Effect of deuteration on the single dose pharmacokinetic properties and postoperative analgesic activity of methadone

Abstract

Although methadone is effective in the management of acute pain, the complexity of its absorption-distribution-metabolism-excretion profile limits its use as an opioid of choice for perioperative analgesia. Because deuteration is known to improve the pharmacokinetic, pharmacodynamic and toxicological properties of some drugs, here we characterized the single dose pharmacokinetic properties and post-operative analgesic efficacy of d₉-methadone.

The pharmacokinetic profiles of d₉-methadone and methadone administered intravenously to CD-1 male mice revealed that deuteration lead to a 5.7- and 4.4-fold increase in the area under the time-concentration curve and maximum concentration in plasma, respectively, as well as reduction in clearance (0.9 ± 0.3 L/h/kg vs 4.7 ± 0.8 L/h/kg). The lower brain-to-plasma ratio of d₉-methadone compared to that of methadone (0.35 ± 0.12 vs 2.05 ± 0.62) suggested that deuteration decreases the transfer of the drug across the blood-brain barrier. The estimated LD₅₀ value for a single intravenous dose of d₉-methadone was 2.1-fold higher than that for methadone. Moreover, d₉-methadone outperformed methadone in the efficacy against postoperative pain by primarily activating peripheral opioid receptors. Collectively, these data suggest that the replacement of three hydrogen atoms in three methyl groups of methadone improved its pharmacokinetic properties, improved safety, and enhanced its analgesic efficacy.

1. Introduction

Methadone is a long-lasting synthetic opioid used as a maintenance therapy to treat opioid addiction and as a pain reliever. Recently, Kreutzwiser and Tawfic provided a comprehensive review article focusing on the use of methadone in the management of acute pain and chronic non-cancer and chronic cancer pain [1]. Accordingly, the existing data do not support the use of methadone as a first-line aid in non-cancer and cancer pain management [1]. On the other hand, several studies demonstrated the efficacy and safety of methadone in the management of acute pain during intra- and post-operative periods [2–6]. However, despite existing evidence that methadone decreases pain after surgery, the use of methadone as an opioid of choice for postoperative pain management is hindered by the complex absorption-distribution-metabolism-excretion profile of the drug [1].

The main challenges in the therapeutic use of methadone are associated with considerable inter-individual variation in its effect and toxicity. High inter-patient variability in the plasma concentration of methadone among those taking the same dose of the drug [7,8] is attributed mainly to variability in the rates of its hepatic and intestinal biotransformation [9–11]. Methadone undergoes sequential N-demethylation into its inactive derivatives 2-ethylidene-1,5-dimethyl-3,3,-diphenylpyrrolidine (EDDP) and 2-ethyl-5-methyl-3,3-diphenyl-1-pyrroline (EMDP) by several cytochrome P450 (CYP) enzymes, of which hepatic CYP3A4, CYP2B6, CYP2D6 and CYP2C19 are the major contributors [12]. The activity of these enzymes varies greatly among individuals due to genetic and environmental factors. While several reports showed correlation between functional genetic variants of methadone-metabolizing CYPs and the concentrations of methadone in plasma [12,13], it was later established that CYP2B6, a highly polymorphic CYP isoform, is the main determinant of clinical disposition of methadone in humans. Thus, a formulation of methadone that improves its pharmacokinetic (PK) and pharmacodynamic (PD) properties could potentially improve the clinical utility of this drug.

Deuterium is a naturally occurring, stable, non-radioactive hydrogen isotope. In recent years, the use of deuterated compounds in the clinical development of medications has become a fast-growing field on the premise of the proven safety of deuterium [14–16] and the ability of deuteration to improve the PK properties of drugs [17–25]. Previously, Elison et al. reported that the substitution of deuterium for hydrogen in the N-CH₃ group of morphine resulted in a reduction in the rate of oxidative N-demethylation of d₃-morphine [26]. On the other hand, Hsia et al. reported similar rates of absorption, distribution, and excretion of d₃-methadone and methadone in rats, suggesting that the replacement of hydrogen atoms at C-1 with deuterium atoms in methadone did not affect its PK and PD properties [27]. These examples illustrate the differential effects of the incorporation of deuterium on the PK of drugs.

Therefore, the aims of this study were to evaluate whether the replacement of three hydrogen atoms in each of three methyl groups of methadone (Figure 1) affects the pharmacokinetic properties and organ distribution of d₉-methadone following its single intravenous administration as well as its safety profile and analgesic potency.

Figure 1.

Chemical structures of d₉-methadone and its metabolite d₆-EDDP.

EDDP, 2-ethylidene-1,5-dimethyl-3,3,-diphenylpyrrolidine.

2. Materials and Methods

2.1. Chemicals

Protein concentrations were measured using Bradford protein assay reagent (Bio-Rad Laboratories, Hercules, CA, USA) with bovine serum albumin as a standard. (±)-Methadone hydrochloride and (±)-d₉-methadone (1 mg/ml solution in methanol) were purchased from Sigma-Aldrich (St. Louis, MO, USA). All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise mentioned.

2.2. Animals

All experimental procedures were approved by the Animal Care and Use Committee at the University of Texas Medical Branch (UTMB) at Galveston and were in accordance with guidelines published by the US National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85–23, revised 1996). Hsd:ICR (CD-1) mice were purchased from Envigo (Indianapolis, IN) and C57BL/6N mice (7–10 weeks, both sexes) were purchased from Charles River Laboratories (Houston, TX). All animals were housed in temperature-controlled rooms (23°C) with a 12:12 hour light/dark cycle and with unlimited access to food and water in an American Association for Accreditation of Laboratory Animal Care (AAALAC)-accredited animal facility.

