| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
To determine whether intraarticular pretreatment with N-methyl-D-aspartic (NMDA) receptor antagonist ketamine or memantine currently used in humans has prophylactic analgesia in arthritic pain, we examined the effects of their intraarticular injection before carrageenan injection into the knee joint on pain-related behavior and spinal c-Fos expression in rats. Injection of ketamine (0.2 mg and 1 mg) or memantine (0.1 mg, 0.2 mg, and 1 mg) into the knee joint, but not the abdominal cavity, immediately before carrageenan injection (2%, 40 µL) significantly prevented pain-related behavior. The intraarticular injection of ketamine (1 mg) or memantine (0.2 mg) also suppressed c-Fos expression in the laminae I-II and laminae V-VI at the L3-4 spinal level. Subsequent statistical analyses revealed that the degree of the spinal c-Fos expression was correlated with the extent of the pain-related behavior. These results suggest that peripheral administration of NMDA receptor antagonists has prophylactic analgesic effects in arthritic pain, which might be associated with the decrease of central nociceptive signaling. Because ketamine and memantine are currently used in humans and considered clinically safe, they may have therapeutic value in the treatment of joint pain. IMPLICATIONS: This study demonstrated that peripheral pretreatment with ketamine and memantine suppressed both arthritic pain-related behavior and spinal c-Fos expression. These results indicate that both ketamine and memantine have prophylactic analgesia and provide support for their potential use as peripheral analgesics for joint pain.
Accumulating evidence suggests that glutamate and its receptors are involved in peripheral nociceptive transduction (16). It has been demonstrated that glutamate receptors are localized on primary afferent axons and are increased after the induction of inflammation (1). Furthermore, glutamate is released into the knee joint and that the glutamate content of the medial articular nerve is increased after the induction of arthritis (5,6). Peripheral administration of N-methyl-D-aspartic (NMDA) receptor antagonists attenuated nociceptive behaviors in several pain models (24,79). c-Fos expression, at the spinal level, is one of the long-term intracellular events that is used as an indirect marker of nociceptive processes (10). Neuronal excitation leads to a rapid and transient induction of c-Fos (11). Once expressed, c-Fos protein acts as a transcriptional factor binding to DNA, thereby regulating the expression of nearby promoters that, in turn, regulates the activation of a target gene (12). Previous studies have demonstrated that inflammation of various peripheral regions, including the knee joint, evoked c-Fos expression in the dorsal horn of the spinal cord (13,14). Also, intraplantar injection of MK-801 suppressed spinal c-Fos expression induced by formalin injection into the paw (15). Therefore, spinal c-Fos expression may be involved in central nociceptive processes or integration associated with inflammatory pain. It is unknown whether intraarticular injection of clinically available NMDA receptor antagonists may produce prophylactic analgesia in arthritic pain, and the analgesia induced by these antagonists may be associated with the decrease of central nociceptive signaling. Accordingly, we examined whether intraarticular pretreatment with ketamine and memantine suppressed arthritic pain-related behavior and spinal c-Fos expression induced by carrageenan injection into the knee joint. We used ketamine and memantine instead of MK-801, which has toxic properties and low safety margins that preclude its clinical use (16,17).
