing glutamate transport could reverse stress-induced visceral hypersensitivity following maternal separation. Vehicle-treated separated rats show a higher response to colorectal distension than vehicle-treated nonseparated animals, in line with previously described works16 (F(3,28) ⫽ 4.058; P ⬍ .05, Figure 1A). Remarkably, single intraperitoneal administration of riluzole (5 mg/kg) counteracts maternal separation-associated hypersensitivity, restoring a number of abdominal contractions comparable to vehicle-treated nonseparated animals. Notably, riluzole had no detectable effect in nonseparated rats (F(3,28) ⫽ 4.058; P ⬎ .05), suggesting that riluzole reverses specific pronociceptive alterations in maternal separation without modulating the entire nociceptive pathway. In order to specifically address this point, we performed testing for mechanical and thermal nociception. As shown in Figure 1B, riluzole has no effect on thermal or mechanical nociception in naïve rats in Hot-Plate (F(5,49) ⫽ 0.166), Randall–Sellito (F(5,42) ⫽ 0.126), or Hargreaves (F(5,42) ⫽ 0.572) tests. Some authors have reported impairment in motor coordination (sedative or ataxic effect) following riluzole,18 whereas others have observed no effect.11,15,19 Therefore, in order to rule out the possibility that the riluzole effect we observe could be due to such a process, we evaluated the consequences of riluzole on motor coordination in naïve rats using the rotarod test. As shown in Figure 1C, riluzole treatment has no influence on motor performances at either 30 minutes or 180 minutes postinjection.
Figure 1. Riluzole reverses stress-induced visceral hypersensitivity. (A) Maternally separated (MS) rats and nonseparated (NS) controls were treated with vehicle or riluzole (5 mg/kg, intraperitoneally), and visceral sensitivity to colorectal distension (10 minutes at 40 mm Hg) was determined as the number of abdominal contractions. An increase in abdominal contractions in response to distension is observed in separated animals. Riluzole counteracts the visceral hypersensitivity observed in separated rats (F(3,28) ⫽ 4.058; Newman–Keuls test, *P ⬍ .05) with no detectable effect in nonseparated animals (n ⫽ 9 for vehicle NS and MS, n ⫽ 8 and 6 for riluzole NS and riluzole MS, respectively). (B) Riluzole has no effect on acute nociception in the Hot Plate, Randall– Sellito, or Hargreaves tests (n ⫽ 8 per group). (C) Rotarod testing following administration of various doses of riluzole shows no impairment of motor coordination after treatment with 2.5, 5, or 10 mg/kg riluzole.
Functional analysis in primary cultures of rat astrocytes confirms that riluzole increases the activity of EAAT (Figure 2). Reuptake of glutamate analogue [3H]D-aspartate was dose-dependently increased by riluzole (up to a 37% augmentation for a riluzole concentration of 1 M), an effect fully blocked by preincubation with TBOA (data not shown). At higher doses of riluzole, this effect was not observed as described previously.20,21 The expressions of glutamate transporters EAAT-1 and EAAT-2 in our in vitro system were confirmed by Western blot (Figure 2C). Together, these results show that riluzole is effective at reversing stress-induced visceral hypersensitivity and, given the activating effect of riluzole on glutamate transport, this suggests that glutamate reuptake could be altered as a result of maternal separation.
BASIC– ALIMENTARY TRACT
Data from pixel quantifications (immunofluorescence and Western blot) were compared using unpaired 2-tailed Student’s t test. Behavioral as well as reuptake results were analyzed using 1-way analyses of variance. Analyses of variance were followed by Newman–Keuls or Dunnett’s multiple comparison post-hoc tests when appropriate. All analyses were performed using GraphPadPrism (GraphPad Software, San Diego, CA). Differences were considered significant for P ⬍ .05. Data are expressed as mean ⫾ standard error of mean. No outlier animal or sample was discarded from any group.
2422
GOSSELIN ET AL
GASTROENTEROLOGY Vol. 138, No. 7
Basal Spinal Glutamate Reuptake Is Required to Prevent Visceral Hypersensitivity
Figure 2. Effect of riluzole on [3H]D-aspartate transport in vitro. (A) Riluzole dose-dependently increases [3H]D-aspartate reuptake in cultured astrocytes with a significant effect for 1 M riluzole (F(8,27) ⫽ 29.8; Dunnett’s test, *P ⬍ .05 vs no riluzole treatment; n ⫽ 4 wells). (B) Western blots showing that primary cultures of spinal astrocytes expressing the astrocytic marker glial fibrillary acidic protein (GFAP) also express the glutamate transporters excitatory amino-acid transporter (EAAT)-1 and EAAT-2.
In order to determine the consequence of spinal glutamate transport reduction on the maintenance of basal visceral sensitivity, we intrathecally administrated the glutamate transport blocker TBOA and monitored visceromotor response in naïve rats (Figure 5A). Intrathecal TBOA produces a dose-dependent increase in visceral sensitivity to distention with a 2-fold augmentation in contractions for an injection of 12 ng TBOA (F(2,30) ⫽ 6.737; P ⬍ .01 vehicle vs TBOA 12 ng). No other behavioral effect of TBOA was detected. The ability of TBOA to inhibit EAAT was further confirmed in vitro by functional reuptake analysis of the nonbiologically active, but transportable glutamate analogue [3H]D-aspartate in primary cultures of rat astrocytes (Figure 4B). As described previously, TBOA produces a robust reduction in the uptake of [3H]D-aspartate (P ⬍ .0001).
Maternal Separation Results in a Selective Reduction in Spinal EAAT-1 Expression
BASIC– ALIMENTARY TRACT
We investigated whether maternal separation affects spinal astrocytic glutamate transporter integrity. A significant reduction in EAAT-1 immunoreactivity is detected by both Western blot (P ⬍ .0028, Figure 3A) and immunofluorescence (P ⬍ .0461, Figure 3B) in the spinal cord of separated animals. The reduction in EAAT-1 immunoreactivity is observed in all dorsal spinal layers (target of the sensory input), even though EAAT-1 immunoreactivity retains its classical pattern with a higher expression in the superficial layers (Figure 3B). Remarkably, no change was detected in the other astroglial glutamate transporter, EAAT-2 or GAT-1 and -3 (Figure 3A). Finally, dual-immunofluorescence shows that GFAP colocalizes with EAAT-1 (Figure 3C).
Maternal Separation Does Not Induce a Global Change in Spinal Astrocytic Phenotype Chronic pain is associated with alterations in spinal astrocytic phenotype, the core change of which is the up-regulation in GFAP.22 Yet, spinal GFAP expression has not been studied in visceral pain so far. Strikingly, we found that early-life stress did not influence spinal GFAP expression (Figure 4). Indeed, tissues from separated and nonseparated animals exhibit comparable intensities in GFAP immunoreactivity and similar densities of GFAPexpressing cells (P ⫽ .5581 and P ⫽ .9628, respectively; Figure 4A and B). This absence of variation in GFAP expression was further confirmed by Western blot analysis (P ⫽ .2246; Figure 4C). Because perturbations in astrocytic physiology may be linked to a modification in the apparent cell morphology, we also quantified the mean length of GFAP containing processes and show a lack of effect of early-life stress as well (P ⫽ .5843; Figure 4C).
Figure 3. Maternal separation is associated with a reduction in excitatory amino-acid transporter (EAAT)-1 in spinal astrocytes. (A) Western-blot analysis of spinal EAAT and ␥-amino butyric acid transporter (GAT) in separated rats. A significant decrease in spinal EAAT-1 is detected in comparison to nonseparated rats (**P ⬍ .01, Student’s t test, n ⫽ 5), with no change in EAAT-2 or GAT transporters. (B) Immunofluorescence analysis showing a drop in EAAT-1 immunoreactivity in the dorsal spinal cord of separated rats (*P ⬍ .05, Student’s t test, n ⫽ 5). Scale bar: 200 m. (C) Double immunofluorescence showing the colocalization (yellow, right images) of glial fibrillary acidic protein (GFAP) (red, left images) with EAAT-1 (green, middle images) in the spinal cord of nonseparated rats (upper panels) or maternally separated rats (lower panels). Scale bar: 20 m.
Figure 4. Maternal separation is not associated with alterations in the gross spinal astrocytic phenotype. (A) Microphotograph showing glial fibrillary acidic protein (GFAP) immunoreactivity in the dorsal spinal cord of nonseparated (left panel) and maternally separated rats (right panel). Scale bar: 200 m. (B) No change in spinal GFAP immunoreactivity was noticeable in pixel density, cell density per mm2, or in the mean length of astrocytic processes following maternal separation (n ⫽ 5 rats). (C) Western blot confirming that maternal separation had no effect on spinal GFAP expression. All Western blot results were expressed as a ratio over actin and normalized to nonseparated rats (n ⫽ 5 rats).
