The mainstay of treatment for acute pain today remains the opioids—compounds that produce analgesia by binding to opiate receptors to reduce nociception (1). This large class of drugs comprises natural, semisynthetic (natural compounds chemically altered to improve their action), and completely synthetic compounds. However, tolerance (where escalating doses are needed to achieve a similar level of pain relief) to these drugs arises rapidly when they are administered chronically. For this reason, patients with chronic pain can be extremely difficult to treat. Pain specialists often resort to other adjuvants to augment the pain control achieved with opioids.
Increasing interest is being drawn to drugs that modulate the function of the N-methyl-d-aspartate (NMDA) receptors to augment pain-control regimens. NMDA receptors are members of the ligand-gated ion channel superfamily whose natural agonists are the amino acids glutamate and aspartate; they are important mediators of excitatory neurotransmission in the central nervous system (CNS) (2). There are five principal subunits that contribute to functional NMDA receptors. The first cloned subunit, termed “NR1,” is expressed ubiquitously in the CNS and may be alternatively spliced into eight possible forms. NR1 is thought to be paired with one or more of four secondary subunits—NR2A, 2B, 2C, or 2D—to produce functional receptors responsible for ligand-activated native currents (2). The secondary subunits display regional and developmental differences in their expression patterns (3) that, along with their electrophysiologic differences when paired with NR1, generate the heterogeneity of NMDA-activated currents in the CNS (4).
NMDA receptors have minimal baseline activity in pain circuits during normal activity. However, excitatory glutamate transmission is significantly enhanced in chronic pain syndromes. Specifically, they have been implicated in pain processing, neuronal plasticity, and the generation and maintenance of central hypersensitivity (5,6). In addition, an interaction between an upregulated glutamatergic system and the development of opioid tolerance has been observed in animals and humans (7–9).
On the basis of these findings, noncompetitive NMDA antagonists may contribute to the treatment of chronic pain directly by blocking glutamatergic neurotransmission and indirectly by decreasing opioid-induced tolerance (7). High-affinity noncompetitive NMDA receptor antagonists, such as MK-801 and phencyclidine, are potent neuropathic pain inhibitors (10). However, their psychoactive properties cause severe psychomimetic side effects at large doses. Recently, more attention has focused on antagonists such as ketamine and dextromethorphan, which have a lower affinity for the NMDA receptor and a resulting decrease in severe side effects (11,12)
There has been some suggestion that methadone also has the ability to block NMDA receptors (13,14). This dual quality may provide methadone with a unique mechanism of action. Only binding studies have established the interaction of methadone with NMDA receptors (15); in fact, methadone displaces the noncompetitive NMDA antagonist MK-801 with a binding Ki of 0.85 μM (5). We hypothesized that methadone can inhibit NMDA receptors at therapeutic concentrations and that these effects may be subunit specific. Therefore, we investigated the ability of methadone and its stereoisomers to inhibit the function of various subtypes of NMDA receptors that display regional expression differences. The results of these studies shed light on the effect of methadone’s inhibition of NMDA receptors as a possible mechanism of action.
All experimental procedures involving the South African clawed frog (Xenopus laevis) were approved by the Institutional Animal Use and Care Committee of the University of California-San Francisco and are similar to those previously described (16). Unfertilized Xenopus oocytes (Nasco, Fort Atkinson, WI) were removed from adult female frogs anesthetized with 0.35% tricaine and washed once in oocyte Ringer’s II solution ([mM] 82.5 NaCl, 2 KCl, 1 MgCl2, and 10 HEPES/Tris, pH 7.4), followed by collagenase (Type A; Boehringer Mannheim, Indianapolis, IN) treatment (1.5 mg/mL) for 1–1.5 h at room temperature with constant agitation to remove the follicular cell layer. Oocytes were then washed twice in oocyte Ringer’s II solution, twice in Ca2+-free modified Barth’s solution ([mM] 88 NaCl, 1 KCl, 2.4 NaHCO3, and 10 HEPES), and twice in Barth’s solution ([mM] 88 NaCl, 1 KCl, 0.82 MgSO4, 0.40 CaCl2, 0.33 Ca(NO3)2, 2.4 NaHCO3, and 10 HEPES/Tris) to which tetracycline 50 μg/mL and gentamicin 50 mg/mL had been added. For heteromeric NMDA receptor studies, 5′-capped complementary RNAs were synthesized by in vitro transcription (mMessage mMachine™; Ambion, Austin, TX) from expression plasmids encoding rat NMDA receptor subunit complementary cDNA clones (NR1-1a splice variant). NR1 and one of the NR2A, 2B, 2C, and 2D cRNAs were diluted 1:3 in ribonuclease-free water and mixed in a 1:1 volumetric ratio. Mature oocytes (Stage V and VI) were injected within 16 h after removal with approximately 50 nL of diluted transcript by using an automated microinjector (PV 830; WPI, Sarasota, FL). Sorted oocytes were stored in Barth’s solution at 16°C with gentle shaking (Belly Button; Stovall Life Sciences, Greensboro, NC).