2.3. In vivo toxicity of d₉-methadone vs methadone

The toxicity of d₉-methadone and methadone administered intravenously (i.v.) via tail vein was assessed in CD-1 male mice using the Up-and-Down method [30]. Following the administration of the drug, each animal was observed for 20 minutes and scored using the Clinical Scores scale (1–5) as described in detail in our recent report [31]. We first evaluated the toxicity of methadone, for which the starting dose administered to the first animal was selected using a constant multiplicative factor of 1.2 and was five steps below 41 mg/kg; the latter value represented the LD₅₀ of methadone which was administered intraperitoneally (i.p.) to Swiss mice [27]. The dose for the next animal was increased or decreased by a factor of 1.2 depending on whether the animal survived or died, respectively. The dosing was stopped when five reversals occurred in six animals tested consecutively. The estimated LD₅₀ value was calculated using the maximum likelihood method imbedded in the AOT425 software v.1.0 (Westat, Inc. and U.S. Environmental Protection Agency, Washington, DC). The outcome was deemed the same for all animals with the Clinical Score of 5 for those sacrificed by CO₂ asphyxiation for humane reasons and for those that had died during the testing. For d₉-methadone, we utilized similar approach, however the starting dose was selected five steps above the LD₅₀ value that we determined for methadone.

2.4. Pharmacokinetic and tissue distribution of d₉-methadone vs methadone

The PK parameters of d₉-methadone and methadone were determined following the i.v. administration in CD-1 male mice. N=3 experiments were conducted for each studied compound. For the PK study, (±)-methadone hydrochloride was diluted in Dulbecco’s phosphate buffered saline (DPBS, cell culture grade, Fischer Scientific, PA, USA) to achieve a final concentration of 0.5 mg/mL. To prepare (±)-d₉-methadone solution, 5 mL of the 1 mg/ml stock in methanol, then the dried residue was reconstituted in 3 ml of 20 mM hydrochloric acid, and then PBS was added to achieve a final concentration of 0.5 mg/ml of d₉-methadone. At the day of each experiment, each mouse (n=9 animals per experiment) received a 2 mg/kg dose of d₉-methadone or methadone i.v. via tail vein. The dosage volume was adjusted based on a body weight of each mouse and was equivalent to 100 μL of the drug solution administered to a mouse weighing 25 g. At predetermined time points (5 min, 15 min, 30 min, 1 h, 2 h, 3 h, 4 h, 6 h, and 8 h), n=1 animal was sacrificed by CO₂ asphyxiation. The blood samples were obtained by cardiac puncture and centrifuged immediately to separate the plasma. The lungs, liver, spleen, heart, and a kidney were resected from each animal and weighed (for paired organs, both lungs were collected and processed together, while only one kidney was harvested from each animal). The plasma samples and the organs were stored at −80°C until analysis.

2.5. Tissue sample preparation

After thawing, the organs were weighed and homogenized with deionized water containing 1% formic acid. The ratios of the organ weight to the volume of deionized water (w/v) in the homogenates were: 1:1 for the brain, 1:3 for the liver, heart, lungs and kidney, and 1:4 for the spleen.

2.6. Sample Analysis

A liquid chromatography–mass spectrometry (LC-MS/MS) method for the quantification of methadone, d₉-methadone, EDDP and d₆-EDDP was developed and validated in our laboratory. The samples were analyzed with a Thermo Scientific^™ TSQ Quantiva^™ Triple Quadrupole mass spectrometer (Waltham, MA). The system was controlled by Xcalibur 3.0 software. Separation of the analytes was achieved by a Waters Symmetry C₁₈ HPLC column at 45°C. The mobile phase was made of (A) acetonitrile and (B) 0.1% formic acid aqueous solution (v/v) with gradient elution of 0–5 min, 25% A to 50% A with a flow rate of 350 μL/min. Before each injection, the HPLC column was washed with 90% A, and then equilibrated with 25% A for 5 min. The following internal standards were utilized: d₉-methadone (for quantitative determination of methadone), methadone (for d₉-methadone), and d₃-EDDP (for EDDP). Due to the unavailability of synthesized d₆-EDDP as a standard compound, d₃-EDDP was used in lieu of d₆-EDDP to construct a calibration curve, with EDDP serving as internal standard. Multiple reaction monitoring (MRM) was set up at m/z 310→265 for methadone, m/z 319→268 for d₉-methadone, m/z 278→249 for EDDP, m/z 284→249 for d₆-EDDP, and m/z 281→249 for d₃-EDDP.

The samples of plasma and tissue collected from the mice were processed as follows: 250 μL of plasma or 200 μL of organ homogenates were spiked with 10 μL of the corresponding internal standard working solution and 10 μL of Li₂CO₃ (10 mg/ml stock), and then the mixture was vortexed for 30 sec. Afterwards, 800 μL of acetonitrile was added and vortexed for 30 sec. The mixture was centrifuged at 12,000 × g for 15 min, and then the supernatant was transferred to a 2-mL centrifuge tube and dried at 50°C under a stream of nitrogen. The residue was reconstituted with 150 μL of initial mobile phase and centrifuged at 12,000 × g for 5 min. 100 μL of the supernatant was transferred to an HPLC injection vial and 10 μL of the sample was injected into the LC-MS/MS system.

The calibration and quality control standards of methadone, d₉-methadone, EDDP and d₃-EDDP (in lieu of d₆-EDDP) were prepared by spiking the appropriate working solution of the standards with the blank plasma or the tissue homogenates, and the samples were processed as described above. The calibration curves for methadone, d_9-methadone, EDDP and d₃-EDDP were found to be linear in the ranges of 3.0–2400 ng/mL in plasma, 5.0–3900 ng/g in brain, 18.0–14000 ng/g in spleen, and 14.0–12000 ng/g in liver, kidney, lung and heart samples. The intra- and inter-day accuracy of the method was within 90% to 114%, and the precision was above 93%. All biological samples were processed and analyzed along with quality control samples at high, medium and low concentration levels.