Discussion Taken together, the data presented herein add another dimension to the therapeutic use of riluzole, establishing its potential relevance to the treatment of visceral hypersensitivity. Our present results are in accordance with data showing that riluzole attenuates hyperalgesia in neuropathic pain models at doses devoid of side effects, an action chiefly connected to the activation in glutamate reuptake.11,15 The present demonstration that riluzole reverses visceral hypersensitivity is important because riluzole has only been approved to date by international drug regulatory authorities for the treatment of amyotrophic lateral sclerosis.14 Nevertheless, beyond this, analgesic effects of riluzole have been also described, in particular, it was shown to reduce glutamate concentration in cerebrospinal fluid upon systemic administration.15 A remarkable feature of its presently described action is the reduction in visceral sensitivity in separated rats only, with no noticeable effect in nonseparated animals. This suggests that riluzole per se does not impede basal visceromotor responses or the nociceptive response to noxious somatic stimulus, but rather interferes with hypersensitivity in separated animals. In support of this assumption, we further show that riluzole has no effect on acute nociception in naïve animals. In line with our present data, others have reported that riluzole did not affect acute nociception.23
RILUZOLE AND VISCERAL HYPERSENSITIVITY
2423
Furthermore, an absence of analgesic effect of preventive administration of riluzole on acute pain has been also reported in human naïve subjects.24 More intriguingly, 1 study has indicated no pain-alleviating effect of riluzole in 1 human trial on neuropathic subjects.25 However, neuropathic insults likely result in more intense spinal glial changes than early-life stress, as illustrated by the global change in glial morphology observed in neuropathic pain and in contrast to the more discreet alterations described in our present work. Hence, astrocytic pronociceptive changes that occurred after years of neuropathy in this specific study may have been too profound to be reversed by the sole increased activity in EAAT. This hypothesis is emphasized by the recent report from Maeda et al, indicating that viral gene transfer of EAAT-2 in the rat spinal cord was able to prevent but not reverse allodynia in the nerve ligation model.26 Interestingly, riluzole has also recently emerged as a candidate in human clinical trials to treat mood disorders, because off-label use of this drug has provided satisfactory results in patients.27,28 Notably, psychoform symptoms and mood disorders are common comorbidities among patients with visceral pain29 and alteration in astrocytic function in specific brain regions has been repeatedly put forward as a putative mechanism in these psychiatric diseases.30,31 Riluzole might therefore act on pathophysiological pathways shared between mood disorders and visceral hypersensitivity, such as glutamatergic neurotransmission, chiefly via an increased glial glutamate transport. Interestingly, we have recently shown that riluzole fails to alter anxiety behavior in a number of tasks (O’Connor and Cryan, unpublished), thus its pain-
Figure 5. Inhibition of spinal glutamate transport produces visceral hypersensitivity. (A) Intrathecal injection of glutamate transporter blocker DL-threo--benzyloxyaspartate (TBOA) results in a visceral hypersensitivity to colorectal distension (F(2,30) ⫽ 6.7; Dunnett’s test, *P ⬍ .05, **P ⬍ .01, vs vehicle treatment; n ⫽ 10 for vehicle and 1.2 ng TBOA, n ⫽ 13 for 12 ng TBOA). (B) TBOA blocks [3H]D-aspartate reuptake in rat primary spinal astrocytes (***P ⬍ .0001, Student’s t test, n ⫽ 4 wells).
BASIC– ALIMENTARY TRACT
June 2010
2424
GOSSELIN ET AL
BASIC– ALIMENTARY TRACT
alleviating ability is not consequential to any potential anxiolytic effects. One striking feature of the astrocytic response to maternal separation is the lack of gross phenotypic changes in the spinal cord. Notably, we did not detect any modification in GFAP immunoreactivity at either the level of synthesis or morphologically. This is a remarkable observation, as the vast majority of chronic pain models used to investigate glial involvement have shown a strong astrocytic reaction in association with increased nociception, especially a large GFAP up-regulation.32 Our present data suggest that discreet changes in astrocytes, such as a reduction in EAAT-1 expression, may take place in spinal astrocytes without any detectable increase in GFAP. Interestingly, using a model of water avoidance stress, it was recently shown that other glial cells, microglia, mediate stress-induced visceral hyperalgesia in the absence of change in the expression of classical microglial activation marker cd-11b.33 Therefore, it is possible that stressinduced visceral hypersensitivity is maintained by glia through molecular mechanisms that do not involve the global cell activation observed in neuropathic or inflammatory pain. Nevertheless, in our protocol, visceral responses are assessed up to 10 weeks after maternal separation, therefore, we cannot rule out the possibility that modifications in GFAP expression might occur soon after stress and resolve thereafter. The demonstration herein that visceral pain perception relies on spinal glutamate scavenging is in accordance with multiple lines of evidence reporting that a reduction in spinal glutamate uptake may produce pain.34 Intrathecal administration of EAAT blockers in naïve animals results in spontaneous somatic pain, as well as hyperalgesia and allodynia, implying that a continuous spinal glutamate uptake has a key basal antinociceptive action.9,35 Our data providing the first demonstration that intrathecal blocking of glutamate transport increases visceral response to mechanical noxious stimuli confirm the existence of a constant antinociceptive action of spinal EAAT. This control exerted by glutamate transporters has been further emphasized by recent studies reporting that EAAT overexpression reduces pain sensitivity.26,36 Interestingly, Lin and colleagues also reported that EAAT-overexpressing mice present a lower visceral sensitivity, pointing out the preventive influence of glutamate scavenging on visceral pain. Altogether, our present data extend and confirm the concept that spinal glutamate transport exerts a key role in the control of nociception.34 The importance of glial glutamate uptake is further highlighted by the down-regulation of spinal EAAT transporters in various models of chronic pain,10,11,37 similarly to the reduction in EAAT-1 described here in our maternally separated animals. Indeed, following chronic constriction injury, neuropathic rats show a long-lasting reduction in spinal EAAT-1 and EAAT-2 immunoreactivity in the spinal cord.11 Additionally, down-regulation of spinal EAAT-1 has also been described in hypersensitivity
GASTROENTEROLOGY Vol. 138, No. 7
due to chronic morphine tolerance10 and in chemotherapy (taxol)-induced hyperalgesia.37 However, to our knowledge, our data are the first to describe a change in EAAT in an animal model relevant to human stressinduced visceral hypersensitivity. Various hypotheses can be put forward to explain the reduction in EAAT-1 following maternal separation. First, a strong regulation exerted by inflammatory mediators on EAAT has been described consistently.38 This is of particular interest because it has been repeatedly claimed that glial involvement in chronic pain may be mediated by increases in immune factors.22 Interestingly, an augmentation in systemic immunity occurs following maternal separation in the rat,16 as well as in IBS patients.39 In addition, glutamate itself down-regulates EAAT-1.40 If the actual regulatory mechanisms accounting for the differential regulation of EAAT-1 over EAAT-2 in maternal separation are not elucidated, it is worth mentioning that opposite regulations have been described already.41 The unraveling of such mechanisms upstream to EAAT-1 regulation may provide further targets against visceral hypersensitivity. Moreover, the analysis of riluzole-induced alterations in glutamate levels using spinal microdialysis techniques in stressed animals, as has been shown in normal animals,12 will further confirm the functional role of reduced EAAT-1 in these animals. In conclusion, the data provided in the present work point to a possible key implication of spinal glutamate transport in visceral pain and suggest that riluzole may be a good candidate to alleviate visceral hypersensitivity. Given the clinical availability of riluzole, human trials aiming at testing the potency of riluzole in functional gastrointestinal disorders, such as IBS, are now warranted. References 1. Bradesi S, Mayer EA. Novel therapeutic approaches in IBS. Curr Opin Pharmacol 2007;7:598 – 604. 2. Clarke G, Quigley EM, Cryan JF, et al. Irritable bowel syndrome: towards biomarker identification. Trends Mol Med 2009;15:478–489. 3. Lowman BC, Drossman DA, Cramer EM, et al. Recollection of childhood events in adults with irritable bowel syndrome. J Clin Gastroenterol 1987;9:324 –330. 4. Hyland NP, Julio-Pieper M, O’Mahony SM, et al. A distinct subset of submucosal mast cells undergoes hyperplasia following neonatal maternal separation: a role in visceral hypersensitivity? Gut 2009;58:1029 –1030; author reply 1030 –1031. 5. O’Malley D, Dinan TG, Cryan JF. Alterations in colonic corticotropinreleasing factor receptors in the maternally separated rat model of irritable bowel syndrome: differential effects of acute psychological and physical stressors. Peptides 2010;31:662–670. 6. Coutinho SV, Plotsky PM, Sablad M, et al. Neonatal maternal separation alters stress-induced responses to viscerosomatic nociceptive stimuli in rat. Am J Physiol Gastrointest Liver Physiol 2002;282:G307– G316. 7. Price DD, Zhou Q, Moshiree B, et al. Peripheral and central contributions to hyperalgesia in irritable bowel syndrome. J Pain 2006;7:529 –535. 8. Danbolt NC. Glutamate uptake. Prog Neurobiol 2001;65:1–105. 9. Liaw WJ, Stephens RL Jr, Binns BC, et al. Spinal glutamate uptake is critical for maintaining normal sensory transmission in rat spinal cord. Pain 2005;115:60 –70.