Electrophysiological experiments were performed at room temperature (20°C–22°C). A single defolliculated oocyte was placed in a continuous-flow recording chamber (25-μL volume) and superfused with 3–5 mL/min frog Ringer’s solution (FR [mM]; 115 NaCl, 2.5 KCl, 1.8 CaCl2, and 10 HEPES, pH 7.4) containing glycine 10 μM. Glycine was added as a coagonist necessary for receptor activation by NMDA. The oocyte was impaled with two glass electrodes (KG-33; Garner Glass, Claremont, CA) pulled with a micropipette puller (P-87; Sutter Instruments, Novato, CA) and filled with 3 M KCl to a resistance of 0.4–1.5 MΩ. Oocytes were voltage-clamped at a holding potential of −60 mV by using a two-electrode voltage clamp amplifier (Axoclamp 2A; Axon Instruments, Foster City, CA). Signals were filtered by using an 8-pole low-pass Bessel filter (Frequency Devices, Haverhill, MA) set at a 40-Hz cutoff before sampling at 100 Hz. The resulting signals were digitized and stored on a Power Macintosh 7100 (Apple Computer, Cupertino, CA) by using data-acquisition software (MacLab; ADInstruments, Milford, MA). Holding currents for oocytes ranged from 0 to 50 nA, and peak agonist currents ranged from 80 to 4000 nA. At least four oocytes were used for each data point, and oocytes from at least 2 frogs were studied for each experiment.
Racemic methadone and morphine were purchased from the University of California-San Francisco pharmacy per regulations. S(+) methadone was purchased from Sigma Chemicals (St. Louis, MO). Both S(+) and R(−) methadone were also obtained from RTI International (Research Triangle Park, NC) after approval was obtained from the National Institute on Drug Abuse. All NMDA receptor clones, NR1, 2A, 2B, 2C, and 2D cDNA sequences were provided Dr. Shigetada Nakamura (Department of Biological Sciences, Kyoto University, Kyoto, Japan).
Agonist response was quantified as the difference between baseline current and peak current during 15 s of superfusion (3–5 mL/min) with 10 μM NMDA inFR. Methadone racemate, its isomers S(+) and R(−) methadone, and morphine were diluted in FR with 10 μM glycine to the required concentrations. Antagonists were preapplied for 15 s before a combined agonist (10 μM NMDA) and antagonist superfusion. A control measurement with 10 μM NMDA in FR agonist application preceded and followed each antagonist application. Extended FR washout periods of 2–10 min followed each postantagonist application to achieve a pre- and postantagonist agonist response within 90% of each other. The mean value of the pre- and postantagonist applications was taken as the average control current. Antagonist response was compared with control current by using the following equation:EQUATION
The concentration-response relations for each drug were fitted by nonlinear regression analysis with Graph Pad Prism software 3.0 for Macintosh (Graph Pad Software, San Diego, CA), which derived the 50% inhibitory concentrations (IC50) and Hill coefficients. Results are represented as mean±sd except where noted. To test for significant differences between inhibition of the four receptor subtypes or between the different antagonists, one-way analysis of variance, followed by Tukey’s test when appropriate, was used with the same software package. P <0.05 was considered significant.
Four subtypes of NMDA receptor channels were expressed in Xenopus oocytes by using a combination of NR1 with each NR2 subunit-specific transcript (NR1/2A, NR1/2B, NR1/2C, and NR1/2D). Examples of NMDA-activated currents in an NR1/2B-expressing oocyte under a 2-electrode voltage clamp (holding potential of −60 mV) are shown in Figure 1. Peak inward currents occurred rapidly on application of agonists. NR1/2B-expressing oocytes were exposed to a range of NMDA concentrations (0.1–100 μM) for 15 s (Fig. 1A). Concentration-dependent currents reached a near-maximum response at 10 μM. At NMDA concentrations larger than 10 μM, very long washout periods were necessary to achieve repeatable responses. Very similar concentration-response effects were also obtained with NMDA on NR1/2A, NR1/2C, and NR1/2D subunit combinations (data not shown). Therefore, in all subsequent experiments, the agonist concentration was set at 10 μM NMDA to produce robust yet repeatable agonist responses.
