Secondary Logo

Journal Logo

Apraclonidine Attenuates the Increases in Spinal Excitatory Amino Acid Release in Rats with Adjuvant-Induced Inflammation

Lin, Chung-Ren, MD*; Wang, Cheng-Haung, MD*; Wu, P.-C., BA*; Wen, Zhi-Hong, BA*; Buerkle, Hartmut, MD, PhD; Yang, Lin-Cheng, MD*

doi: 10.1097/00000539-200203000-00041
Pain Medicine: Research Report

The release of excitatory amino acids (EAAs), nitric oxide, and prostaglandins plays a critical role in the development of peripheral tactile and thermal hypersensitivity after the induction of knee joint inflammation. In this study, we used a model of chronic spinal microdialysis to examine the effect of complete Freund’s adjuvant (CFA)-induced inflammation on the spinal release of EAAs and also assessed the antinociceptive effect of a new α2-adrenergic agonist, apraclonidine, by using this model. Male Sprague-Dawley rats were implanted with microdialysis catheters. CFA was injected into the plantar surface of the left hindpaw to induce inflammation. Concentrations of amino acids in dialysate and thermal and tactile withdrawal latency were evaluated for 1 wk. Intraplantar injection of CFA evoked a significant release of glutamate, aspartate, and citrulline for 6 days. Three milligrams of intraperitoneal apraclonidine significantly suppressed the release of EAAs and citrulline. Apraclonidine was given intraperitoneally 2–3 days after CFA injection. Prominent thermal and tactile allodynia was observed for 6 days. Our results show that the significant modulatory effect of the α2-adrenergic agonist apraclonidine on the release of EAAs may account for its antinociceptive properties in adjuvant-induced inflammation.

*Department of Anesthesiology, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University, Kaohsiung Hsien, Taiwan; and †Department of Anesthesiology, Westfaelische Wilhelms-Universität Muenster, Germany

Supported in part by Chang Gung Memorial Hospital Grant CMRP980 and National Science Council Grant NMRP9030G.

October 22, 2001.

Address correspondence and reprint requests to Lin-Cheng Yang, MD, Department of Anesthesiology, Chang Gung Memorial Hospital, 123 Ta-Pei Road, Niao Shung Hsiang, Kaohsiung Hsien, 833, Taiwan. Address e-mail to

Pathological pain states are primarily associated with significant activation in spinal excitatory amino acid (EAA) receptors and subsequent intracellular events (1). The release of EAAs into the spinal extracellular milieu plays an important role in the modulation of peripheral inflammation and in the transmission of peripheral painful stimuli within the spinal cord. Activation of glutamate receptors underlies central hypersensitivity, whereby the level of transmission of noxious messages is potentiated (2). It is important to note that the inhibitory and EAA receptor-mediated events in the spinal cord seem to determine the level of pain transmission (3). Thus, information concerning amplification systems of the N-methyl-d-aspartate (NMDA) receptor of the spinal cord is a step toward understanding why and how a painful response is not always matched to the stimulus (4). The roles of EAAs, citrulline, and prostaglandin E2 have been defined in the development of thermal and tactile allodynia for a 24-h period after complete Freund’s adjuvant (CFA)-induced knee joint inflammation using a technique of chronic spinal cord microdialysis (5). The process of achieving a progressive or an acutely evolving pain state is not static but rather is based on the magnitude, character, and duration of the stimulus applied. Moreover, the roles of inhibitory transmitter systems can also change. Opioid, glycine, and γ-aminobutyric acid (GABA) transmission in the spinal cord can vary in different pain states (1,6). Whether adjuvant-induced plantar inflammation evokes a prolonged increase in the concentration of EAAs for days has not been investigated.