2.7. Pharmacokinetic analysis

The PK parameters for methadone, d₉-methadone, and their metabolites EDDP and d₆-EDDP were determined using Kinetica software (v. 5.1, Thermo Fisher Scientific, Waltham, MA) [32–33]. The maximum plasma concentration (C_max) and time to C_max (T_max) were determined by visual inspection of the concentration versus time curve for which the data for each experiment utilized multiple mice (one mouse per time point, total of n=3 experiments). The area under the plasma concentration vs. time curve over the 8-hour study period (AUC_0–8h) was determined by non-compartmental analysis, utilizing the trapezoidal method. For methadone and d₉-methadone, the clearance (CL) was calculated as the dose divided by the AUC_0−∞, with the AUC_0−∞ being calculated as the AUC_0–8h plus an extrapolated AUC_8h−∞ predicted as the concentration at 8 h divided by the terminal rate constant fit through the latter time points. The total volume of distribution (V_ss) was calculated as the clearance multiplied by the mean residence time, which is the AUC_0−∞ divided by the area under the moment curve (AUMC_0−∞). The half-life was calculated as ln(2) divided by the terminal rate constant. The AUC_0–8h for the parent drugs and their respective metabolites in each studied organ were derived using time vs. tissue concentration data from multiple mice (one animal per time point, total of n=3 experiments).

2.8. Metabolism of methadone and d₉-methadone in vitro

The activity of mouse liver microsomes (MLM) and human liver microsomes (HLM) in catalyzing the N-demethylation of methadone to EDDP and d₉-methadone to d₆-EDDP was determined following the protocol described previously [28]. The stock solutions of methadone and d₉-methadone were prepared as described above for the PK study, except that potassium phosphate buffer was used instead of PBS. Methadone and d₉-methadone were used in a range of concentrations to construct saturation curves for the formation of their corresponding metabolites as follows: methadone, 7.5–100 μM with MLM and 15–300 μM with HLM; and d₉-methadone, 1.875–60 μM with MLM and 7.5–100 μM with HLM. N=3 experiments were conducted for each drug, with each point in triplicate. Each reaction mixture (total volume of 0.2 mL in 100 mM potassium phosphate buffer, pH 7.4) containing 0.025 mg of pooled HLM (XenoTech, LLC, Kansas City, KS) or 0.05 mg of pooled MLM from male CD-1 mice (XenoTech, LLC, Kansas City, KS) and either methadone or d₉-methadone was pre-incubated for 5 minutes at 37°C. The reaction was initiated by the addition of NADPH-regenerating system (0.4 mM NADP; 4 mM glucose-6-phosphate; 1U/mL glucose-6-phosphate dehydrogenase and 2 mM MgCl₂) and carried out at 37°C. The duration of the reaction was 10 minutes for MLM and 20 minutes for HLM. The reaction was terminated by placement of the tubes on ice and the addition of 20 μL of Li₂CO₃ (10 mg/mL working solution) and 1.2 mL of ice-cold acetonitrile containing 1% formic acid to each tube as well as the respective internal standard (see section 2.6). The samples were processed as described in section 2.6; the final residues were reconstituted with 200 μL of the initial mobile phase and analyzed by means of LC-MS/MS for the quantitative determination of either EDDP or d₆-EDDP as described in section 2.6. For each concentration of methadone and d₉-methadone, the control reactions were conducted in the absence of NADPH-regenerating system, and the background quantities of EDDP and d₆-EDDP were subtracted from the respective EDDP and d₆-EDDP quantities determined in the reactions with NADPH. To minimize the number of freeze/thaw cycles of the liver microsomal proteins, the pooled microsomes were aliquoted and stored at −80°C as recommended by the manufacturer, and individual aliquots were defrosted for each experiment. The mouse and human liver microsomes utilized in this work were from the same respective lots to eliminate the influence of potential donor-associated variability in the activity of the hepatic enzymes. The experimental conditions were optimized using 30 μM of methadone for the MLM and HLM protein amounts and incubation time to assure that the formation of EDDP was within the linear range (assuming that the formation of d₆-EDDP from d₉-methadone was also linear under the same experimental conditions). The apparent Michaelis constant (K_m) and maximum velocity (V_max) values were estimated using SigmaPlot software, version 14.5.0.101 (Systat Software, Inc., Palo Alto, CA) using pooled data from n=3 saturation curves for each of the substrates. The intrinsic metabolic clearance CL_int was calculated as V_max/K_m [29]; and the deuterium effect on the reaction was evaluated based on a ratio of V_max and CL_int parameters for the non-deuterated substrate to those of the deuterated substrate.

2.9. Animal models of pain

To model postoperative pain in mice, plantar incision was performed under isoflurane anesthesia. Specifically, a 4–8 mm incision was made through the skin, fascia, and muscle of the plantar side of one hind paw, longitudinally starting from the proximal edge of the heel. The skin was apposed with sutures (Fine Science Tools, Foster City, CA). The suture was removed at the end of postoperative day 2.

2.10. Behavioral testing

Mechanical pain sensitivity after plantar incision was assessed by measuring paw withdrawal behaviors in response to von Frey filament (VFF) probing. A mouse was placed in a plastic chamber on top of a mesh screen platform. The VFF of 0.98 mN was applied perpendicularly to the hind paw. At baseline (i.e., before surgery), mice rarely withdraw their hind paw from this VFF probing, whereas after surgery, the operated hind paw develops nociceptive hypersensitivity manifested as the increased number of withdrawals from ten trials of probing [34–35]. The percent of withdrawal responses out of the ten trials was recorded.