Channel sensitivities to methadone were then examined by eliciting NMDA currents with and without a coapplication of 10 μM racemic methadone. Similar inhibitory effects were also observed in oocytes expressing other heteromeric NMDA subunit combinations. Reversible inhibition of NMDA-elicited currents was observed in NR1/2B-expressing oocytes (Fig. 1B).
Racemic methadone reversibly inhibited NMDA-induced currents in the four heteromeric NMDA receptor channels with similar concentration-dependent relationships. All subtype IC50 values were in the low micromolar range, and Hill coefficients were between 0.68 and 0.77 (Fig. 2A, Table 1). The NR1/2A and NR1/2B subtypes were statistically more sensitive to racemic methadone than the NR1/2C and NR1/2D subtypes (P < 0.05). In comparison, morphine was a much less potent inhibitor of the same NMDA channel receptor subtypes (Fig. 2B, Table 1). Morphine IC50 values for each heteromeric subunit combination ranged from 26.5 to 159 μM—approximately 8 to 16 times larger than racemic methadone. By analysis of variance, the NR1/2A and NR1/2B subtypes were more sensitive to morphine than the NR1/2C and NR1/2D subtypes (P < 0.01).
To further elucidate the inhibitory properties of methadone on the heteromeric NMDA channels, we studied whether methadone inhibits by a competitive or noncompetitive mechanism. Increasing concentrations of NMDA were not able to overcome the inhibition produced by an IC50 concentration of racemic methadone on the NR1/2A subunit combination (Fig. 3). In addition, the 50% effective concentration values for NMDA in the presence or absence of methadone were not different. From this we confirmed in our functional assay the previously reported noncompetitive inhibition reported in binding assays (5).
We also tested the ability of the R(−) and S(+) methadone stereoisomers to inhibit NMDA currents. The observed currents displayed reversible dose-dependent inhibition. Sensitivities to both R(−) methadone and S(+) methadone were similar in potency to racemic methadone on most subunit combinations with IC50 values in the low micromolar range (Fig. 4, Table 1). R(−) methadone was more potent on the NR1/2A and NR1/2B subtypes than on the NR1/2C and NR1/2D subtypes (P < 0.01), whose IC50 values were more than 150 times larger than those of NR1/2A. The NR1/2A subtype showed strong stereoselectivity, with an S/R IC50 ratio of 62.7, which differed significantly from the ratios of NR1/2B, NR1/2C, and NR1/2D subtypes (0.91, 0.42, and 0.65, respectively). The only significant difference in S(+) methadone inhibition of the NMDA subtypes was observed between the NR1/2B and NR1/2D channels (P < 0.01).
In this study, we examined the actions of morphine, racemic methadone, and the methadone stereoisomers on NMDA receptors. Specifically, we quantified the ability of these drugs to inhibit the electrophysiologic function of various NMDA receptor subtypes and found that racemic methadone is 8–16 times more potent than morphine in producing this effect. For each NMDA subtype, morphine is significantly less potent than methadone and its stereoisomers.
These results position methadone as the most potent opioid for inhibiting NMDA receptor function. Previously, Yamakura et al. (17) found that meperidine, fentanyl, morphine, codeine, and naloxone inhibited NMDA receptor function, with IC50 values in the hundreds of micromolar concentration range. The concentrations at which methadone produces NMDA receptor inhibition overlap with the established clinical range for this drug. Gourlay et al. (18) determined that serum levels of approximately 300 nM for methadone and 120 nM for morphine produced significant analgesia in 18 cancer pain patients. Steady-state levels in patients chronically taking methadone can exceed 1 μM, with peak levels more than 3 μM (19). Methadone appears to be efficiently transported across the blood-brain barrier, with cerebrospinal fluid concentrations in methadone-maintenance patients as large as 73% of their serum concentrations (20). From our studies, these concentrations of methadone could block up to 50% of some NMDA receptor subtypes, a significant fraction of excitatory glutamatergic neurotransmission. In contrast, morphine and other previously studied opioids would appear to have a negligible effect on NMDA receptors within their clinical concentration ranges.