The facilitated central sensitivity of the chronic pain state is triggered by the pre- and postsynaptic release of EAA and may be altered by the presynaptic and postsynaptic α2-adrenoceptor agonist clonidine (7). However, clonidine induces sympathetic flow inhibition and a decrease in blood pressure as a result of α2-adrenoceptor stimulation in the brainstem (8). In addition to the cardiovascular side effects, there are sedation and dry mouth, which may limit its clinical application. New interest has focused on peripherally restricted analgesics such as the α2-agonist apraclonidine, which mimics the activity of clonidine (9,10). Our group studied the effects of intraperitoneal (IP) apraclonidine on spinal release of EAAs and the development of thermal and tactile allodynia in rats with CFA-induced inflammation.

Back to Top | Article Outline


All studies were performed according to protocols approved by the Institutional Animal Care and Use Committee of Chang Gung University. Male Sprague-Dawley rats (350–375 g; Sprague Dawley, National Science Council, Taiwan) were used in this study. Animals were anesthetized in a Plexiglas box with 3% isoflurane in an oxygen/room air mixture (1:1). A loop-style microdialysis probe (9 cm long) was implanted intrathecally by using a previously described technique (11). Briefly, rats were placed in a stereotaxic headholder, and the cisternal membrane was exposed. The loop catheter was inserted through a cisternal incision and passed 9 cm into the intrathecal space. This placed the tip of the catheter at the L5 spinal segment. After implantation, all animals were allowed to recover for a minimum of 2 days before being used in an experiment. Rats showing motor weakness or signs of paresis upon recovery from anesthesia were killed. The recovery rate of the microdialysis for each catheter was measured before each experiment. The catheters were perfused with artificial cerebrospinal fluid (ACSF) at a rate of 5 μL/min, and the membranes were exposed to a dextrose solution of a known concentration. The dextrose solution was stirred during perfusion, and the temperature of the solution was maintained at 37°C.

After 2 days, animals were reanesthetized with 2.5% isoflurane, and the dialysis catheters were perfused with ACSF at a rate of 5 μL/min. The ACSF contained (in mM) 151.1 Na+, 2.6 K+, 0.9 Mg2+, 1.3 Ca2+, 122.7 Cl, 21.0 HCO3, 2.5 HPO4, and 3.5 dextrose. The ACSF was bubbled with 95% oxygen/5% CO2 before each experiment to adjust the final pH to 7.2. All experimental manipulations were preceded by a 30-min washout period and then followed by two control sample (10 min each) collections. After the baseline samples were collected, the animals were randomly divided into the following groups and treated as follows: Group A (n = 6), control noninjected animals; Group B (n = 6), saline intradermal injection (0.2 mL) at the plantar surface of the left hindpaw; Group C (n = 6), CFA intradermal injection (0.2 mL) at the plantar surface of the left hindpaw; Group D (n = 6), saline injected IP 2 days after CFA injection; Group E (n = 6), apraclonidine (Lotus Laboratories, Inc., Taiwan) (0.3 mg) injected IP 2 days after CFA injection; Group F (n = 6), apraclonidine (3 mg) injected IP 2 days after CFA injection; Group G (n = 6), apraclonidine (3 mg) injected IP 3 days after CFA injection.

Animals were allowed to recover, and dialysate samples were collected 3 h after the induction of CFA-induced inflammation. After 3 h, the animals were disconnected from the syringe pump and returned to their cages. At Days 2, 3, and 6, animals were reconnected to the syringe pump and allowed a 30 min washout. Dialysate samples were collected. Ten microliters of dialysate was assayed with on-line high-performance liquid chromatography (HP1100; Hewlett-Packard, Palo Alto, CA) coupled with a fluorescence detector (12). The detection sensitivity was up to 10−8 M. Sensitivity was 5–10 pmol/10-μL tube.

After the baseline dialysate samples were collected, animals, while under isoflurane anesthesia, were injected on the plantar surface of the left hindpaw with 0.2 mL of CFA (containing Mycobacterium butyricum) at a concentration of 5 mg/mL. Immediately after the injection, the animals were allowed to recover (full recovery time 5–10 min), and all survived for 1 wk.