Methadone (3 mg/kg) or d₉-methadone (3 mg/kg) was administered i.p. either once before plantar incision (i.e., intraoperative treatment) or repeatedly at 4-hour, 1-day, 2-day, and 3-day time points post-surgery (i.e., postoperative treatments). In a separate experiment that included the postoperative treatment, naloxone methiodide (NLXM, 10 mg/kg) was injected i.p. 30 minutes before the administration of either methadone or d₉-methadone at 1-day post-surgery.

2.11. Statistical analysis

For the metabolism of methadone and d₉-methadone by human liver microsomes in vitro, the K_m and V_max parameters were estimated for each substrate using pooled data from n=3 saturation curves (each data point represented the average of n=3 values from three saturation curves), and were compared descriptively. Likewise, CL_int values were derived from the mean K_m and V_max values and compared descriptively.

The PK parameters for methadone and its metabolite EDDP were compared to the corresponding PK parameters of d₉-methadone and its metabolite d₆-EDDP by the Student’s t-test using Excel 2016 (Microsoft, Redmond, WA). P-values <0.05 were considered significant. The power of analysis was estimated post-hoc using G-power software, version 3.1 [36].

For sample size calculation in the behavioral study, we conducted preliminary experiments using three mice of plantar incision model in each of intra- and post-operative treatment regimens (both sexes at 1:2 ratio). The preliminary intra-operative treatment resulted in the Cohen’s d value of 2.1 between the methadone and d₉-methadone groups in comparing the % of withdrawals measured 4 hours post-incision, indicating that at least five mice per group are needed to achieve statistical power of 0.8 at α=0.05 in this experimental setup (both sexes at 2:3 ratio). In the postoperative treatment experiment comparing % withdrawals before and after the treatment with d₉-methadone (i.e., a paired comparison), Cohen’s d was >20, requiring at least two mice per group for the statistical power of 0.80 at α=0.05. Thus, we used n=3 mice per group for this experimental design (both sexes at 1:2 ratio). We used a generalized linear mixed model to analyze the number of withdrawals out of 10 trials (binomial distribution) repeatedly measured at multiple time points (AR1 covariance structure) with random intercepts for subject (i.e., mouse) variances. Degrees of freedom were allowed to vary across tests by Satterthwaite approximation.

3. Results

3.1. In vivo toxicity of d₉-methadone vs methadone.

The estimated LD₅₀ value for a single intravenous dose of methadone and d₉-methadone were 11.6 mg/kg, 95% CI (11.5, 13.4) and 24.8 mg/kg, 95% CI (20.6, 28.7), respectively.

3.2. Pharmacokinetic and tissue distribution of d₉-methadone vs. methadone.

Figures 2A–D show the plasma concentrations of methadone and d₉-methadone and their corresponding metabolites EDDP and d₆-EDDP as a function of time following i.v. administration of the studied drugs in male CD-1 mice. The plasma concentrations of both methadone and d₉-methadone declined in a biexponential manner; the calculated PK parameters are presented in Table 1. The exposure to d₉-methadone (AUC_0–8h) and the C_max of d₉-methadone exceeded those of methadone by 6 folds (P < 0.05) and 4 folds (P < 0.0001), respectively. The V_ss and CL values of d₉-methadone were reduced by 8 and 5 folds (P < 0.01), respectively, as compared to methadone, while the difference in the elimination half-lives for both drugs did not attain statistical significance. EDDP and d₆-EDDP, the corresponding metabolites of methadone and d₉-methadone, were determined in the plasma of mice over the entire study period Figures 2C & 2D. The C_max of d₆-EDDP in plasma was 5-fold lower than the C_max of EDDP (P < 0.001), while the difference between the T_max values of the metabolites did not attain statistical significance (Table 1). The AUC_0–8h of d₆-EDDP was 2-fold lower than the AUC_0–8h of EDDP (P < 0.01).

Figure 2.

Plasma concentration-time profiles for methadone and d₉-methadone on a linear scale (A) and logarithmic scale (B) and their metabolites EDDP and d₆-EDDP on a linear scale (C) and logarithmic scale (D) after intravenous administration to male CD-1 mice. Each mouse was administered 2 mg/kg of d₉-methadone or methadone intravenously via the tail vein. At the predetermined time points of 5 min, 15 min, 30 min, 1 h, 2 h, 3 h, 4 h, 6 h, and 8 h, animals were sacrificed by CO₂ asphyxiation and blood samples were obtained by cardiac puncture. The quantification of methadone, d₉-methadone, EDDP and d₆-EDDP in plasma samples was performed using an LC-MS/MS method. Data are presented as the mean ± s.d. of 3 separate experiments using 1 animal per time point. N=3 animals were used for each time point. EDDP, 2-ethylidene-1,5-dimethyl-3,3,-diphenylpyrrolidine.

Table 1.

Pharmacokinetic parameters of methadone and d₉-methadone, and their respective metabolites, EDDP and d₆-EDDP, in the plasma of male CD-1 mice after a single intravenous injection of the studied drugs.

Parent Drug Parameters	Methadone	d9-Methadone	P-value
Dose (mg/kg)	2.0	2.0	—
Cmax (ng/mL)	521 ± 55	2270 ± 181	0.0001
Tmax (h)	0.08 ± 0	0.08 ± 0	—
AUC0–8h (ng·h/mL)	414 ± 67	2362 ± 785	0.0128
AUC0−∞ (ng·h/mL)	436 ± 75	2389 ± 817	0.0146
CL (L/h/kg)	4.7 ± 0.8	0.9 ± 0.3	0.0019
Vss (L/kg)	8.9 ± 2.0	1.2 ± 0.0	0.0028
Half-life (h)	2.2 ± 0.8	1.1 ± 0.2	0.0796
Metabolite Parameters	EDDP	d6-EDDP	P-value
Cmax (ng/mL)	233 ± 28	48 ± 7	0.0004
Tmax (h)	0.14 ± 0.10	0.33 ± 0.14	0.1242
AUC0–8h (ng·h/mL)	257 ± 29	138 ± 25	0.0058

Methadone and d₉-methadone, and their respective metabolites, EDDP and d₆-EDDP, were quantified in the plasma samples of male CD-1 mice following a single intravenous injection of 2.0 mg/kg of the parent drug. The plasma concentration-time profiles of the drugs and their metabolites are shown in Figure 2. The PK parameters were determined using Kinetica software (v. 5.1). The data are presented as the mean ± s.d. of three experiments. EDDP, 2-ethylidene-1,5-dimethyl-3,3,-diphenylpyrrolidine.