10. Mao J, Sung B, Ji RR, et al. Chronic morphine induces downregulation of spinal glutamate transporters: implications in morphine tolerance and abnormal pain sensitivity. J Neurosci 2002;22:8312–8323. 11. Sung B, Lim G, Mao J. Altered expression and uptake activity of spinal glutamate transporters after nerve injury contribute to the pathogenesis of neuropathic pain in rats. J Neurosci 2003;23:2899–2910. 12. Tawfik VL, Regan MR, Haenggeli C, et al. Propentofylline-induced astrocyte modulation leads to alterations in glial glutamate promoter activation following spinal nerve transection. Neuroscience 2008;152:1086 –1092. 13. Milligan ED, Watkins LR. Pathological and protective roles of glia in chronic pain. Nat Rev Neurosci 2009;10:23–36. 14. Wokke J. Riluzole. Lancet 1996;348:795–799. 15. Coderre TJ, Kumar N, Lefebvre CD, et al. A comparison of the glutamate release inhibition and anti-allodynic effects of gabapentin, lamotrigine, and riluzole in a model of neuropathic pain. J Neurochem 2007;100:1289 –1299. 16. O’Mahony SM, Marchesi JR, Scully P, et al. Early life stress alters behavior, immunity, and microbiota in rats: implications for irritable bowel syndrome and psychiatric illnesses. Biol Psychiatry 2009;65: 263–267. 17. Gosselin RD, O’Mahony S, Gibney S, et al. Altered spinal glial activation in genetically anxious Wistar-Kyoto rats: implications for visceral pain in irritable bowel syndrome (IBS). Society for Neuroscience, Neurocience Meeting Planner, San Diego, CA, 2007. 18. Mantz J, Cheramy A, Thierry AM, et al. Anesthetic properties of riluzole (54274 RP), a new inhibitor of glutamate neurotransmission. Anesthesiology 1992;76:844 – 848. 19. Blackburn-Munro G, Ibsen N, Erichsen HK. A comparison of the anti-nociceptive effects of voltage-activated Na⫹ channel blockers in the formalin test. Eur J Pharmacol 2002;445:231–238. 20. Azbill RD, Mu X, Springer JE. Riluzole increases high-affinity glutamate uptake in rat spinal cord synaptosomes. Brain Res 2000;871:175– 180. 21. Fumagalli E, Funicello M, Rauen T, et al. Riluzole enhances the activity of glutamate transporters GLAST, GLT1 and EAAC1. Eur J Pharmacol 2008;578:171–176. 22. McMahon SB, Cafferty WB, Marchand F. Immune and glial cell factors as pain mediators and modulators. Exp Neurol 2005;192:444–462. 23. Schmidt AP, Tort AB, Silveira PP, et al. The NMDA antagonist MK-801 induces hyperalgesia and increases CSF excitatory amino acids in rats: reversal by guanosine. Pharmacol Biochem Behav 2009;91:549 –553. 24. Hammer NA, Lilleso J, Pedersen JL, et al. Effect of riluzole on acute pain and hyperalgesia in humans. Br J Anaesth 1999;82:718–722. 25. Galer BS, Twilling LL, Harle J, et al. Lack of efficacy of riluzole in the treatment of peripheral neuropathic pain conditions. Neurology 2000;55:971–975. 26. Maeda S, Kawamoto A, Yatani Y, et al. Gene transfer of GLT-1, a glial glutamate transporter, into the spinal cord by recombinant adenovirus attenuates inflammatory and neuropathic pain in rats. Mol Pain 2008;4:65. 27. Pittenger C, Coric V, Banasr M, et al. Riluzole in the treatment of mood and anxiety disorders. CNS Drugs 2008;22:761–786. 28. Zarate CA, Manji HK. Riluzole in psychiatry: a systematic review of the literature. Expert Opin Drug Metab Toxicol 2008;4:1223–1234. 29. North CS, Hong BA, Alpers DH. Relationship of functional gastrointestinal disorders and psychiatric disorders: implications for treatment. World J Gastroenterol 2007;13:2020 –2027. 30. Gosselin RD, Gibney S, O’Malley D, et al. Region specific decrease in glial fibrillary acidic protein immunoreactivity in the brain of a rat model of depression. Neuroscience 2009;159:915–925.
RILUZOLE AND VISCERAL HYPERSENSITIVITY
2425
31. Rajkowska G, Miguel-Hidalgo JJ. Gliogenesis and glial pathology in depression. CNS Neurol Disord Drug Targets 2007;6:219–233. 32. Honore P, Rogers SD, Schwei MJ, et al. Murine models of inflammatory, neuropathic and cancer pain each generates a unique set of neurochemical changes in the spinal cord and sensory neurons. Neuroscience 2000;98:585–598. 33. Bradesi S, Svensson CI, Steinauer J, et al. Role of spinal microglia in visceral hyperalgesia and NK1R up-regulation in a rat model of chronic stress. Gastroenterology 2009;136:1339 –1348, e1– e2. 34. Tao YX, Gu J, Stephens RL Jr. Role of spinal cord glutamate transporter during normal sensory transmission and pathological pain states. Mol Pain 2005;1:30. 35. Weng HR, Chen JH, Cata JP. Inhibition of glutamate uptake in the spinal cord induces hyperalgesia and increased responses of spinal dorsal horn neurons to peripheral afferent stimulation. Neuroscience 2006;138:1351–1360. 36. Lin Y, Tian G, Roman K, et al. Increased glial glutamate transporter EAAT2 expression reduces visceral nociceptive response in mice. Am J Physiol Gastrointest Liver Physiol 2009;296:G129–G134. 37. Weng HR, Aravindan N, Cata JP, et al. Spinal glial glutamate transporters downregulate in rats with taxol-induced hyperalgesia. Neurosci Lett 2005;386:18 –22. 38. Tilleux S, Hermans E. Neuroinflammation and regulation of glial glutamate uptake in neurological disorders. J Neurosci Res 2007;85:2059 –2070. 39. Dinan TG, Quigley EM, Ahmed SM, et al. Hypothalamic-pituitary-gut axis dysregulation in irritable bowel syndrome: plasma cytokines as a potential biomarker? Gastroenterology 2006;130:304–311. 40. Rosas S, Vargas MA, Lopez-Bayghen E, et al. Glutamate-dependent transcriptional regulation of GLAST/EAAT1: a role for YY1. J Neurochem 2007;101:1134 –1144. 41. Zerangue N, Arriza JL, Amara SG, et al. Differential modulation of human glutamate transporter subtypes by arachidonic acid. J Biol Chem 1995;270:6433– 6435.
Received August 20, 2009. Accepted March 4, 2010. Reprint requests Address requests for reprints to: John F. Cryan, PhD, Department of Pharmacology and Therapeutics, School of Pharmacy, University College Cork, Cavanagh Pharmacy Building, College Road, Cork, Ireland. e-mail: [email protected]; fax (353) 21490 1656. Dr Gosselin is currently at the Department of Anesthesiology, University Hospital Center and Department of Cell Biology and Morphology, University of Lausanne, Lausanne, Switzerland. Acknowledgments The authors are grateful to Pat Fitzgerald, Sinead Gibney, and Siobhain O’Mahony for the excellent technical assistance with behavioral experiments and to Nora Dwyer for help with optical quantification of EAAT-1 immunoreactivity. Conflicts of interest The authors disclose the following conflicts: The authors received industry support from GlaxoSmithKline. The Alimentary Pharmabiotic Centre is partly funded by GlaxoSmithKline. Funding JFC and TGD are supported in part by Science Foundation Ireland in the form of a Centre grant (Alimentary Pharmabiotic Centre).