The present study was performed in accordance with a protocol approved by the University Committee on Laboratory Animals. Male Sprague-Dawley rats (190210 g; n = 113; purchased from Sam, Korea) were housed in a cage of 5 in a temperature-controlled room (22°C25°C) illuminated from 7:00 AM to 7:00 PM. Food and water were available ad libitum. To induce arthritis, 2% lambda-carrageenan (40 µL, in saline) was injected into the knee joint of the right hind limb under enflurane anesthesia (0.5%2%). To be sure that carrageenan injection induced arthritis, we measured the diameters of right and left knee joints before and after the carrageenan injection. The knee joint diameter was defined as the distance between lateral and medial collateral ligament regions. One day before and 4, 8, and 12 h, and 1, 3, and 5 days after the injection of carrageenan, weight-load tests for pain-related behavior were performed. In the inflamed joint, weight loading generated pain perception, which is the major symptom in arthritis. So we used a convenient method (18) for assessment of arthritic pain in freely walking rats. This allowed measurement of weight load on each limb while the rat was walking through a path, the bottom of which was equipped with strain gauge weight sensors. The output of each load cell was fed to a digital amplifier (Cyber Amp 380, Axon Instruments, Forest City, CA) for appropriate amplification and filtering. The processed signal was sent to a personal computer via an analog-digital converter (1401 Plus, CED Ltd, Cambridge, United Kingdom) and plotted as time-weight curve using software (Spike 2, CED). While an rat was walking through the path, the investigator identified the plates on which the rat placed the paw of interest. Output signals from the load cells attached to these plates were selected for plotting of time-weight curves. The test was repeated 2 or 3 times until at least 5 time-weight curves were obtained for a given limb. Ketamine (0.1 mg, 0.2 mg, and 1 mg; 40 µL; n = 31) and memantine (0.1 mg, 0.2 mg, and 1 mg, 40 µL; n = 32) were dissolved in saline and injected into the knee joint cavity of right hind limb immediately before carrageenan injection. The selection of doses used in this study was based on previous studies (7,9) and our preliminary experiments. Control rats were injected with either saline (40 µL; n = 10) or morphine (0.4 mg; 40 µL; n = 10). To confirm that intraarticular injection of ketamine and memantine produced their effects via a peripheral site of action, we also examined the effects of intraperitoneal injection of ketamine (1 mg; 40 µL; n = 5) and memantine (0.2 mg; 40 µL; n = 5). The drug treatments and weight-load measurements were performed in a double-blind fashion. All drugs used in this study were purchased from Sigma Chemical Co. (St. Louis, MO). Spinal c-Fos expression was examined in four different groups. The drugs were injected into the knee joint as follows: (a) saline (n = 4), (b) saline + carrageenan (n = 6), (c) ketamine 1 mg + carrageenan (n = 6), and (4) memantine 0.2 mg + carrageenan (n = 4). The doses of ketamine and memantine were selected according to their stable effects during all the time points in behavioral tests. Ketamine, memantine, or saline was injected into the knee joint immediately before the carrageenan injection. The volume of each injection was 40 µL. The rats were killed after weight load tests 4 h after the carrageenan injection. The time point, 4 h, was chosen because the decrease of weight load was most severe at 4 h after the carrageenan injection. Under urethane anesthesia (1.25 g/kg intraperitoneally), they were transcardially perfused with heparinized saline (0.9%) followed by 4% paraformaldehyde in 0.1 M of phosphate buffer (PB; pH value of 7.2). The L3 and L4 spinal segments were excised and postfixed for 68 h in the same perfusate and cryoprotected overnight in 0.1 M of PB containing 30% sucrose at 4°C. The segments were sectioned at 1525 µm on a freezing microtome. The sections were reacted with anti-Fos antibody, polyclonal antiserum generated in rabbits and directly against the Fos protein (Oncogene Science, Boston, MA), according to the ABC method (Vector Elite Kit, Vector, Burlingame, CA). They were rinsed with 1% bovine serum albumin and 10% normal goat serum (NGS) for 1 h and incubated in Fos antisera (Ab-5 solution diluted 1:10,000) for 48 h at 4°C. The sections were then rinsed sequentially in 0.05 M of PB saline (PBS) and 3% NGS (30 min each) and incubated in diluted biotinylated goat anti-rabbit immunoglobulin G PBS solution for 1 h. After this, they were rinsed sequentially in 1% and 3% NGS and incubated in an avidin-biotinylated horseradish peroxidase complex for 1 h. After 3 10-min washes in PBS, the sections were incubated in a solution of diaminobenzidine (0.05%) containing 0.01% hydrogen peroxide for 510 min. Reacted sections were dehydrated, cleared, cover slipped, and examined with light microscopy. For each rat, 10 sections that had the largest numbers of Fos-like immunoreactive (Fos-LI) neurons in the superficial laminae (laminae I and II) and the neck of the dorsal horn (laminae V and VI) associated with nociceptive transduction were selected, and the numbers of these neurons were counted without considering the intensity of staining and averaged for each of the regions. The analysis of variance test was used to analyze the behavioral test and immunohistochemical scores. Post hoc comparison test (Tukey test) was further performed to identify the source of the variance. Correlation test (Pearson test) was used to determine the degree of the correlation between the weight load of the inflamed leg and ipsilateral spinal Fos expression. All data were presented as the mean ± SEM. P < 0.05 was considered significant.