The ratio of the AUC_0–8h of the metabolite to the AUC_0–8h of the parent drug was lower for d₉-methadone (0.06 ± 0.01) than for methadone (0.62 ± 0.04, P < 0.0001).

Following i.v. administration of the studied drugs, the extent of the exposure of the organs to d₉-methadone ranked as following: lungs > kidneys ≈ liver > spleen > heart > brain, while the rank order for methadone was: kidneys ≈ lungs > spleen > liver > heart > brain. The exposure of the lungs and liver of the mice to d₉-methadone was greater than the exposure to methadone, while the exposure of the heart, kidneys, and spleen to d₉-methadone was lower than that of methadone (Table 2, P < 0.05). For all the results of the PK and tissue distribution experiments that were statistically significant by the standards of this study, the achieved statistical power was above 0.85 at α=0.05.

Table 2.

AUC_0–8h of methadone, d₉-methadone, and their respective metabolites EDDP and d₆-EDDP in organs of male CD-1 mice after a single intravenous injection of 2.0 mg/kg of either methadone or d₉-methadone.

Parent Drug Tissue AUC0–8h (ng·h/g)	Methadone	d9-Methadone	P-value
Brain	877 ± 395	780 ± 73	0.6971
Liver	3272 ± 570	5052 ± 117	0.0061
Lung	7694 ± 460	10799 ± 969	0.0074
Heart	1890 ± 115	1426 ± 230	0.0351
Kidney	7784 ± 469	5110 ± 408	0.0017
Spleen	4841 ± 494	3451 ± 148	0.0095
Metabolite Tissue AUC0–8h (ng·h/g)	EDDP	d6-EDDP	P-value
Brain	8.96 ± 5.26	<LLOQ§	<0.05
Liver	5608 ± 218	4375 ± 513	0.0186
Lung	9672 ± 856	9886 ± 614	0.7433
Heart	2155 ± 144	1243 ± 254	0.0057
Kidney	2815 ± 226	1562 ± 207	0.0021
Spleen	622 ± 45	411 ± 33	0.0028

The data reflect the exposure of individual organs to the parent drug and its metabolite and are presented as the mean ± s.d. of three experiments.

^§The concentration of d₆-EDDP in the brain was below the lower limit of quantitation at every time point for all three animals. AUC, area under the tissue concentration vs. time curve; EDDP, 2-ethylidene-1,5-dimethyl-3,3,-diphenylpyrrolidine; LLOQ, lower limit of quantification.

The ratios of AUC_0–8h in the organs to the AUC_0–8h in the plasma indicated substantial distribution and exposure of the mouse organs to methadone in comparison to reduced distribution ratios for d₉-methadone (Figure 3, P < 0.01 for all organs). Despite similar exposure of the brain to d₉-methadone and methadone (as reflected by similar AUC values), the brain-to-plasma AUC ratio of d₉-methadone was 0.35 ± 0.12 while the brain-to-plasma AUC ratio for methadone was 2.05 ± 0.62 (Figure 3, P < 0.01), suggesting lower proportional transfer of d₉-methadone across the blood-brain barrier (BBB).

Figure 3.

Ratio of AUC_0–8h of methadone or d₉-methadone in different organs to the AUC_0–8h in plasma.

Each mouse was administered 2 mg/kg of d₉-methadone or methadone intravenously via the tail vein. At predetermined time points (5 min, 15 min, 30 min, 1 h, 2 h, 3 h, 4 h, 6 h, and 8 h), n=1 animal per time point was sacrificed by CO₂ asphyxiation, and the tissue samples were collected. The quantification of methadone and d₉-methadone in tissue samples was performed using an LC-MS/MS method. AUC values for methadone and d₉-methadone in each organ were determined using Kinetica software (v. 5.1). Data are presented as the mean + s.d. of 3 separate experiments using 1 animal per time point. N=3 animals were used for each time point.

AUC, area under the concentration vs. time curve. ** P < 0.01; *** P < 0.001.

The extent of the exposure of the organs to d₆-EDDP and EDDP, the respective metabolites of d₉-methadone and methadone, ranked as follows: lungs > liver > kidney ≈ heart > spleen. The exposure of the lungs to d₆-EDDP and EDDP was similar, while the exposure to d₆-EDDP in the liver, heart, kidney and spleen was significantly lower than the exposure to EDDP (Table 2, P<0.05). The concentration of d₆-EDDP in the brain was below the lower limit of quantification (LLOQ).

3.3. Metabolism of methadone and d₉-methadone.

The rates of the formation of EDDP and d₆-EDDP from methadone and d₉-methadone, respectively, were dependent on the substrate concentrations and exhibited typical saturation kinetics (Figure 4A–D). The apparent K_m and V_max values are shown in Table 3. The apparent V_max of the formation of d₆-EDDP from d₉-methadone by MLM was 2.5-fold lower than the V_max of the formation of EDDP from methadone. Meanwhile, the apparent K_m of d₉-methadone was 2.6-fold lower than the K_m of methadone. Similar results were obtained using HLM: the estimated apparent K_m and V_max values for methadone N-demethylation were 2- and 2.5-folds lower than the corresponding values estimated for d₉-methadone N-demethylation (Table 3). Thus, the deuteration effects of the drug’s N-demethylation V_max (^HV_max/^DV_max) and CL_int (^HCL_int/^DCL_int) were 2.5 and 1.3, respectively, in MLM, and 2.5 and 1, respectively, in HLM, and were deemed as isotope effects of low magnitude (<3.6) according to Miwa and Walsh [37].