BASIC– ALIMENTARY TRACT
June 2010
Riluzole Normalizes Early-Life Stress-Induced Visceral Hypersensitivity in Rats: Role of Spinal Glutamate Reuptake Mechanisms ROMAIN–DANIEL GOSSELIN,* RICHARD M. O’CONNOR,* MONICA TRAMULLAS,* MARCELA JULIO–PIEPER,* TIMOTHY G. DINAN,*,‡ and JOHN F. CRYAN*,§,储 *Laboratory of Neurogastroenterology, Alimentary Pharmabiotic Centre; ‡Departments of Psychiatry and §Pharmacology and Therapeutics, and 储School of Pharmacy, University College Cork, Cork, Ireland
BASIC– ALIMENTARY TRACT
BACKGROUND & AIMS: The molecular basis underlying visceral hypersensitivity in functional irritable bowel syndrome remains elusive, resulting in poor treatment effectiveness. Because alterations in spinal non-neuronal (astrocytic) glutamate reuptake are suspected to participate in chronic pain, we asked whether such processes occur in visceral hypersensitivity. METHODS: Visceral hypersensitivity was induced in Sprague–Dawley rats by maternal separation. Separated adults were given a systemic administration of riluzole (5 mg/kg), an approved neuroprotective agent activating glutamate reuptake. Visceral hypersensitivity was assessed using colorectal distension (40 mm Hg). Somatic nociception was quantified using Hot Plate, Randall–Sellito, and Hargreaves tests. Spinal proteins were quantified using immunofluorescence and Western blot. The dependence of visceral sensory function upon spinal glutamate transport was evaluated by intrathecal injection of glutamate transport antagonist DL-threo--benzyloxyaspartate (TBOA). For in vitro testing of riluzole and TBOA, primary cultures of astrocytes were used. RESULTS: We show that riluzole counteracts stress-induced visceral hypersensitivity without affecting visceral response in nonseparated rats or altering nociceptive responses to somatic pain stimulation. In addition, maternal separation produces a reduction in glial excitatory amino acid transporter (EAAT)-1 with no change in EAAT-2 or ␥-amino butyric acid transporters. Stress was not associated with changes in glial fibrillary acidic protein or astrocytic morphology per se. Furthermore, visceral normosensitivity relies on spinal EAAT, as intrathecal TBOA is sufficient to induce hypersensitivity in normal rats. CONCLUSIONS: We identify spinal EAAT as a therapeutic target, and establish riluzole as a candidate to counteract gastrointestinal hypersensitivity in disorders such as irritable bowel syndrome. Keywords: Irritable Bowel Syndrome; Maternal Separation; Astrocytes; Spinal Cord.
V
isceral hypersensitivity is a core symptom of functional gastrointestinal disorders such as irritable bowel syndrome (IBS) and is a pressing clinical issue that remains an ongoing challenge for the pharmaceutical industry.1,2 The ineffectiveness of current medications against visceral hypersensitivity is attributable at least in
part to the lack of a unifying hypothesis regarding the chain of events generating pain. Despite this absence of consensus, stress occurring in early life is known to have a key detrimental upstream influence on nociceptive pathways and in the manifestation of IBS symptoms.3–5 Hence, the enhanced visceromotor response following pup isolation from dams (maternal separation) is viewed as an IBS-relevant model in animals.6 In addition, central sensitization (ie, an increased activation of spinal nociceptive neurons to peripheral stimuli) may be an end mechanism accounting for visceral hyperalgesia in IBS,7 presumably due to a sustained glutamatergic sensory neurotransmission. However, glutamate receptor antagonists show a constellation of side effects strongly restricting their utilization in patients. Interestingly, beyond glutamate receptor targeting, the efficacy of glutamatergic neurotransmission may be modulated by acting on glutamate reuptake systems. Indeed, the proper and regulated action of glutamate relies on its fast elimination from extracellular synaptic milieu as a result of the scavenging machinery, namely the excitatory amino-acid transporters (EAAT)-1 and EAAT-2.8 These transporters are expressed by non-neuronal glial cells, primarily astrocytes, whose endfeet cover virtually all synapses in the central nervous system. Remarkably, the importance of spinal EAAT in nociception has been emphasized by recent evidence showing that selective intrathecal inhibition of glial glutamate transport in naïve rats results in pain behaviors.9 Moreover, a reduction in EAAT expression occurs in chronic pain states10 –12 as a component of the profound phenotypic alterations taking place in spinal astrocytes, which encompasses an increased synthesis of glial fibrillary acidic protein (GFAP) and a release in sensitizing soluble factors.13 These alterations are consistently observed in a variety of pain models, giving astrocytes an emerging key role in the maintenance of chronic pain. However, there is a paucity of data regarding the Abbreviations used in this paper: EAAT, excitatory amino-acid transporter; GAT, ␥-amino butyric acid transporter; GFAP, glial fibrillary acidic protein; IBS, irritable bowel syndrome; PBS, phosphate-buffered saline; TBOA, DL-threo--benzyloxyaspartate. © 2010 by the AGA Institute 0016-5085/$36.00 doi:10.1053/j.gastro.2010.03.003
involvement of astrocytes in visceral pain, especially with regard to the possible role of glutamate uptake system. Riluzole (2-amino-6-trifluoromethoxybenzothiazole) is a neuroprotective drug already approved for amyotrophic lateral sclerosis.14 Riluzole is a potent activator of glutamate reuptake both in vitro and in vivo. In addition, pain-modulatory properties of riluzole have been reported,11,15 but its influence on visceral pain remains unknown. Therefore, in the present study, we assessed whether riluzole could reduce visceral hypersensitivity induced by maternal separation and how this could shed some insight on underlying mechanisms in stress-induced visceral pain. We found that riluzole given intraperitoneally normalizes visceral hypersensitivity in maternally separated rats. Moreover, it failed to affect visceral responses in nonseparated animals and had no effects on somatic nociception; maternal separation produces a reduction in EAAT-1 in the spinal cord; stress did not result in changes in EAAT-2, ␥-amino butyric acid transporter (GAT), and ␥-amino butyric acid expression or in astrocytic morphological alterations; direct intrathecal inhibition of EAAT using DL-threo--benzyloxyaspartate (TBOA) is sufficient to produce a visceral hypersensitivity in naïve rats. Taken together, the data presented here add another dimension to the use of riluzole in therapeutics, establishing its relevance to counteract visceral hypersensitivity, and point to spinal glial glutamate reuptake as a pharmacological target for pain in functional bowel disorders.
Materials and Methods Animals Male Sprague–Dawley rats weighing 280 –300 g were housed and bred in the local animal facility with food and water ad libidum, on a 12:12-hour reversed dark–light cycle with temperature at 20°C ⫾ 1°C. Animals were group-housed by 4 to 5 per cage in usual plastic cages with sawdust bedding in an enriched environment with shredded paper and a cardboard roll. All experiments were in full accordance with the European Community Council Directive (86/609/EEC). Behavioral assessments were always carried out randomly by an experimenter blind to treatment groups.
Drug Administration Riluzole and TBOA were purchased from Tocris (Bristol, UK). For intrathecal injection, rats were sedated by light isofurane anesthesia and vehicle (phosphatebuffered saline [PBS], 10 L) or TBOA (1.2 g or 12 g in 10 L vehicle) was administrated. Correct positioning was assessed by a tail flick upon needle insertion. Balloons were immediately inserted in preparation for colorectal distension. The absence of spinal damage was assessed prior to distension, and animals presenting any sign of paralysis or spontaneous pain behaviors were immediately culled. Colorectal distensions were carried out 8 –10 minutes after intrathecal injections. This tim-
RILUZOLE AND VISCERAL HYPERSENSITIVITY
2419
ing was selected from previous studies that showed that it was at the longer range used to see maximal effects of TBOA delivered intrathecally.9 For riluzole studies, rats were weighed and a slight volume correction was applied to the stock solution (1.5 mg/mL, in PBS) for the intraperitoneal injection to achieve a final administration of 5 mg/kg riluzole. Control rats were injected with vehicle (1 mL PBS). Colorectal distension was carried out 30 minutes later. For nociceptive testing (Hot Plate, Randall– Sellito, and Hargreaves tests), riluzole was injected 30 minutes prior to testing.