Carrageenan injection into the knee joint cavity of the right hindlimb significantly reduced the weight load on this limb. This observation was consistent with previous reports (18,19). Intraarticular administration of ketamine, memantine, and morphine partially prevented the reduction of the weight load induced by carrageenan injection (Fig. 1). The effects of ketamine at the doses of 0.2 and 1 mg were statistically significant at all the time points examined (Fig. 1A). The effects of memantine against carrageenan-induced reduction of weight load were similar to those of ketamine. At all of the doses tested (0.1, 0.2, and 1 mg), their effects at 4 h and 8 h after carrageenan injection were statistically significant. At the 0.2-mg dose, the effect of memantine against weight-load reduction lasted for 3 days (Fig. 1B). However, intraperitoneal injections of ketamine (1 mg) and memantine (0.2 mg) had no effect on the weight-load reduction induced by carrageenan (Fig. 1). In addition, ketamine and memantine had no significant effects on carrageenan-induced increase of knee-joint diameter (Table 1).
Fos-LI neurons were located mostly in laminae I-II and laminae V-VI of the dorsal horn ipsilateral to the carrageenan-injected knee joint (Fig. 3). Fos expression was scarce in rats injected with saline alone (laminae I-II: 2.75 ± 0.25 neurons per section; laminae V-VI: 1.5 ± 0.29 neurons per section; Fig. 2A and 3). However, it was abundant in carrageenan-injected rats (laminae I-II: 15.00 ± 0.52 neurons per section; laminae V-VI: 25.00 ± 1.15 neurons per section; Fig. 2B and 3 ).
Intraarticular injection of ketamine (1 mg) and memantine (0.2 mg) suppressed carrageenan-induced c-Fos expression. Ketamine reduced the number of Fos-LI neurons in both laminae I-II (6.67 ± 0.49 neurons per section) and laminae V-VI (13.17 ± 0.54 neurons per section) (Fig. 2C and 3). Memantine also reduced the number of Fos-LI neurons in laminae I-II (6.5 ± 0.29 neurons per section) and laminae V-VI (12.00 ± 1.22 neurons per section) (Fig. 2D and 3). To see if pain-related behavior was related to spinal c-Fos expression, the Pearson correlation test was performed for the two measured variables, i.e., the weight load of the injected leg and the number of Fos-LI neurons on the ipsilateral side. For this, the data from the rats injected with (1) saline alone, (2) saline + carrageenan, (3) ketamine + carrageenan, and (4) memantine + carrageenan were all combined. As illustrated in Figure 4, the weight load had a strong negative correlation with the Fos expression in both laminae I-II (r = 0.936) and laminae V-VI (r = 0.963).
The present study demonstrated that intraarticular pretreatment with ketamine and memantine, which are currently used in humans, prevented arthritic pain-related behavior induced by carrageenan injection into the knee joint. In addition, we showed that there was correlation between the degree of pain-related behavior and spinal c-Fos expression induced by intraarticular injection of ketamine and memantine in individual rats 4 h after carrageenan. Therefore, our results indicate that peripheral administration of NMDA receptor antagonists has prophylactic analgesic effects in arthritic pain, which may be associated with the decrease of central nociceptive signaling. In accordance with our results, previous studies have reported that local cutaneous administration of MK-801 (2), ketamine, and memantine (7) reduced nociceptive behaviors in rats induced by formalin injection into the paw. Similarly, intraplantar injection of MK-801 suppressed spinal c-Fos expression induced by formalin injection into the paw (15). However, intraarticular injection with ketamine and memantine did not alter the carrageenan-induced increase of knee-joint diameter, indicating that intraarticular pretreatment with ketamine and memantine has no antiinflammatory effects. In the present study, the antinociceptive effects induced by these antagonists did not seem to be mediated by their central actions caused by the systemic leakage because intraperitoneal injection of the same dose and volume of these antagonists had no effects on the weight-load change. Intraplantar injection of MK-801, a potent NMDA receptor antagonist, is effective in reducing nociceptive behaviors in several different pain models (2,3,8). However, it could not be applied to the clinical setting because it has undesirable side effects such as ataxia, psychotomimetic actions, cognitive impairment, and neurotoxicity (16,17). In contrast, the NMDA antagonists used in the present study are considered clinically safe. In humans, ketamine is used as an anesthetic and has analgesic effects on various pain states at a subanesthetic dose (2022), and memantine has been used to treat movement disorders and dementia (23). In animals, systemic (24), intrathecal (25), or intraplantar (7) injection of