Regular Articles
The Measurement of Meloxicam and Meloxicam Metabolites in Rat Plasma Using a High-Performance Liquid Chromatography-Ultraviolet Spectrophotometry Method
2017 Volume 65 Issue 2 Pages 121-126
Details
2017 Volume 65 Issue 2 Pages 121-126
A high-performance liquid chromatography-ultraviolet spectrophotometry (HPLC-UV) method for the determination of meloxicam (MEL) and meloxicam metabolites (5′-hydroxy meloxicam (5-HMEL) and 5′-carboxy meloxicam (5-CMEL)) has been developed. After extraction of MEL, 5-HMEL, and 5-CMEL from rat plasma using Oasis HLB cartridges, the extracts were separated with a Luna C18 (2) 100 A column (5 µm, 4.6×150 mm, Phenomenex) using a mobile phase of 50 mM phosphate buffer (pH 2.15, solvent A) and acetonitrile (solvent B) at a flow rate of 0.8 mL/min in a linear gradient. The detection wavelength was 360 nm, and the internal standard (IS) was piroxicam. Each calibration curve was linear in the range of 40 to 1000 ng/mL (r2>0.999). The extraction rates of MEL, 5-HMEL, and 5-CMEL were greater than 86.9%. The intra- and inter-day accuracies were in the range of 95.0 to 119.0%, and the precision was 0.2 to 17.0%. To the best of our knowledge, this is the first report of the quantitative and qualitative measurement of meloxicam and each metabolite using an HPLC-UV method.
The aim of this study was to establish a procedure to study the pharmacokinetic (PK) model of meloxicam (MEL). MEL (4-hydroxy-2-methyl-N-(5-methyl-2-thiazolyl)-2H-1,2-benzothiazine-3-carboxamide-1,1-dioxide) is a nonsteroidal anti-inflammatory drug (NSAID) that has selective inhibitory effects on the inducible isoform of cyclooxygenase-2. MEL is as effective as other NSAIDs in the treatment of rheumatoid arthritis and osteoarthritis in humans1); it is used as an anti-inflammatory agent and analgesic in animals.2) Generally, administrated meloxicam is first metabolized to a 5′-hydroxymethyl metabolite (5-HMEL) by CYP2C9 (major) and CYP3A4 (minor); it is then metabolized to a 5′-carboxy metabolite (5-CMEL).3)
Studies of the human plasma concentration/time profile of MEL after a single oral administration indicate that the graph exhibits bimodal peaks.4,5) These reports suggest that MEL undergoes enterohepatic circulation. Taking these findings into account, we are planning to investigate the process by which MEL is metabolized to 5-HMEL and 5-CMEL (e.g., glucuronide conjugate, CYP metabolism). To investigate the PK model of MEL, measurement of the concentrations of MEL and its metabolites (5-HMEL and 5-CMEL) in biological samples is necessary. Several reports have examined the concentration of MEL in human and animal plasma4–9) and the metabolism of MEL.2,3) However, these studies did not concomitantly measure 5-HMEL and 5-CMEL. The development of a system that enables the detection of MEL, 5-HMEL, and 5-CMEL is required.
Various analytical methods for determining MEL in biological samples have been reported. The majority of these methods employ HPLC-ultraviolet (HPLC-UV)2–6) or HPLC-fluorescence (HPLC-FL)10) spectrophotometry. Other reports employed LC-MS,2) liquid chromatography-tandem mass spectrometry (LC-MS/MS),7–9) flow injection analysis-chemiluminescence (FIA-CL),11) or electrochemical methods.12) Simple and rapid sample preparation is an advantage of HPLC. Because MEL exhibits low fluorescence intensity,10) HPLC-UV was judged to be the best method for analyzing MEL and its metabolites. In this paper, a simple HPLC-UV method for the determination of MEL, 5-HMEL, and 5-CMEL is reported. These compounds were successfully measured in the same system.