Colorectal Distension Animals were sedated using isofurane, and 7-cm–long deflated balloons attached to polyethylene tubing were inserted intrarectally through the anus. Rats were left to recover for 10 minutes and the tubing was connected to a barostat machine. The protocol consisted in a tonic distension at a constant pressure of 40 mm Hg during 10 minutes. Visceromotor response to distension was quantified as the number of abdominal contractions resulting in animal arching during the procedure. At the end of the distension, balloons were deflated and the animals were culled.
Hot-Plate Test Rats were placed with all 4 paws on the hot plate (Incremental Hot Plate; Stoelting, Dublin, Ireland) heated at 55°C and the latency to hindpaw lick or shake was measured. The cutoff time was set at 20 seconds to avoid tissue damage. Baseline latencies were recorded 30 minutes prior to the experiment.
Paw Pressure Test, Randall–Sellito Responses to noxious mechanical stimuli were measured with a Digital Randall–Sellito test (Stoelting). Rats were handled and partly restrained. Following forepaw positioning, an increasing pressure was applied to the dorsal surface. Nociceptive threshold was defined as the pressure in grams eliciting the paw withdrawal. Measurements were performed in triplicate. A cutoff weight was set at 400 g. Baseline responses were recorded 1 day before drug administration.
Paw Withdrawal Test, Hargreaves Paw withdrawal response to noxious radiant heat stimuli was assessed using Plantar Test (Plantar Test Analgesia Meter; Stoelting). Rats were placed in individual Plexiglas chambers on a glass platform heated at 37°C and habituated for 5 minutes. A mobile radiant heat source was located under the platform and focused onto the mid-plantar surface of each hindpaw sequentially. The paw withdrawal latency was defined as the time from radiant heat onset to paw withdrawal. The apparatus was calibrated to give a paw withdrawal latency of 7–9 seconds prior to drug injection. A total of 6 paw withdrawal latencies (in triplicate, each separated by at
BASIC– ALIMENTARY TRACT
June 2010
2420
GOSSELIN ET AL
least 5 minutes) were measured. A cutoff time of 20 seconds was used to avoid tissue damage. Baseline responses were taken the day before drug administration.
Rotarod Testing of Motor Coordination
BASIC– ALIMENTARY TRACT
All animals were trained prior to accessing performance on the rotarod (HARotaRod System). Rotating speed was fixed at 12 rpm. Training comprised of 2 days: on day 1, animals were placed on the rotarod (perpendicular to the axis of rotation, facing away from the direction of rotation so that the animal had to walk forward to stay on the bar) until they completed 1 trial of 180 seconds; on day 2, animals completed 3 sessions of 180 seconds within 1 hour. After the final trial on day 2, animals were placed on the rotarod multiple times for habituation, until the rats stepped voluntarily onto the rotating bar. On the test day (day 3), 3 trials were conducted, all trials having a cutoff point of 180 seconds. Trial 1 was conducted 3 minutes prior to treatment; any animals that did not successfully complete trial 1 were omitted from the study. Trial 2 was conducted 30 minutes after treatment and trial 3 was conducted 180 minutes after treatment. For riluzole treatment, final doses of 2.5 mg/kg, 5.0 mg/kg, and 10 mg/kg were used. Control animals were injected with 1 mL vehicle alone (PBS). All trials were scored manually by the researcher.
Maternal Separation Maternal separation was performed as described previously.16 Briefly, rat pups were separated from their dams as a whole litter during 180 minutes every day over a period of 10 days between postnatal days 2 and 12. Separations were conducted between 9 AM and 12 AM in plastic cages placed on top of heater pads (30°C–33°C) and placed in a separated room. Control rats consisted of nonhandled pups, left untouched with their mothers. After postnatal day 12, rats were left undisturbed except for routine cage cleaning every two days until they were 8 –10 weeks old.
Immunolabeling Rabbit primary antibodies against EAAT-1 and EAAT-2 were from Abcam (Cambridge, UK); rabbit antibodies against GFAP, GAT-1, and GAT-3 were from Chemicon-Millipore (Tullagreen, Ireland); mouse antiGFAP was from Cell Signaling (Danvers, MA); and mouse anti-actin was purchased from Sigma (Dublin, Ireland). Secondary antibodies were from Jackson Immunoresearch (Newmarket, Suffolk, UK). Immunofluorescence was carried out as described previously17 with minor modifications. Briefly, rats were fixed by transcardial perfusion of 4% paraformaldehyde in PBS under lethal anesthesia. Postfixed and sucrose-cryoprotected samples were frozen in dry ice and cut in 18-m slices using a cryostat (Leica, Ashbourne, Ireland). Mounted slices were blocked using 10% normal goat serum in PBS 0.05% Triton x-100 and incubated overnight at 4°C with the
GASTROENTEROLOGY Vol. 138, No. 7
primary antibody in the same blocking solution (rabbit anti-GFAP: 1/2000; rabbit anti–EAAT-1: 1/1000; mouse anti-GFAP 1/500). After extensive washing, the slides were incubated with fluorescent dye-coupled secondary antibody, washed, and mounted in fluorescence medium (DAKO, Hazel Grove, UK). Images were taken using the Cell-F software (Olympus, Dublin, Ireland) and analyzed using ImageJ software (National Institutes of Health, http://rsb.info.nih.gov/ij/).
Western Blot The Western blot procedure was conducted as described previously.17 Briefly, spinal cords were frozen and stored at ⫺80°C until protein extraction and quantification. Twenty micrograms of proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred onto a polyvinylidene fluoride membrane, and blocked (PBS with 0.1% Tween 20 and 5% Milk). For immunodetection, overnight incubation at 4°C with primary antibodies (1/2000) was performed in blocking solution. After washing, membranes were incubated with horseradish peroxidase– conjugated secondary antibodies (1/10,000) and in electrochemiluminescence reagents (Pierce, IL). Images were obtained using a luminescent image analyzer (LAS-3000; Fugifilm, Ireland). For the subsequent detection of -actin, the membranes were processed in the same manner as in mouse anti-actin antibody (1/5000; Sigma, Dublin, Ireland). Immunoblots were quantified using ImageJ software; 5 rats were used in each group.
Primary Culture of Astrocytes and Aspartate Reuptake Spinal cords were collected in Dulbecco’s modified Eagle’s medium (Sigma, Dublin, Ireland) under aseptic conditions from decapitated newborn rats at postnatal days 3 to 4. Tissues were mechanically dissociated in Dulbecco’s modified Eagle’s medium supplemented with 5% fetal bovine, L-glutamine (2 mM; Sigma), and gentamycin (Sigma). The cell suspension was plated onto polyD-lysine– coated flasks (T-75) using 1 spinal cord per flask and kept in a 95% CO2/5% H2O incubator at 37°C. Medium was changed by half every 3 days, and after 7 or 10 days in vitro, flasks were shaken 3 hours at 200 rpm to detach microglial cells. Adherent astrocytes were trypsinized and seeded at 5.104 cells/mL on poly-Dlysine– coated 6-well microplates. Medium was changed by half every 2 days with complete medium without serum to improve cell differentiation, and astrocytes were used 6 days after their passage. The enrichment in astrocytes was confirmed by a GFAP immunolabeling ⬎95%. Reuptake experiments were conducted as follows. Medium was fully changed, and astrocytes were first equilibrated 10 minutes at 37°C with prewarmed Hank’s Balanced Salt Solution (Sigma, Dublin, Ireland) completed with MgCl2 (1.2 mM) and HEPES (10 mM) and supplemented or not with TBOA or riluzole. Radiolabeled aspartate (50 nM final, [3H]D-aspartate, 12.2 Ci/mmol;
June 2010
RILUZOLE AND VISCERAL HYPERSENSITIVITY
2421
Amersham, Buckinghamshire, UK) was added simultaneously with nonradioactive aspartate (5 M final) from prewarmed 10⫻ solutions in Hank’s Balanced Salt Solution. After 7 minutes, the transport was stopped by 2 rinses using ice-cold Hank’s Balanced Salt Solution and cells were lysed with 1 mL 0.5 M NaOH ⫹ 0.05% sodium dodecyl sulfate in water. Protein content was evaluated using Qubit technology (Invitrogen, Carlsbad, CA), and reuptake activity was expressed in cpm/mg of proteins.