To assess thermal paw withdrawal latency (TPWL), a modified technique originally described by Hargreaves et al. (13) was used. The animal was placed on a glass plate (maintained at 30°C) for 5–10 min for adaptation. The latency between application of a focused light beam and the hindpaw withdrawal response was measured to the nearest 0.1 s. The cutoff time in the absence of a response was 20 s. This value was then assigned as the response latency.

To assess mechanical paw withdrawal thresholds (MPWT), animals were placed in a plastic cage with a wire mesh bottom and allowed to acclimate for approximately 15 min. The midplantar right (injected) and left (noninjected) paws were then touched with one of a series of eight von Frey hairs (0.41, 0.70, 1.20, 2.00, 3.63, 5.50, 8.50, and 15.1 g) (Stoetling) for 5 s. Stimuli for each paw were applied in 10- to 15-s intervals. A sharp withdrawal of the paw was considered as a positive response. The TPWL and MPWL were measured before CFA injection and then at 3 h, 2 days, 3 days, and 6 days after CFA injection. The experimenter was blinded to treatment classification throughout the study, although inflammation related to the adjuvant was evident during behavioral testing.

At the end of the 6-day experimental period, animals were deeply anesthetized and perfused with saline followed by 4% paraformaldehyde. Twenty-four hours later, the spinal cords were removed from the vertebral column and the localizations of the loop and injection catheters verified. For histopathological analysis of the spinal cords, the L2 to L5 spinal cord segments were dissected and postfixed in 1% buffered OsO4 and embedded in Araldite. Ten subserial semithin sections (1 μm thick) were then cut and stained with p-phenylenediamine. Sections were evaluated for the presence of dark-staining neurons or a vacuolized type of degeneration.

All data were analyzed with a commercially available computer program (SPSS 8.0; SPSS Inc., Chicago, IL). Statistical analysis of biochemical results was performed with nonparametric analysis of variance (Kruskal-Wallis test followed by the Mann-Whitney U-test). Comparison between individual treatment groups and time points was done with Wilcoxon’s signed rank test. To compare groups, a post hoc Fisher least significant difference test was used. A P value of <0.05 was considered significant. EAA release data and mechanical and thermal nociceptive data were expressed as percentage change from baseline ± sd.

Back to Top | Article Outline


In the noninjected animals (Group A) , no significant changes in behavior were seen in this control group. TPWL ranged from 9.6 to 12.7 s without significant differences during the whole 6-day testing period. MPWL ranged from 10.4 to 15 g. There was no significant change during the testing period.

In the saline-injected animals (Group B), TPWL ranged from 9.1 to 12.6 s without significant differences during the whole 6-day testing period. Injection of saline into the plantar surface resulted in a modest but not significant decrease in TPWL and MPWL of the ipsilateral paw at 3 h after injection, with nondetectable changes thereafter. Baseline MPWT ranged from 9.1 to 15 g. No significant changes in mechanical threshold ipsilateral to the saline injection were noted during the entire testing period.

In the CFA plantar-injected animals (Group C), the baseline TPWL was 10.8 ± 0.56 s and decreased 47%, 44%, 38%, and 64% at 3 h, 2 days, 3 days, and 6 days after CFA injection, respectively (Fig. 1) (P < 0.05). Similarly, a decrease of the MPWT on the side of injection was seen at 3 h after injection (P < 0.05) which also lasted for the entire 6-day testing period (Fig. 1) (P < 0.05).

Figure 1

Figure 1

Post-CFA injection treatment with apraclonidine (3 mg IP) significantly increased both TPWL and MPWT at Days 2 and 3 after CFA-induced inflammation (Fig. 2, A and B). In the spinal microdialysis study, baseline concentrations of all amino acids except GABA were detectable in all animals tested, with no significant differences among experimental groups (data not shown).