Figure 4.

N-demethylation of methadone and d₉-methadone into their respective metabolites, EDDP (A,C) and d₆-EDDP (B,D) in vitro.

N-demethylation of methadone and d₉-methadone into their respective metabolites, EDDP and d₆-EDDP were tested in vitro using mouse liver microsomes (MLM) (A and B, respectively) and human liver microsomes (HLM) (C and D, respectively). The substrates were used in the following ranges: methadone, 7.5–100 μM with MLM and 15–300 μM with HLM; and d₉-methadone, 1.875–60 μM with MLM and 7.5–100 μM with HLM. The substrates were incubates with the microsomal proteins at pH 7.4 in the presence of NADPH and in the absence of the cofactor (control reactions) at 37°C for either 10 min (MLM) or 20 min (HLM). The CYP-catalyzed formation of EDDP and d₆-EDDP was calculated by subtracting the metabolite quantities determined in the control reactions from those determined in the reactions with NADPH. The rates of the formation of EDDP and d₆-EDDP were dependent on the substrate concentrations and were fitted into the Michaelis-Menten model using SigmaPlot (version 14.5.0.101 for Windows, Systat Software Inc.). The insets show the corresponding Eadie-Hofstee plots. Data are presented as the mean ± s.d. from n=3 experiments, with each point in triplicate. The estimated kinetic parameters of the reactions are shown in Table 1. [V], reaction velocity, pmol/min/mg.protein; [S], substrate concentration, μM; EDDP, 2-ethylidene-1,5-dimethyl-3,3,-diphenylpyrrolidine; MLM, mouse liver microsomes; HLM, human liver microsomes.

Table 3.

Kinetic parameters of the formation of EDDP and d₆-EDDP from methadone and d₉-methadone, respectively, by mouse and human liver microsomes.

Subcellular fraction	Substrate	Km (μM)		Vmax (pmol·min−1·mg of protein−1)		CLint (Vmax/Km)
		Best fit value	95% CI	Best fit value	95% CI
MLM	Methadone	53.5 ± 5.1	(42.3, 64.6)	2631 ± 121	(2368, 2893)	49.2
	d 9 -Methadone	20.7 ± 2.1	(16.2, 25.2)	1033 ± 45	(937, 1129)	49.7
HLM	Methadone	82.9 ± 3.7	(74.7, 91.0)	1116 ± 17	(1080, 1152)	13.5
	d 9 -Methadone	41.3 ± 5.9	(28.4, 54.1)	438 ± 27	(379, 497)	10.6

The effect of the range of methadone and d₉-methadone concentrations on the respective formation of EDDP and d₆-EDDP was examined in mouse liver microsomes (MLM) and human liver microsomes (HLM) (Figure 4). The kinetic parameters were estimated by fitting the data in the Michaelis-Menten model using SigmaPlot software (version 14.5.0.101, Systat Software Inc., Palo Alto, CA). Data are presented as the mean ± S.E. of n=3 experiments.

EDDP, 2-ethylidene-1,5-dimethyl-3,3,-diphenylpyrrolidine; HLM, human liver microsomes; MLM, mouse liver microsomes; K_m, Michaelis constant; V_max, maximum velocity; CI, confidence interval; CL_int, intrinsic metabolic clearance.

3.4. Analgesic efficacy of d₉-methadone against postoperative pain in mouse models.

Methadone has been shown to decrease postoperative pain with either intra- or post-operative treatment regimens [38–40]. As shown in Figure 5, mice treated with 3 mg/kg of d₉-methadone demonstrated lower postoperative mechanical hypersensitivity on the day of surgery than methadone-treated mice (t(17)=3.37, P=0.004 at 2 hr; t(17)=2.33, P=0.033 at 4 hr). On the next day, however, there was no difference in the degree of postoperative mechanical hypersensitivity between the two groups (t(17)=0.33, P=0.75).

Figure 5.

The magnitude of postoperative mechanical pain hypersensitivity after intra-operative treatment with methadone or d₉-methadone.

Using an intra-operative regimen, we tested if d₉-methadone produces greater relief of postoperative pain than methadone. Under anesthesia, immediately before plantar incision, mice received either methadone or d₉-methadone (3 mg/kg, i.p.) as an intra-operative treatment. Data are presented as the mean ± s.d. of 3 experiments. N=3 animals were used for each time point.

* P < 0.05; ** P < 0.01.

Methadone given postoperatively at a dose of 3 mg/kg was unable to reduce the postoperative mechanical hypersensitivity (Figure 6A). On the other hand, the same dose of d₉-methadone alleviated hypersensitivity after each systemic injection once daily for 4 days (Figure 6B: F(3,22)=18.1, P<0.001 between times post-injection). Of note, no development of the apparent analgesic tolerance was observed in the ‘once daily for 4 days’ treatment regimen applied in this study (F(3,20)=0.335, P=0.8 between treatment days).

Figure 6.

The magnitude of post-operative mechanical pain hypersensitivity after post-operative treatment.

Using a post-operative regimen, we tested if d₉-methadone produces greater relief of postoperative pain than methadone. After plantar incision, mice received 3 mg/kg, i.p., of either (A) methadone or (B) d₉-methadone as a post-operative treatment. (C) Effect of naloxone methiodide (NLXM), a peripherally restricted opioid receptor antagonist, on the magnitude of post-operative mechanical pain hypersensitivity produced by d₉-methadone. NLXM was administered postoperatively and 30 minutes prior to d₉-methadone injection. NLXM completely blocked the effect of d₉-methadone Data are presented as the mean ± s.d. of 3 experiments. N=3 animals were used for each time point.