MEL sodium salt, 5-HMEL (5′-hydroxymethyl 5′-desmethyl meloxicam), and 5-CMEL (5′-carboxy meloxicam) were purchased from Santa Cruz Biotechnology, Inc. (U.S.A.). Acetonitrile, methanol, and distilled water were all of HPLC grade, and the 28% ammonia solution was a JIS guaranteed reagent. These chemicals were obtained from Kanto Chemical Co., Inc. (Japan). Piroxicam (biochemistry grade) was used as the internal standard (IS); dipotassium hydrogenphosphate, potassium dihydrogenphosphate, phosphoric acid, and dimethyl sulfoxide (DMSO) were obtained from Wako Pure Chemical Industries, Ltd. (Japan). The Oasis HLB 1 cc (30 mg) extraction cartridge was purchased from Waters (U.S.A.).
Animals and Drug AdministrationSix male 8-week-old Sprague–Dawley rats (254 to 270 g) were purchased from Sankyo Lab Service Corporation (Japan). A cannula was inserted into the right femoral artery and right jugular vein of each rat by the vendor. The rats were housed in a temperature-controlled room in a 12 h light–dark cycle and were allowed free access to food and water except for the night before administration of MEL until 4 h after administration. The experiment was performed at least 1 d after purchase. The animals underwent inhalation induction of anaesthesia with 5% isoflurane, and anesthesia was maintained with 4% isoflurane. MEL in 5 mM NaOH (2.0 mg/kg) was administered in a single oral dose (n=3). 5-HMEL in 5 mM NaOH (20 mg/kg) was injected into the jugular vein as a single dose through the cannula (n=3). The sampling times were 0, 0.16, 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, and 10 h after oral administration and 0, 0.16, 0.5, 1, 1.5, 2, and 3 h after intravenous administration (n=3). Rat plasma was separated by centrifugation at 4°C and 10000×g for 10 min and stored at −20°C until use. After sampling, the collected blood was replaced with an equal volume of saline. At the end of the experiment, all animals were euthanized by intraperitoneal injection of 1 mL (200 mg/kg) sodium pentobarbital. The animal experiment was approved by the Nihon University Animal Care and Use Committee (Tokyo, Japan).
Preparation of Standard and Working SamplesPrimary stock standard solutions of MEL (0.1 mg/mL), 5-HMEL (0.1 g/mL), and 5-CMEL (0.1 mg/mL) were prepared in 100 µL DMSO and then diluted with 9.9 mL of 0.1 M phosphate buffer (pH 7.4). The IS primary stock standard solution (0.1 mg/mL) was dissolved in methanol and diluted to 500 ng/mL with methanol to form the working solution. The MEL primary stock standard solution was stored at room temperature and shielded from light. The 5-HMEL primary stock standard solution was stored at −20°C. The stock solutions of 5-CMEL and IS were stored at 4°C and were also shielded from light. Each primary stock standard solution was diluted with 0.1 M phosphate buffer (pH 7.4) to obtain a series of working solutions (200, 400, 1000, 2000, 4000, and 10000 ng/mL).
Plasma calibration standards were freshly prepared by spiking 35 µL of blank pooled rat plasma pre-thawed at room temperature with 5 µL working solutions of MEL, 5-HMEL, and 5-CMEL (final concentrations: 20, 40, 100, 200, 400, and 1000 ng/mL). The blank rat plasma was stored at −20°C until use.
Chromatography Conditions and InstrumentsThe HPLC system consisted of a Shimadzu LC-20 A instrument (Japan). MEL, 5-HMEL, 5-CMEL, and IS were analyzed on a Luna C18 (2) 100 A column (5 µm, 4.6×150 mm, Phenomenex) with SecurityGuard cartridges (C18, 4×3.0 mm, Phenomenex), maintained at 35°C. The mobile phase comprised 50 mM phosphate buffer (pH 2.15, solvent A) and acetonitrile (solvent B) at a flow rate of 0.8 mL/min. Isocratic elution (solvent A : solvent B=60 : 40) was used to calculate the selectivity and recovery. The linear gradient program (for linearity, accuracy, and precision) was 0 to 13 min (40 to 60% solvent B), 13 to 13.01 min (60 to 40% solvent B), and 13.01 to 23 min (40% solvent B consistent). The injection volume was 10 µL, and the detection wavelength was set at 360 nm.