Statistical Analysis
Results Riluzole Counteracts Stress-Induced Visceral Hypersensitivity We explored the possibility that increasing glutamate transport could reverse stress-induced visceral hypersensitivity following maternal separation. Vehicle-treated separated rats show a higher response to colorectal distension than vehicle-treated nonseparated animals, in line with previously described works16 (F(3,28) ⫽ 4.058; P ⬍ .05, Figure 1A). Remarkably, single intraperitoneal administration of riluzole (5 mg/kg) counteracts maternal separation-associated hypersensitivity, restoring a number of abdominal contractions comparable to vehicle-treated nonseparated animals. Notably, riluzole had no detectable effect in nonseparated rats (F(3,28) ⫽ 4.058; P ⬎ .05), suggesting that riluzole reverses specific pronociceptive alterations in maternal separation without modulating the entire nociceptive pathway. In order to specifically address this point, we performed testing for mechanical and thermal nociception. As shown in Figure 1B, riluzole has no effect on thermal or mechanical nociception in naïve rats in Hot-Plate (F(5,49) ⫽ 0.166), Randall–Sellito (F(5,42) ⫽ 0.126), or Hargreaves (F(5,42) ⫽ 0.572) tests. Some authors have reported impairment in motor coordination (sedative or ataxic effect) following riluzole,18 whereas others have observed no effect.11,15,19 Therefore, in order to rule out the possibility that the riluzole effect we observe could be due to such a process, we evaluated the consequences of riluzole on motor coordination in naïve rats using the rotarod test. As shown in Figure 1C, riluzole treatment has no influence on motor performances at either 30 minutes or 180 minutes postinjection.
Figure 1. Riluzole reverses stress-induced visceral hypersensitivity. (A) Maternally separated (MS) rats and nonseparated (NS) controls were treated with vehicle or riluzole (5 mg/kg, intraperitoneally), and visceral sensitivity to colorectal distension (10 minutes at 40 mm Hg) was determined as the number of abdominal contractions. An increase in abdominal contractions in response to distension is observed in separated animals. Riluzole counteracts the visceral hypersensitivity observed in separated rats (F(3,28) ⫽ 4.058; Newman–Keuls test, *P ⬍ .05) with no detectable effect in nonseparated animals (n ⫽ 9 for vehicle NS and MS, n ⫽ 8 and 6 for riluzole NS and riluzole MS, respectively). (B) Riluzole has no effect on acute nociception in the Hot Plate, Randall– Sellito, or Hargreaves tests (n ⫽ 8 per group). (C) Rotarod testing following administration of various doses of riluzole shows no impairment of motor coordination after treatment with 2.5, 5, or 10 mg/kg riluzole.
Functional analysis in primary cultures of rat astrocytes confirms that riluzole increases the activity of EAAT (Figure 2). Reuptake of glutamate analogue [3H]D-aspartate was dose-dependently increased by riluzole (up to a 37% augmentation for a riluzole concentration of 1 M), an effect fully blocked by preincubation with TBOA (data not shown). At higher doses of riluzole, this effect was not observed as described previously.20,21 The expressions of glutamate transporters EAAT-1 and EAAT-2 in our in vitro system were confirmed by Western blot (Figure 2C). Together, these results show that riluzole is effective at reversing stress-induced visceral hypersensitivity and, given the activating effect of riluzole on glutamate transport, this suggests that glutamate reuptake could be altered as a result of maternal separation.
BASIC– ALIMENTARY TRACT
Data from pixel quantifications (immunofluorescence and Western blot) were compared using unpaired 2-tailed Student’s t test. Behavioral as well as reuptake results were analyzed using 1-way analyses of variance. Analyses of variance were followed by Newman–Keuls or Dunnett’s multiple comparison post-hoc tests when appropriate. All analyses were performed using GraphPadPrism (GraphPad Software, San Diego, CA). Differences were considered significant for P ⬍ .05. Data are expressed as mean ⫾ standard error of mean. No outlier animal or sample was discarded from any group.
2422
GOSSELIN ET AL
GASTROENTEROLOGY Vol. 138, No. 7
Basal Spinal Glutamate Reuptake Is Required to Prevent Visceral Hypersensitivity
Figure 2. Effect of riluzole on [3H]D-aspartate transport in vitro. (A) Riluzole dose-dependently increases [3H]D-aspartate reuptake in cultured astrocytes with a significant effect for 1 M riluzole (F(8,27) ⫽ 29.8; Dunnett’s test, *P ⬍ .05 vs no riluzole treatment; n ⫽ 4 wells). (B) Western blots showing that primary cultures of spinal astrocytes expressing the astrocytic marker glial fibrillary acidic protein (GFAP) also express the glutamate transporters excitatory amino-acid transporter (EAAT)-1 and EAAT-2.
In order to determine the consequence of spinal glutamate transport reduction on the maintenance of basal visceral sensitivity, we intrathecally administrated the glutamate transport blocker TBOA and monitored visceromotor response in naïve rats (Figure 5A). Intrathecal TBOA produces a dose-dependent increase in visceral sensitivity to distention with a 2-fold augmentation in contractions for an injection of 12 ng TBOA (F(2,30) ⫽ 6.737; P ⬍ .01 vehicle vs TBOA 12 ng). No other behavioral effect of TBOA was detected. The ability of TBOA to inhibit EAAT was further confirmed in vitro by functional reuptake analysis of the nonbiologically active, but transportable glutamate analogue [3H]D-aspartate in primary cultures of rat astrocytes (Figure 4B). As described previously, TBOA produces a robust reduction in the uptake of [3H]D-aspartate (P ⬍ .0001).
Maternal Separation Results in a Selective Reduction in Spinal EAAT-1 Expression
BASIC– ALIMENTARY TRACT
We investigated whether maternal separation affects spinal astrocytic glutamate transporter integrity. A significant reduction in EAAT-1 immunoreactivity is detected by both Western blot (P ⬍ .0028, Figure 3A) and immunofluorescence (P ⬍ .0461, Figure 3B) in the spinal cord of separated animals. The reduction in EAAT-1 immunoreactivity is observed in all dorsal spinal layers (target of the sensory input), even though EAAT-1 immunoreactivity retains its classical pattern with a higher expression in the superficial layers (Figure 3B). Remarkably, no change was detected in the other astroglial glutamate transporter, EAAT-2 or GAT-1 and -3 (Figure 3A). Finally, dual-immunofluorescence shows that GFAP colocalizes with EAAT-1 (Figure 3C).
Maternal Separation Does Not Induce a Global Change in Spinal Astrocytic Phenotype Chronic pain is associated with alterations in spinal astrocytic phenotype, the core change of which is the up-regulation in GFAP.22 Yet, spinal GFAP expression has not been studied in visceral pain so far. Strikingly, we found that early-life stress did not influence spinal GFAP expression (Figure 4). Indeed, tissues from separated and nonseparated animals exhibit comparable intensities in GFAP immunoreactivity and similar densities of GFAPexpressing cells (P ⫽ .5581 and P ⫽ .9628, respectively; Figure 4A and B). This absence of variation in GFAP expression was further confirmed by Western blot analysis (P ⫽ .2246; Figure 4C). Because perturbations in astrocytic physiology may be linked to a modification in the apparent cell morphology, we also quantified the mean length of GFAP containing processes and show a lack of effect of early-life stress as well (P ⫽ .5843; Figure 4C).
Figure 3. Maternal separation is associated with a reduction in excitatory amino-acid transporter (EAAT)-1 in spinal astrocytes. (A) Western-blot analysis of spinal EAAT and ␥-amino butyric acid transporter (GAT) in separated rats. A significant decrease in spinal EAAT-1 is detected in comparison to nonseparated rats (**P ⬍ .01, Student’s t test, n ⫽ 5), with no change in EAAT-2 or GAT transporters. (B) Immunofluorescence analysis showing a drop in EAAT-1 immunoreactivity in the dorsal spinal cord of separated rats (*P ⬍ .05, Student’s t test, n ⫽ 5). Scale bar: 200 m. (C) Double immunofluorescence showing the colocalization (yellow, right images) of glial fibrillary acidic protein (GFAP) (red, left images) with EAAT-1 (green, middle images) in the spinal cord of nonseparated rats (upper panels) or maternally separated rats (lower panels). Scale bar: 20 m.
Figure 4. Maternal separation is not associated with alterations in the gross spinal astrocytic phenotype. (A) Microphotograph showing glial fibrillary acidic protein (GFAP) immunoreactivity in the dorsal spinal cord of nonseparated (left panel) and maternally separated rats (right panel). Scale bar: 200 m. (B) No change in spinal GFAP immunoreactivity was noticeable in pixel density, cell density per mm2, or in the mean length of astrocytic processes following maternal separation (n ⫽ 5 rats). (C) Western blot confirming that maternal separation had no effect on spinal GFAP expression. All Western blot results were expressed as a ratio over actin and normalized to nonseparated rats (n ⫽ 5 rats).