As methadone’s analgesic effect is stringently mediated by opioid receptors in the periphery [41], we next investigated if the relief of postoperative pain by d₉-methadone would be similarly mediated by peripheral opioid receptors. Naloxone methiodide (NLXM), a peripherally restricted opioid receptor antagonist, was administered to the operated animals 30 minutes before the treatment with d₉-methadone. As a result, NLXM completely blocked the effect of d₉-methadone (Figure 6C).

4. Discussion

It is estimated that postoperative pain occurs in 20 to 40% of surgical procedures, and pain following surgery remains one of the leading adverse outcomes that impacts postoperative recovery. Unfortunately, the use of repeated i.v. doses (i.e., i.v. boluses) of short-acting opioids such as morphine, oxycodone, and fentanyl result in fluctuating blood concentrations of opioid analgesics and, consequently, relatively brief periods of adequate pain relief. An alternative approach to the postoperative use of short-acting opioids is the use of a single-dose of a long-acting opioid such as methadone. However, the extensive first-pass metabolism of methadone as well as high inter-individual variability in its biotransformation (specifically, its N-demethylation) [42–45] diminish the suitability of methadone in the management of postoperative pain. Therefore, a formulation of methadone that decreases N-demethylation of methadone could potentially improve the clinical utility of this drug.

Deuterium modification has a potential to alter the pharmacokinetic parameters of a compound. A successfully designed deuterated compound may demonstrate an improvement over the parent (nondeuterated) compound by exhibiting increased AUC, increased C_max and half-life, as well as reduced [46]. Therefore, the first aim of this study was to determine the effect of deuteration on the PK parameters and organ distribution of d₉-methadone following its single i.v. administration to male CD-1 mice in comparison to nondeuterated methadone. The data obtained in this investigation revealed that the replacement of three hydrogen atoms in three methyl groups of methadone affected its PK properties. The AUC_0–8h and C_max of d₉-methadone were greater than the AUC_0–8h and C_max of methadone, and the clearance of d₉-methadone was lower than that of methadone. The decreased V_ss for d₉-methadone suggested that this deuterated analog of methadone remains mainly in the plasma, which is in agreement with its increased C_max and AUC_0–8h values in the plasma. Furthermore, the AUC_0–8h value determined for the metabolite, d₆-EDDP, demonstrated that deuteration of methadone also decreased the formation of d₆-EDDP and consequently resulted in a decreased metabolite-to-parent drug ratio. The observed reduction in the clearance of d₉-methadone can in part be explained by the reduced metabolic rate of the hepatic N-demethylation of the deuterated compound. Interestingly, the reduced clearance of d₉-methadone was not accompanied by an increase in half-life. The reduced clearance accompanied by the reduced volume of distribution for d₉-methadone may be explained by the differences in the extent of metabolism, differences in permeability across biological membranes, or by differences in binding to plasma proteins, which can all influence the extravascular/tissue distribution of the drug. An increase in plasma protein binding for d₉-methadone compared to methadone would be expected to result in a decrease in the volume of distribution for d₉-methadone. However, since changes in the unbound fraction in plasma do not always reflect the changes in the unbound fraction in tissue, a side-by-side evaluation of plasma protein binding of d₉-methadone and methadone along with tissue binding would yield comparison of the unbound PK parameters and will be addressed in a future investigation.

The PK study revealed that the levels of d₉-methadone and methadone in the brain were similar; however, the brain-to-plasma ratio for d₉-methadone was 5.8-fold lower than the brain-to-plasma ratio of methadone. We determined that in mice the brain exposure to methadone exceeded the systemic exposure by 2-fold (brain-to-plasma ratio was 2.05), which was in agreement with previous studies reporting the brain-to-plasma ratio of 2.7 in rats [47] and brain-to-plasma ratios of 2.85 [48] and 2.3 [49] in humans determined postmortem. On the other hand, the AUC of d₉-methadone in the brain of mice was lower than that in the plasma (brain-to-plasma ratio of 0.35), suggesting that deuteration decreased the permeability of methadone across the BBB. In principle, the polarity of deuterated compounds increases with the extent of deuterium-for-hydrogen substitution in the C-H bond [50]. Thus, decreased permeability of d₉-methadone across biological membranes including the BBB would be expected. On the other hand, it is well known that P-glycoprotein (P-gp, ABCB1) limits the delivery of clinically used racemic methadone across the BBB [51]. Thus, it appears that the effect of deuteration on the lipophilicity and on the interaction of d₉-methadone with P-gp are the likely reasons of the lower brain-to-plasma ratio of d₉-methadone compared to that of methadone. To the best of our knowledge, there are no reports describing how incorporation of deuterium affects the permeability of drugs across biological membranes containing P-gp; thus, interaction of d₉-methadone with P-gp should be further investigated.

Similar to the brain-to-plasma AUC ratio, the organ-to-plasma AUC ratios in the liver, lung, heart, kidney, and spleen for d₉-methadone were also reduced several fold as compared to those of methadone (Figure 3), further suggesting the deuteration-associated changes in the proportion of the unbound and bound fractions of d₉-methadone. In addition, the effect of deuteration on the interaction of d₉-methadone with membrane efflux transporters of the organs could not be ruled out. The reduced exposure of the kidneys, spleen, and heart to d₉-methadone as determined by the respective AUCs (Table 2) suggest decreased potential for d₉-methadone to induce cardio- [52] and nephrotoxicity [53], which represent known adverse effects associated with the use of methadone. On the other hand, our study revealed that the exposure of the liver and lungs to d₉-methadone was higher than to non-deuterated methadone. Since the treatment with methadone has not been linked to elevated serum enzymes or idiosyncratic acute liver injury [54], the administration of d₉-methadone would not be expected to lead to hepatotoxicity.