Figure 2

Figure 2

In the noninjected animals (Group A), there was no significant effect on the release of all amino acids in comparison to baseline. In the saline-injected animals (Group B), injection of saline into the skin evoked no detectable change in the concentration of glutamate, aspartate, glycine, or citrulline at any time points.

In the CFA-injected animals (Group C), there was a significant and prolonged increase in the concentrations of glutamate, aspartate, glycine, and citrulline. At 3 h after CFA injection, glutamate concentration increased to 188% ± 39% of baseline values. At Day 2, the glutamate concentration was 174% ± 27% above baseline. At Days 3 and 6 after CFA injection, glutamate concentrations were 170% ± 27% and 132% ± 25% of baseline, respectively. Concentrations of aspartate were significantly (P < 0.05) increased for the whole 6-day experimental period. The largest increase was detected at 3 h (181% ± 52%) and 3 days (219% ± 40%). Glycine revealed a modest increase at 3 h after injection (143 ± 39%, P < 0.05) and then gradually normalized, with no detectable changes at Day 2 after injection (126% ± 22%). Citrulline concentration was significantly (P < 0.05) increased at 3 h after injection (156% ± 34%), with the peak seen at Day 3 (173% ± 44%). Citrulline levels were restored to normal at Day 6 after injection (106% ± 46%;P > 0.05) (Fig. 3A).

Figure 3

Figure 3

In the animals posttreated with apraclonidine at Day 2 after CFA injection (Groups D, E, and F), significant suppression of release of glutamate, aspartate, and citrulline was seen in the Apraclonidine 3 mg group (P < 0.05). However, there was no significant change in glycine level (P > 0.05) (Fig. 3B).

In comparison with saline-treated animals, posttreatment with apraclonidine 3 mg IP at Day 3 after CFA injection significantly decreased the concentration of citrulline, aspartate, and glutamate (P < 0.05). Glycine levels showed no significant difference in concentrations when compared with those of the nontreated group (Fig. 3C).

Light microscopy analysis of the spinal cord sections taken from the L2 to L5 segments showed no evidence of pathological changes in all spinal cord levels examined when specifically analyzed for the presence of “dark” forms of neuronal degeneration or the presence of cytoplasmic vacuolization. All neuronal pools, including large alpha-motorneurons and small- and medium-sized interneurons, showed fully preserved nucleus and nucleolus, with no detectable changes in neuropils (data not shown).

Back to Top | Article Outline


Development of central sensitization plays a key role in the mechanism of peripheral hypersensitivity in experimental arthritis (1). After the induction of peripheral inflammation, the ensuing central sensitization is related to the activation of glutamate receptors in the spinal cord (2). Yet glutamate also participates in the peripheral modulation of thermal hyperalgesia (14–16). In arthritis induced by CFA or formalin, both centrally and peripherally administered glutamate antagonists reduce the degree of joint inflammation and thermal hyperalgesic response (15). In this study, the concentrations of glutamate and aspartate in the plantar area of the rat hindpaw were not measured, but possibly peripheral NMDA and non-NMDA glutamate receptors contributed to central sensitization. Behaviorally, IP administration of apraclonidine significantly blocked the development of tactile and thermal hypersensitivity at Days 2 and 3 after CFA injection.