The ability of methadone to activate μ-opioid receptors and at the same time inhibit the activity of NMDA receptors suggests a particular utility of methadone in treating chronic pain. NMDA receptor antagonists diminish the glutamatergic transmission found in hypersensitivity states associated with neuropathic pain (21). Methadone is currently accepted as a second-line opioid for severe cancer pain (18) and has been reported effective for treatment of chronic pain after burn injury (22). Methadone, then, unlike other conventional opioids, appears capable of acting by both opioid and glutamatergic mechanisms.
This dual quality of methadone may also indicate an intrinsic ability to limit its own propensity to induce tolerance. Much recent evidence supports a role for NMDA receptors in the development of opioid tolerance. In fact, the noncompetitive NMDA antagonists MK-801, ketamine, dextromethorphan, and phencyclidine prevent the development of opioid tolerance in rats (23). These two proposed effects—inhibition of hyperactive glutamatergic pathways in chronic pain and an intrinsic self-limitation of tolerance—provide a further rationale for an increased clinical role for methadone in the treatment of chronic pain.
In addition, one of the stereoisomers, R(−) methadone, showed very high potency for one NMDA receptor subtype (NR1/2A), displaying a S/R IC50 ratio >60. In comparison, S-ketamine is only three times more potent than R-ketamine as an anesthetic (24), which parallels the stereoselectivity found at the molecular level with NMDA receptor subtypes (25). This finding contradicts, to some extent, previously reported receptor binding studies that showed both stereoisomers only weakly displacing MK-801 from preparations (26). However, this previous study examined methadone binding to rat cortical and spinal cord homogenates, purifying all subtypes of NMDA receptors together, whereas we expressed each subtype individually. Thus, we were able to discern subtype-specific stereoselectivity.
The stereoselectivity of R(−) methadone on the NR1/2A subtype may suggest a role for this stereoisomer in clinical practice. R(−) methadone has already been shown to block the development of morphine tolerance in rodents (13). The importance of the NR1/2A subtype in chronic pain mechanisms or in opioid tolerance has not been established, but this subunit combination is probably the most widely expressed form of NMDA receptor in the CNS (2) and may be selectively targeted for inhibition by R(−) methadone.
The principal limitation of this study is that we examined only a fraction of the NMDA receptor subunit combinations that are likely expressed in native tissues. In addition to other splice variants of the NR1 subunit, more than one NR2 subunit may coassemble to produce functional NMDA receptors. Future studies will examine whether functional receptors made from such pairings display equivalent stereoselective methadone sensitivity as we have found for the NR1/2A subtype.
In summary, we have identified noncompetitive NMDA receptor blocking capability in the strong opioid methadone within the clinically relevant range. The noncompetitive nature of this inhibition is similar to that found for drugs such as ketamine and phencyclidine, which have strong analgesic properties. These findings suggest a unique efficacy of methadone, in addition to its favorable pharmacokinetic profile, in treating chronic pain syndromes.
The authors thank Dr. Shigetada Nakamura, Department of Biological Sciences, Kyoto University, Kyoto, Japan, for providing the rat NMDA receptor clones and Alexander Reicher for editorial assistance.
1. Foley KM. Opioids and chronic neuropathic pain. N Engl J Med 2003; 348: 1279–81.
2. Dingledine R, Borges K, Bowie D, Traynelis SF. The glutamate receptor ion channels. Pharmacol Rev 1999; 51: 7–61.
3. Akazawa C, Shigemoto R, Bessho Y, et al. Differential expression of five N
-methyl- d -aspartate receptor subunit mRNAs in the cerebellum of developing and adult rats. J Comp Neurol 1994; 347: 150–60.
4. Monyer H, Burnashev N, Laurie DJ, et al. Developmental and regional expression in the rat brain and functional properties of four NMDA receptors. Neuron 1994; 12: 529–40.
5. Ebert B, Thorkildsen C, Andersen S, et al. Opioid analgesics as noncompetitive N
-methyl- d -aspartate (NMDA) antagonists. Biochem Pharmacol 1998; 56: 553–9.
6. South SM, Kohno T, Kaspar BK, et al. A conditional deletion of the NR1 subunit of the NMDA receptor in adult spinal cord dorsal horn reduces NMDA currents and injury-induced pain. J Neurosci 2003; 23: 5031–40.