Solid Phase Extraction (SPE)A Waters Oasis HLB 1 cc (30 mg) extraction cartridge was used for SPE. The cartridge was conditioned with 1 mL methanol and equilibrated with 1 mL water before loading the sample. Fifty microliters of plasma or the plasma calibration standard, 20 µL of IS working solution, and 180 µL of water were added to the sample tube. After vortex mixing, 250 µL of the mixture was loaded onto the HLB cartridge. The cartridge was rinsed and dried under vacuum using TOMY-ME-100 (2000 rpm, 3 min). Various conditions were examined using several different rinse solutions to produce the best results. The compounds were eluted with 1 mL methanol. The solutions were passed through the cartridge by gravity flow in all steps. The eluate was evaporated to dryness in vacuo with a centrifugal dryer (2000 rpm, 40°C, 45 min). The residue was reconstituted in 50 µL of 0.1 M phosphate buffer (pH 7.4), vortexed for 15 s, and sonicated for 5 min. Ten microliters of the solution was injected into the HPLC system for analysis. The same procedure was used for the method validation. If the value obtained was greater than 1000 ng/mL, the sample was appropriately diluted for HPLC with 0.1 M phosphate buffer (pH 7.4).
Method ValidationThe selectivity, recovery, linearity, accuracy, and precision of the method were validated. The validation was conducted according to the National Institute of Health Sciences (NIHS) guidelines.13) The selectivity was evaluated by examining the separation of MEL, 5-HMEL, 5-CMEL, and IS from the plasma matrix components of blank rat plasma. The recovery was determined for five replicates of each sample at a concentration of 1000 ng/mL. The recoveries were determined by comparing the absolute peak areas of the extracted samples with those of the pre-spiked standards. The calibration curves were constructed by plotting the MEX, HMEL, or CMEL peak area divided by the IS peak area. The linearity of the calibration curves was evaluated by linear regression analysis. The lower limits of quantitation (LLOQs) of MEL, 5-HMEL, and 5-CMEL were experimentally defined as the lowest concentration of the calibration curve that could be measured with acceptable accuracy and precision.
The intra- and inter-day accuracy and precision of this method were investigated using working solutions of MEL, 5-HMEL, and 5-CMEL. Accuracy was expressed as a percentage of the measured concentration relative to the theoretical concentration.
 |
The PK parameters of MEL were estimated by a non-compartmental method. The maximum plasma concentration (Cmax) and the time at which the Cmax was observed (tmax) were used to determine the actual blood sampling times. The plasma concentrations of 5-HMEL after intravenous administration were extrapolated to time zero (C0). The area under the plasma concentration–time curve (AUC) was obtained using the linear trapezoidal rule. The elimination rate constant (ke) was estimated from the least-squares regression slope of the terminal plasma concentration. The AUC from zero to infinity (AUCinf) was calculated as AUCinf=AUC+Clast/ke (Clast is the last plasma concentration measured). The half-life (t1/2) was calculated as ln 2/ke.