As an opioid drug, methadone administered in excessive amounts can induce respiratory depression, which constitutes the major cause of death due to drug overdose in people taking prescription or illicit opioids [55]. Similar to humans, high doses of methadone can lead to respiratory depression in mice as well [55].

In this study, despite similar exposure of the murine brain to d₉-methadone and methadone (Table 2), the estimated LD₅₀ value for a single i.v. dose of d₉-methadone was 2.1-fold higher than the estimated LD₅₀ value for a single i.v. dose of methadone. Despite a multitude of pathophysiological studies of opioid-induced respiratory depression, Baldo et al. highlighted that the underlying mechanisms leading to the fatalities associated with opioid-induced respiratory depression are still not fully understood at this time [56]. Recent studies suggest that multiple sites of the brain can independently exert a depressive effect on breathing [56, 57], and the effect of d₉-methadone on the μ opioid receptors of the inspiratory-generating regions of the brainstem should be investigated. Nevertheless, the comprehensive dose-dependent toxicokinetic assessment of d₉-methadone is warranted. With respect to the main metabolite of d₉-methadone, d₆-EDDP, the exposure to d₆-EDDP in the liver, heart, kidney and spleen was lower, while in the lungs it was similar to the pharmacologically inactive metabolite of methadone, EDDP.

The second aim of this study was to determine the effect of deuteration on the analgesic potency of d₉-methadone in a model of postoperative pain. Because methadone’s analgesic effect is not mediated by central opioid receptors in the mouse [41] (in other words, the effect is strictly dependent on peripheral opioid receptors as is that of d₉-methadone [Figure 6C]), it is expected that d₉-methadone will be more potent in relieving pain than its nondeuterated counterpart by yielding a higher plasma concentration than methadone when administered at the same dose. The present study clearly supports this concept by showing that d₉-methadone significantly relieved postoperative mechanical hypersensitivity in mice at a dose at which methadone was completely ineffective. Conversely, our results indicate that a greater dose of methadone would have been needed to achieve a similar degree of pain relief, elevating the concentration of the drug in the brain as well, which would consequently pose an increased risk for the central nervous system (CNS)-related side effects. And vice versa, lower doses of d₉-methadone would be required to achieve a similar degree of pain relief caused by methadone. This could further lower the concentrations of deuterated drug in the brain and the organs included in the present study, thus decreasing the risk of possible adverse effects associated with the drug’s administration. Therefore, from the standpoint of developing effective and safe analgesics, our study demonstrated that the deuteration of methadone is a promising pharmacological approach for exploiting the proven involvement of the activated peripheral opioid receptors in the analgesic efficacy of systemically administered methadone. Meanwhile, the deuteration of methadone could reduce the undesired/negative consequences associated with the drug’s activation of opioid receptors in the brain.

Finally, we evaluated whether the substitution of three hydrogen atoms in three methyl groups of methadone would result in a reduced rate of the drug’s oxidative N-demethylation by mouse and human hepatic enzymes in vitro using MLM and HLM preparations. In principle, deuterium-for-hydrogen substitution in a C-H bond should increase the metabolic stability of a compound by decreasing the rate of its oxidative metabolism (C-D bonds are shorter and more stable to oxidative processes) [46]. Accordingly, our data demonstrated a 2.5-fold decrease in the rate of d₉-methadone N-demethylation in vitro as compared to methadone in both MLM and HLM. The lower ratio of AUC_metabolite to AUC_{parent drug} obtained for d₉-methadone methadone in mice (0.06 ± 0.01 vs 0.62 ± 0.04, P < 0.0001) suggest the suitability of extrapolation to humans. The apparent K_m of d₉-methadone was 2- and 2.5-fold lower than that of methadone in humans and mice, respectively, suggesting higher affinity of d₉-methadone to the hepatic CYP enzymes. The in vitro CL_int of d₉-methadone was 1.3-fold lower than the in vitro CL_int of methadone, suggesting a rather modest effect of deuteration on the drug’s N-demethylation in vitro by human hepatic CYPs, while the N-demethylation of methadone by murine hepatic CYPs was not affected by the drug’s deuteration. It should be noted that the in vitro CL_int values in the current study were obtained based on the formation of EDDP and might not reflect changes in the total CL_int of the drug due to deuteration, presenting a limitation of the current study. While N-demethylation by hepatic CYP enzymes is the predominant mechanism of hepatic clearance of methadone in humans [58], the effect of deuteration on other metabolic pathways of the drug (and, consequently, the drug total CL_int) could not be ruled out and will be addressed in a future investigation.

In summary, the data obtained in this investigation showed for the first time that deuterium replacement of three hydrogen atoms in three methyl groups altered the pharmacokinetic properties of methadone, improved its safety, and enhanced its analgesic effect in mice.

Abbreviations:

ADME

absorption-distribution-metabolism-excretion

AUC

area under the curve

BBB

blood-brain barrier

CL

clearance

V_ss

total volume of distribution

C_max

maximum plasma concentration

CYP

cytochrome P450 enzymes

EDDP

2-ethylidene-1,5-dimethyl-3,3,-diphenylpyrrolidine

EMDP

2-ethyl-5-methyl-3,3-diphenyl-1-pyrroline

HLM

human liver microsomes

LC-MS/MS

liquid chromatography – mass spectrometry

MLM

mouse liver microsomes

MRM

Multiple reaction monitoring

NLXM

naloxone methiodide

PK

pharmacokinetic

PD

pharmacodynamic

T_max

time of maximum plasma concentration

VFF

von Frey filament