Apraclonidine injection also blocked the release and increase in spinal amino acid concentrations. These data suggest that the development of tactile and thermal hypersensitivity seen during this kind of inflammation and the corresponding central sensitization can be related to the increased release of spinal EAAs. α2-Adrenoceptors are located on primary afferent terminals (both at peripheral and spinal endings), on neurons in the superficial laminae of the spinal cord, and within several brainstem nuclei implicated in analgesia (17). Whether the analgesic effect of apraclonidine is caused by blockade in a selective or peripheral α2-adrenergic receptor pathway is, however, not clear. Apraclonidine is a derivative of the systemic antihypertensive drug clonidine, which decreases intraocular pressure by decreasing the production of aqueous humor (17,18). Unfortunately, clonidine has been observed to markedly decrease systemic blood pressure and ocular perfusion pressure, even when applied topically (17). Apraclonidine is similar to clonidine, but it does not cross the blood-brain barrier and therefore does not cause systemic hypotension (10). However, the evidence for peripheral selective activity of apraclonidine is still lacking, and that reported in the literature has not been entirely convincing. One study, in cats, that used local ophthalmic application of 30 mL of 0.5% apraclonidine resulted in major side effects that included changes in the heart rate as well as nausea and vomiting in almost 100% of the treated animals (19). In a previous unpublished observation, we could not reveal any antinociceptive central effects of apraclonidine by the local intraplantar injection of 0.3 mg of apraclonidine. This study was conducted to investigate the effects of apraclonidine on in vivo EAA (glutamate, aspartate, and citrulline) concentrations in an animal model of CFA-induced inflammation. Extracellular glutamate concentration increased twofold in the CFA-treated group during the 7-day period of inflammation. IP administration of apraclonidine at Days 2 and 3 after inflammation attenuated the increase in extracellular glutamate concentrations. Clearly, the antinociceptive effect of apraclonidine was correlated with a near complete block in EAA release, whereas EAAs were otherwise increased in this hypersensitivity state.

This work indicates that the α2 agonist apraclonidine blocked the release of EAAs by a peripheral action not involving central α2 receptors. In fact, it has been postulated that peripherally restricted α2 agonists may be useful in the management of neuropathic pain. L-659,066, a peripherally restricted α2-adrenergic receptor antagonist, could only block the analgesic actions of the novel α2 agonist dexmedetomidine in neuropathic rats, with no effect in control animals. This suggests that in neuropathic pain states, dexmedetomidine exerts its analgesic effect via a site outside the central nervous system (20). Moreover, in CFA-induced arthritis, lipid peroxidation and disruption of lysosomal integrity have been implicated in the pathogenesis of inflammatory processes (3). Furthermore, the concentration of glutamate, arginine, and citrulline measured by a microdialysis probe inserted into the joint was doubled, and this increase persisted for at least two hours after knee joint injection with the irritants (21). Apraclonidine may play a role in protecting against these deleterious effects by reducing the pulse generation from periphery and the magnitude of thermal hyperalgesia and tactile allodynia.

It is interesting to note that in animals injected with CFA, glutamate and aspartate levels remained high throughout the 6-day sampling period. This prolonged release might indicate continuing afferent input from the inflammation and gradual sensitization of the spinal cord neurons and increased secondary postsynaptic release. Light-microscopic analysis of the spinal cords showed no evidence of neuronal degeneration—no dark or vacuolized neurons. These results suggested that central sensitization or prolonged afferent input after CFA-induced inflammation did not cause significant damage to the spinal cord neuron. In our experiment, peripheral inflammation promoted spinal nitric oxide synthesis, because citrulline levels were increased from three hours and peaked at Day 3 after experimental arthritis. The early increase in the levels of nitric oxide may increase the responsiveness of spinal neurons to EAAs and potentiate the secondary release of endogenous glutamate.

We conclude that the peripheral α2-adrenoreceptor agonist apraclonidine is effective in attenuating the release of glutamate, aspartate, and citrulline in CFA-induced arthritis. The hypersensitivity state during the entire period seems dependent on increased release of spinal EAAs.