Discussion Taken together, the data presented herein add another dimension to the therapeutic use of riluzole, establishing its potential relevance to the treatment of visceral hypersensitivity. Our present results are in accordance with data showing that riluzole attenuates hyperalgesia in neuropathic pain models at doses devoid of side effects, an action chiefly connected to the activation in glutamate reuptake.11,15 The present demonstration that riluzole reverses visceral hypersensitivity is important because riluzole has only been approved to date by international drug regulatory authorities for the treatment of amyotrophic lateral sclerosis.14 Nevertheless, beyond this, analgesic effects of riluzole have been also described, in particular, it was shown to reduce glutamate concentration in cerebrospinal fluid upon systemic administration.15 A remarkable feature of its presently described action is the reduction in visceral sensitivity in separated rats only, with no noticeable effect in nonseparated animals. This suggests that riluzole per se does not impede basal visceromotor responses or the nociceptive response to noxious somatic stimulus, but rather interferes with hypersensitivity in separated animals. In support of this assumption, we further show that riluzole has no effect on acute nociception in naïve animals. In line with our present data, others have reported that riluzole did not affect acute nociception.23
RILUZOLE AND VISCERAL HYPERSENSITIVITY
2423
Furthermore, an absence of analgesic effect of preventive administration of riluzole on acute pain has been also reported in human naïve subjects.24 More intriguingly, 1 study has indicated no pain-alleviating effect of riluzole in 1 human trial on neuropathic subjects.25 However, neuropathic insults likely result in more intense spinal glial changes than early-life stress, as illustrated by the global change in glial morphology observed in neuropathic pain and in contrast to the more discreet alterations described in our present work. Hence, astrocytic pronociceptive changes that occurred after years of neuropathy in this specific study may have been too profound to be reversed by the sole increased activity in EAAT. This hypothesis is emphasized by the recent report from Maeda et al, indicating that viral gene transfer of EAAT-2 in the rat spinal cord was able to prevent but not reverse allodynia in the nerve ligation model.26 Interestingly, riluzole has also recently emerged as a candidate in human clinical trials to treat mood disorders, because off-label use of this drug has provided satisfactory results in patients.27,28 Notably, psychoform symptoms and mood disorders are common comorbidities among patients with visceral pain29 and alteration in astrocytic function in specific brain regions has been repeatedly put forward as a putative mechanism in these psychiatric diseases.30,31 Riluzole might therefore act on pathophysiological pathways shared between mood disorders and visceral hypersensitivity, such as glutamatergic neurotransmission, chiefly via an increased glial glutamate transport. Interestingly, we have recently shown that riluzole fails to alter anxiety behavior in a number of tasks (O’Connor and Cryan, unpublished), thus its pain-
Figure 5. Inhibition of spinal glutamate transport produces visceral hypersensitivity. (A) Intrathecal injection of glutamate transporter blocker DL-threo--benzyloxyaspartate (TBOA) results in a visceral hypersensitivity to colorectal distension (F(2,30) ⫽ 6.7; Dunnett’s test, *P ⬍ .05, **P ⬍ .01, vs vehicle treatment; n ⫽ 10 for vehicle and 1.2 ng TBOA, n ⫽ 13 for 12 ng TBOA). (B) TBOA blocks [3H]D-aspartate reuptake in rat primary spinal astrocytes (***P ⬍ .0001, Student’s t test, n ⫽ 4 wells).
BASIC– ALIMENTARY TRACT
June 2010
2424
GOSSELIN ET AL
BASIC– ALIMENTARY TRACT
alleviating ability is not consequential to any potential anxiolytic effects. One striking feature of the astrocytic response to maternal separation is the lack of gross phenotypic changes in the spinal cord. Notably, we did not detect any modification in GFAP immunoreactivity at either the level of synthesis or morphologically. This is a remarkable observation, as the vast majority of chronic pain models used to investigate glial involvement have shown a strong astrocytic reaction in association with increased nociception, especially a large GFAP up-regulation.32 Our present data suggest that discreet changes in astrocytes, such as a reduction in EAAT-1 expression, may take place in spinal astrocytes without any detectable increase in GFAP. Interestingly, using a model of water avoidance stress, it was recently shown that other glial cells, microglia, mediate stress-induced visceral hyperalgesia in the absence of change in the expression of classical microglial activation marker cd-11b.33 Therefore, it is possible that stressinduced visceral hypersensitivity is maintained by glia through molecular mechanisms that do not involve the global cell activation observed in neuropathic or inflammatory pain. Nevertheless, in our protocol, visceral responses are assessed up to 10 weeks after maternal separation, therefore, we cannot rule out the possibility that modifications in GFAP expression might occur soon after stress and resolve thereafter. The demonstration herein that visceral pain perception relies on spinal glutamate scavenging is in accordance with multiple lines of evidence reporting that a reduction in spinal glutamate uptake may produce pain.34 Intrathecal administration of EAAT blockers in naïve animals results in spontaneous somatic pain, as well as hyperalgesia and allodynia, implying that a continuous spinal glutamate uptake has a key basal antinociceptive action.9,35 Our data providing the first demonstration that intrathecal blocking of glutamate transport increases visceral response to mechanical noxious stimuli confirm the existence of a constant antinociceptive action of spinal EAAT. This control exerted by glutamate transporters has been further emphasized by recent studies reporting that EAAT overexpression reduces pain sensitivity.26,36 Interestingly, Lin and colleagues also reported that EAAT-overexpressing mice present a lower visceral sensitivity, pointing out the preventive influence of glutamate scavenging on visceral pain. Altogether, our present data extend and confirm the concept that spinal glutamate transport exerts a key role in the control of nociception.34 The importance of glial glutamate uptake is further highlighted by the down-regulation of spinal EAAT transporters in various models of chronic pain,10,11,37 similarly to the reduction in EAAT-1 described here in our maternally separated animals. Indeed, following chronic constriction injury, neuropathic rats show a long-lasting reduction in spinal EAAT-1 and EAAT-2 immunoreactivity in the spinal cord.11 Additionally, down-regulation of spinal EAAT-1 has also been described in hypersensitivity
GASTROENTEROLOGY Vol. 138, No. 7
due to chronic morphine tolerance10 and in chemotherapy (taxol)-induced hyperalgesia.37 However, to our knowledge, our data are the first to describe a change in EAAT in an animal model relevant to human stressinduced visceral hypersensitivity. Various hypotheses can be put forward to explain the reduction in EAAT-1 following maternal separation. First, a strong regulation exerted by inflammatory mediators on EAAT has been described consistently.38 This is of particular interest because it has been repeatedly claimed that glial involvement in chronic pain may be mediated by increases in immune factors.22 Interestingly, an augmentation in systemic immunity occurs following maternal separation in the rat,16 as well as in IBS patients.39 In addition, glutamate itself down-regulates EAAT-1.40 If the actual regulatory mechanisms accounting for the differential regulation of EAAT-1 over EAAT-2 in maternal separation are not elucidated, it is worth mentioning that opposite regulations have been described already.41 The unraveling of such mechanisms upstream to EAAT-1 regulation may provide further targets against visceral hypersensitivity. Moreover, the analysis of riluzole-induced alterations in glutamate levels using spinal microdialysis techniques in stressed animals, as has been shown in normal animals,12 will further confirm the functional role of reduced EAAT-1 in these animals. In conclusion, the data provided in the present work point to a possible key implication of spinal glutamate transport in visceral pain and suggest that riluzole may be a good candidate to alleviate visceral hypersensitivity. Given the clinical availability of riluzole, human trials aiming at testing the potency of riluzole in functional gastrointestinal disorders, such as IBS, are now warranted. References 1. Bradesi S, Mayer EA. Novel therapeutic approaches in IBS. Curr Opin Pharmacol 2007;7:598 – 604. 2. Clarke G, Quigley EM, Cryan JF, et al. Irritable bowel syndrome: towards biomarker identification. Trends Mol Med 2009;15:478–489. 3. Lowman BC, Drossman DA, Cramer EM, et al. Recollection of childhood events in adults with irritable bowel syndrome. J Clin Gastroenterol 1987;9:324 –330. 4. Hyland NP, Julio-Pieper M, O’Mahony SM, et al. A distinct subset of submucosal mast cells undergoes hyperplasia following neonatal maternal separation: a role in visceral hypersensitivity? Gut 2009;58:1029 –1030; author reply 1030 –1031. 5. O’Malley D, Dinan TG, Cryan JF. Alterations in colonic corticotropinreleasing factor receptors in the maternally separated rat model of irritable bowel syndrome: differential effects of acute psychological and physical stressors. Peptides 2010;31:662–670. 6. Coutinho SV, Plotsky PM, Sablad M, et al. Neonatal maternal separation alters stress-induced responses to viscerosomatic nociceptive stimuli in rat. Am J Physiol Gastrointest Liver Physiol 2002;282:G307– G316. 7. Price DD, Zhou Q, Moshiree B, et al. Peripheral and central contributions to hyperalgesia in irritable bowel syndrome. J Pain 2006;7:529 –535. 8. Danbolt NC. Glutamate uptake. Prog Neurobiol 2001;65:1–105. 9. Liaw WJ, Stephens RL Jr, Binns BC, et al. Spinal glutamate uptake is critical for maintaining normal sensory transmission in rat spinal cord. Pain 2005;115:60 –70.