ketamine and memantine could have antinociceptive effects in different pain models. The results of the present study indicate that intraarticular pretreatment with these antagonists is also an effective route for arthritic pain relief. Fos expression has been widely used as an anatomical marker of activated areas in the central nervous system and is involved in central nociceptive processes (1012). In the present study, c-Fos expression was observed immediately after a behavioral test in the same rat, which had the advantage that we could directly demonstrate whether the antinociceptive effects induced by these antagonists were associated with the decrease of central nociceptive signaling. Thus, peripheral pretreatment with these antagonists in this study could have reduced arthritic pain by reducing primary afferent neural activity or central sensitization. The increase of the endogenous glutamate level in the knee joint after arthritis (5) results in activation of glutamate receptors located on primary afferent axons in the knee joint. Pretreatment with NMDA receptor blockers could thus partially prevent peripheral afferent input into the spinal cord. As a consequence, there might be a reduced central sensitization after the initial blockage of peripheral afferent activity, because our results showed that blockage of joint NMDA receptors led to the reduction of spinal c-Fos expression. Supporting our findings, an electrophysiological study showed that peripheral NMDA and non-NMDA receptors were involved in the induction of persistent firing of the spinal wide-dynamic-range neurons induced by bee venom injection (26). The antinociceptive effects of ketamine and memantine in this study were long lasting. Similarly, an earlier study (27) showed that peripheral administration of ketamine produced long-lasting inhibition of development of secondary hyperalgesia induced by burn injury in humans. The long-lasting antinociceptive effects might have been due to the inhibition of induction of peripheral and central sensitization. In other words, peripheral pretreatment with NMDA receptor blockers may inhibit the positive-feedback enhancement of nociceptor excitability triggered by glutamate acting at autoreceptors on the primary afferent nerves. In addition, peripheral pretreatment with NMDA receptor blockers may inhibit the Fos-mediated long-term change in dorsal horn neurons (28,29). However, the analgesic effect of the 1 mg memantine pretreatment was gone since 1 day after the carrageenan injection so the rats showed a similar degree of pain-related behavior as saline-pretreated rats. This indicates that the analgesic effect is time-limited at the largest dose used in this study, and the appropriate dose for a stable effect of memantine was 0.2 mg. Similar to our results, previous studies (30,31) showed that the largest dose of drugs produced weaker or even no effect than the smaller dose. The possible explanation for the bimodal effect of a 1-mg dose of memantine is that the larger dose of memantine primarily acts to antagonize NMDA receptors and also acts on serotonergic and noradrenergic systems (23) that facilitate the peripheral nociception. We could not exclude the possibility that the antinociceptive effects induced by ketamine or memantine observed in this study may also be attributed to other receptors besides NMDA receptors. In this regard, memantine also acts as a dopamine agonist (23), and the intraarticular injection of dopamine could reverse carrageenan-induced incapacitation (32). Ketamine interacts with opioid receptors as opioid agonists (33). In addition, the NMDA receptor antagonists MK-801 (34) and (+)-HA966 (35), as well as ketamine, can interact with opioids in the central nervous system. Other experiments are therefore required to investigate the interaction between NMDA and opioid systems in the periphery in antinociception. In summary, our results showed that intraarticular pretreatment with ketamine and memantine prevented arthritic pain-related behavior. Additionally, there was a correlation between a decrease of pain-related behavior and spinal c-Fos expression induced by intraarticular injection of these antagonists in individual rats four hours after carrageenan injection. These results indicate that peripheral administration of NMDA receptor antagonists has prophylactic analgesic effects on arthritic pain, which may be associated with the decrease of central nociceptive signaling. In addition, this study suggests a potential therapeutic value for the treatment of joint pain, because ketamine and memantine are more often clinically used as NMDA receptor antagonists.
Supported, in part, by a grant of the Korea Health 21 R&D Project, Ministry of Health and Welfare, Republic of Korea (01-PJ1-PG3205000148) and the Brain Korea 21 Project in 2001.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|