Back to Top | Article Outline


1. Sorkin LS, Sluka KA, Dougherty PM, et al. Neural changes in acute arthritis in monkeys. IV. Time-course of amino acid release into the lumbar dorsal horn. Brain Res Brain Res Rev 1992; 17: 39–50.
2. Sluka KA, Jordan HH, Willis WD, Westlund KN. Differential effects of N-methyl-D-aspartate (NMDA) and non-NMDA receptor antagonists on spinal release of amino acids after development of acute arthritis in rats. Brain Res 1994; 664: 77–84.
3. Agha AM, Gad MZ. Lipid peroxidation and lysosomal integrity in different inflammatory models in rats: the effects of indomethacin and naftazone. Pharmacol Res 1995; 32: 279–85.
4. Dickenson AH, Chapman V, Green GM. The pharmacology of excitatory and inhibitory amino acid-mediated events in the transmission and modulation of pain in the spinal cord. Gen Pharmacol 1997; 28: 633–8.
5. Yang LC, Marsala M, Yaksh TL. Characterization of time course of spinal amino acids, citrulline and PGE2 release after carrageenan/kaolin-induced knee joint inflammation: a chronic microdialysis study. Pain 1996; 67: 345–54.
6. Renno WM, Beitz AJ. Peripheral inflammation is associated with decreased veratridine-induced release of GABA in the rat ventrocaudal periaqueductal gray: microdialysis study. J Neurol Sci 1999; 163: 105–10.
7. Eisenach JC, De Kock M, Klimscha W. Alpha sub 2-adrenergic agonists for regional anesthesia: a clinical review of clonidine—1995. Anesthesiology 1996;1984;85:655–74.
8. Oster JR, Epstein M. Use of centrally acting sympatholytic agents in the management of hypertension. Arch Intern Med 1991; 151: 1638–44.
9. Abrams DA, Robin AL, Pollack IP, et al. The safety and efficacy of topical 1% ALO 2145 (p-aminoclonidine hydrochloride) in normal volunteers. Arch Ophthalmol 1987; 105: 1205–7.
10. Coleman AL, Robin AL, Pollack IP, et al. Cardiovascular and intraocular pressure effects and plasma concentrations of apraclonidine. Arch Ophthalmol 1990; 108: 1264–7.
11. Marsala M, Malmberg AB, Yaksh TL. The spinal loop dialysis catheter: characterization of use in the unanesthetized rat. J Neurosci Methods 1995; 62: 43–53.
12. Yang LC, Marsala M, Yaksh TL. Effect of spinal kainic acid receptor activation on spinal amino acid and prostaglandin E2 release in rat. Neuroscience 1996; 75: 453–61.
13. Hargreaves K, Dubner R, Brown F, et al. A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia. Pain 1988; 32: 77–88.
14. Carlton SM, Coggeshall RE. Inflammation-induced changes in peripheral glutamate receptor populations. Brain Res 1999; 820: 63–70.
15. Jackson DL, Graff CB, Richardson JD, Hargreaves KH. Glutamate participates in the peripheral modulation of thermal hyperalgesia in rats. Eur J Pharmacol 1995; 284: 321–5.
16. Omote K, Kawamata T, Kawamata M, Namiki A. Formalin-induced release of excitatory amino acids in the skin of the rat hindpaw. Brain Res 1998; 787: 161–4.
17. Harrison R, Kaufmann CS. Clonidine effects of a topically administered solution on intraocular pressure and blood pressure in open-angle glaucoma. Arch Ophthalmol 1977; 95: 1368–73.
18. Macri FJ, Cevario SJ. Clonidine. Arch Ophthalmol 1978; 96: 2111–3.
19. Miller PE, Rhaesa SL. Effects of topical administration of 0.5% apraclonidine on intraocular pressure, pupil size, and heart rate in clinically normal cats. Am J Vet Res 1996; 57: 83–6.
20. Poree LR, Guo TZ, Kingery WS, Maze M. The analgesic potency of dexmedetomidine is enhanced after nerve injury: a possible role for peripheral alpha 2-adrenoceptors. Anesth Analg 1998; 87: 941–8.
21. Lawand NB, McNearney T, Westlund KN. Amino acid release into the knee joint: key role in nociception and inflammation. Pain 2000; 86: 69–74.
© 2002 International Anesthesia Research Society