7. Wong CS, Cherng CH, Luk HN, et al. Effects of NMDA receptor antagonists on inhibition of morphine tolerance in rats: binding at mu-opioid receptors. Eur J Pharmacol 1996; 297: 27–33.
8. Trujillo KA. Are NMDA receptors involved in opiate-induced neural and behavioral plasticity? A review of preclinical studies. Psychopharmacology 2000; 151: 121–41.
9. Eilers H, Philip LA, Bickler PE, et al. The reversal of fentanyl-induced tolerance by administration of “small-dose” ketamine. Anesth Analg 2001; 93: 213–4.
10. Trujillo KA. Effects of noncompetitive N
-methyl- d -aspartate receptor antagonists on opiate tolerance and physical dependence. Neuropsychopharmacology 1995; 13: 301–7.
11. Baker AK, Hoffmann VL, Meert TF. Dextromethorphan and ketamine potentiate the antinociceptive effects of mu- but not delta- or kappa-opioid agonists in a mouse model of acute pain. Pharmacol Biochem Behav 2002; 74: 73–86.
12. Weinbroum AA, Rudick V, Paret G, Ben-Abraham R. The role of dextromethorphan in pain control. Can J Anaesth 2000; 47: 585–96.
13. Davis AM, Inturrisi CE. d-Methadone blocks morphine tolerance and N
-methyl- d -aspartate-induced hyperalgesia. J Pharmacol Exp Ther 1999; 289: 1048–53.
14. Chizh BA, Schlutz H, Scheede M, Englberger W. The N
-methyl- d -aspartate antagonistic and opioid components of d-methadone antinociception in the rat spinal cord. Neurosci Lett 2000; 296: 117–20.
15. Stringer M, Makin MK, Miles J, Morley JS. d-Morphine, but not l-morphine, has low micromolar affinity for the non-competitive N
-methyl- d -aspartate site in rat forebrain: possible clinical implications for the management of neuropathic pain. Neurosci Lett 2000; 295: 21–4.
16. Gray AT, Zhao BB, Kindler CH, et al. Volatile anesthetics activate the human tandem pore domain baseline K+
channel KCNK5. Anesthesiology 2000; 92: 1722–30.
17. Yamakura T, Sakimura K, Shimoji K. Direct inhibition of the N
-methyl- d -aspartate receptor channel by high concentrations of opioids. Anesthesiology 1999; 91: 1053–63.
18. Gourlay GK, Cherry DA, Cousins MJ. A comparative study of the efficacy and pharmacokinetics of oral methadone and morphine in the treatment of severe pain in patients with cancer. Pain 1986; 25: 297–312.
19. de Vos JW, Geerlings PJ, van den Brink W, et al. Pharmacokinetics of methadone and its primary metabolite in 20 opiate addicts. Eur J Clin Pharmacol 1995; 48: 361–6.
20. Rubenstein RB, Kreek MJ, Mbawa N, et al. Human spinal fluid methadone levels. Drug Alcohol Depend 1978; 3: 103–6.
21. Sang CN. NMDA-receptor antagonists in neuropathic pain: experimental methods to clinical trials. J Pain Symptom Manage 2000; 19: S21–5.
22. Altier N, Dion D, Boulanger A, Choiniere M. Successful use of methadone in the treatment of chronic neuropathic pain arising from burn injuries: a case-study. Burns 2001; 27: 771–5.
23. Trujillo KA, Akil H. Inhibition of opiate tolerance by non-competitive N
-methyl- d -aspartate receptor antagonists. Brain Res 1994; 633: 178–88.
24. Geisslinger G, Hering W, Thomann P, et al. Pharmacokinetics and pharmacodynamics of ketamine enantiomers in surgical patients using a stereoselective analytical method. Br J Anaesth 1993; 70: 666–71.
25. Yamakura T, Sakimura K, Shimoji K. The stereoselective effects of ketamine isomers on heteromeric N
-methyl- d -aspartate receptor channels. Anesth Analg 2000; 91: 225–9.
26. Gorman AL, Elliott KJ, Inturrisi CE. The d- and l-isomers of methadone bind to the non-competitive site on the N
-methyl- d -aspartate (NMDA) receptor in rat forebrain and spinal cord. Neurosci Lett 1997; 223: 5–8.