10. Mao J, Sung B, Ji RR, et al. Chronic morphine induces downregulation of spinal glutamate transporters: implications in morphine tolerance and abnormal pain sensitivity. J Neurosci 2002;22:8312–8323. 11. Sung B, Lim G, Mao J. Altered expression and uptake activity of spinal glutamate transporters after nerve injury contribute to the pathogenesis of neuropathic pain in rats. J Neurosci 2003;23:2899–2910. 12. Tawfik VL, Regan MR, Haenggeli C, et al. Propentofylline-induced astrocyte modulation leads to alterations in glial glutamate promoter activation following spinal nerve transection. Neuroscience 2008;152:1086 –1092. 13. Milligan ED, Watkins LR. Pathological and protective roles of glia in chronic pain. Nat Rev Neurosci 2009;10:23–36. 14. Wokke J. Riluzole. Lancet 1996;348:795–799. 15. Coderre TJ, Kumar N, Lefebvre CD, et al. A comparison of the glutamate release inhibition and anti-allodynic effects of gabapentin, lamotrigine, and riluzole in a model of neuropathic pain. J Neurochem 2007;100:1289 –1299. 16. O’Mahony SM, Marchesi JR, Scully P, et al. Early life stress alters behavior, immunity, and microbiota in rats: implications for irritable bowel syndrome and psychiatric illnesses. Biol Psychiatry 2009;65: 263–267. 17. Gosselin RD, O’Mahony S, Gibney S, et al. Altered spinal glial activation in genetically anxious Wistar-Kyoto rats: implications for visceral pain in irritable bowel syndrome (IBS). Society for Neuroscience, Neurocience Meeting Planner, San Diego, CA, 2007. 18. Mantz J, Cheramy A, Thierry AM, et al. Anesthetic properties of riluzole (54274 RP), a new inhibitor of glutamate neurotransmission. Anesthesiology 1992;76:844 – 848. 19. Blackburn-Munro G, Ibsen N, Erichsen HK. A comparison of the anti-nociceptive effects of voltage-activated Na⫹ channel blockers in the formalin test. Eur J Pharmacol 2002;445:231–238. 20. Azbill RD, Mu X, Springer JE. Riluzole increases high-affinity glutamate uptake in rat spinal cord synaptosomes. Brain Res 2000;871:175– 180. 21. Fumagalli E, Funicello M, Rauen T, et al. Riluzole enhances the activity of glutamate transporters GLAST, GLT1 and EAAC1. Eur J Pharmacol 2008;578:171–176. 22. McMahon SB, Cafferty WB, Marchand F. Immune and glial cell factors as pain mediators and modulators. Exp Neurol 2005;192:444–462. 23. Schmidt AP, Tort AB, Silveira PP, et al. The NMDA antagonist MK-801 induces hyperalgesia and increases CSF excitatory amino acids in rats: reversal by guanosine. Pharmacol Biochem Behav 2009;91:549 –553. 24. Hammer NA, Lilleso J, Pedersen JL, et al. Effect of riluzole on acute pain and hyperalgesia in humans. Br J Anaesth 1999;82:718–722. 25. Galer BS, Twilling LL, Harle J, et al. Lack of efficacy of riluzole in the treatment of peripheral neuropathic pain conditions. Neurology 2000;55:971–975. 26. Maeda S, Kawamoto A, Yatani Y, et al. Gene transfer of GLT-1, a glial glutamate transporter, into the spinal cord by recombinant adenovirus attenuates inflammatory and neuropathic pain in rats. Mol Pain 2008;4:65. 27. Pittenger C, Coric V, Banasr M, et al. Riluzole in the treatment of mood and anxiety disorders. CNS Drugs 2008;22:761–786. 28. Zarate CA, Manji HK. Riluzole in psychiatry: a systematic review of the literature. Expert Opin Drug Metab Toxicol 2008;4:1223–1234. 29. North CS, Hong BA, Alpers DH. Relationship of functional gastrointestinal disorders and psychiatric disorders: implications for treatment. World J Gastroenterol 2007;13:2020 –2027. 30. Gosselin RD, Gibney S, O’Malley D, et al. Region specific decrease in glial fibrillary acidic protein immunoreactivity in the brain of a rat model of depression. Neuroscience 2009;159:915–925.
RILUZOLE AND VISCERAL HYPERSENSITIVITY
2425
31. Rajkowska G, Miguel-Hidalgo JJ. Gliogenesis and glial pathology in depression. CNS Neurol Disord Drug Targets 2007;6:219–233. 32. Honore P, Rogers SD, Schwei MJ, et al. Murine models of inflammatory, neuropathic and cancer pain each generates a unique set of neurochemical changes in the spinal cord and sensory neurons. Neuroscience 2000;98:585–598. 33. Bradesi S, Svensson CI, Steinauer J, et al. Role of spinal microglia in visceral hyperalgesia and NK1R up-regulation in a rat model of chronic stress. Gastroenterology 2009;136:1339 –1348, e1– e2. 34. Tao YX, Gu J, Stephens RL Jr. Role of spinal cord glutamate transporter during normal sensory transmission and pathological pain states. Mol Pain 2005;1:30. 35. Weng HR, Chen JH, Cata JP. Inhibition of glutamate uptake in the spinal cord induces hyperalgesia and increased responses of spinal dorsal horn neurons to peripheral afferent stimulation. Neuroscience 2006;138:1351–1360. 36. Lin Y, Tian G, Roman K, et al. Increased glial glutamate transporter EAAT2 expression reduces visceral nociceptive response in mice. Am J Physiol Gastrointest Liver Physiol 2009;296:G129–G134. 37. Weng HR, Aravindan N, Cata JP, et al. Spinal glial glutamate transporters downregulate in rats with taxol-induced hyperalgesia. Neurosci Lett 2005;386:18 –22. 38. Tilleux S, Hermans E. Neuroinflammation and regulation of glial glutamate uptake in neurological disorders. J Neurosci Res 2007;85:2059 –2070. 39. Dinan TG, Quigley EM, Ahmed SM, et al. Hypothalamic-pituitary-gut axis dysregulation in irritable bowel syndrome: plasma cytokines as a potential biomarker? Gastroenterology 2006;130:304–311. 40. Rosas S, Vargas MA, Lopez-Bayghen E, et al. Glutamate-dependent transcriptional regulation of GLAST/EAAT1: a role for YY1. J Neurochem 2007;101:1134 –1144. 41. Zerangue N, Arriza JL, Amara SG, et al. Differential modulation of human glutamate transporter subtypes by arachidonic acid. J Biol Chem 1995;270:6433– 6435.
Received August 20, 2009. Accepted March 4, 2010. Reprint requests Address requests for reprints to: John F. Cryan, PhD, Department of Pharmacology and Therapeutics, School of Pharmacy, University College Cork, Cavanagh Pharmacy Building, College Road, Cork, Ireland. e-mail: [email protected]; fax (353) 21490 1656. Dr Gosselin is currently at the Department of Anesthesiology, University Hospital Center and Department of Cell Biology and Morphology, University of Lausanne, Lausanne, Switzerland. Acknowledgments The authors are grateful to Pat Fitzgerald, Sinead Gibney, and Siobhain O’Mahony for the excellent technical assistance with behavioral experiments and to Nora Dwyer for help with optical quantification of EAAT-1 immunoreactivity. Conflicts of interest The authors disclose the following conflicts: The authors received industry support from GlaxoSmithKline. The Alimentary Pharmabiotic Centre is partly funded by GlaxoSmithKline. Funding JFC and TGD are supported in part by Science Foundation Ireland in the form of a Centre grant (Alimentary Pharmabiotic Centre).
BASIC– ALIMENTARY TRACT
June 2010