The UV absorptions of MEL, 5-HMEL, 5-CMEL, and IS were measured with a JASCO UV-530 spectrometer (Japan). The wavelengths of maximum absorption were as follows: MEL: 361.6 nm, 5-HMEL: 359.6 nm, 5-CMEL: 362.2 nm, and IS: 325.2 nm. The structures of the compounds and their UV spectra are shown in Fig. 1. These measurement solutions were prepared using primary stock standard solutions. The stock solutions of MEL, 5-HMEL, and 5-CMEL were diluted with 0.1 M phosphate buffer. The IS stock solution was diluted with methanol. All solution concentrations were 10 µg/mL. The detection wavelength of the HPLC-UV system was set at 360 nm. Chromatograms of blank plasma and plasma spiked with MEL, 5-HMEL, 5-CMEL, and IS are shown in Fig. 2. The analysis was conducted using an isocratic eluent (Fig. 2(A)) and gradient conditions (Fig. 2(B)). All compounds were well separated from the plasma components under both eluent conditions. The elution conditions were changed from isocratic to gradient to improve the sensitivity. The signal-to-noise (S/N) ratio for MEL under gradient conditions was twofold greater than that under isocratic conditions. The gradient conditions were selected for linearity, accuracy, and precision and to shorten the measuring times.
Measurements using (A) isocratic and (B) gradient eluents: 1: 400 ng/mL 5-HMEL, 2: 400 ng/mL 5-CMEL, 3: 500 ng/mL IS, 4: 400 ng/mL MEL.
Because HLB, which has both hydrophilic and hydrophobic substitution groups, was adopted for this experiment in the SPE, the retention of substrates was affected by the pH and polarity of the rinse solution. As a first step, 2% formic acid, ammonia sol./water (5 : 95), water, and 5% methanol were used as rinse solutions to evaluate the effects of pH and polarity. The recovery ratios for 2% formic acid, ammonia sol./water (5 : 95), water, and 5% methanol were 57.9 to 86.0%, 73.6 to 92.8%, 75.4 to 82.7%, and 47.9 to 86.6%, respectively, according to preliminary experiments. These results indicated that ammonia sol./water (5 : 95) showed the best recovery ratio (73.6 to 92.8%). The effect of the ammonia concentration of the solution on the percentage recovery was investigated further (Table 1). The ammonia concentration used was ammonia solution : water=10 : 90, as this concentration afforded the highest percentage recovery overall.
| Recovery (%) | |||
|---|---|---|---|
| Ammonia sol./water (5 : 95) | Ammonia sol./water (7.5 : 92.5) | Ammonia sol./water (10 : 90) | |
| MEL | 82.4±3.53 | 86.2±2.11 | 86.9±4.25 |
| 5-HMEL | 87.8±1.26 | 89.5±1.03 | 90.1±3.20 |
| 5-CMEL | 90.8±1.68 | 94.5±0.98 | 93.4±1.39 |
| IS | 85.9±6.17 | 87.7±2.28 | 89.4±2.17 |
During the rinse solution study, 15 to 50% of the substrates were found to elute without being retained in the HLB when 5% methanol was used; therefore, higher concentrations of methanol were evaluated as eluents. Because all the expected analytical peaks appeared without any interference from impurity peaks, 100% methanol was adopted as the eluent (Figs. 2(A), (B)). The percentage recovery was investigated using test samples containing 1000 ng/mL MEL, 5-HMEL, and 5-CMEL and 500 ng/mL IS.
Linearity of the Calibration CurvesTo establish calibration curves for MEL, 5-HMEL, and 5-CMEL, measurements were performed in the range of 20 to 1000 ng/mL in rat plasma. The curves were proportional to the concentration of the analytes over the range of 40 to 1000 ng/mL. Because the accuracy of 20 ng/mL did not meet NIHS guidelines, these points were excluded for this investigation. The equations for the calibration curves are shown in Table 2. The relationships between the MEX, 5-HMEL, or 5-CMEL peak area divided by the IS peak area and the corresponding concentrations were found to be linear. The correlation coefficients were ≥0.9998.
| Equation | R2 | |
|---|---|---|
| MEL | y=(0.0075±0.0003)x+(0.0053±0.0645) | 0.9998 |
| 5-HMEL | y=(0.0044±0.0001)x+(−0.0230±0.0618) | 0.9999 |
| 5-CMEL | y=(0.0075±0.0004)x+(0.0218±0.1086) | 0.9998 |
Table 3 shows the intra- and inter-day accuracy and precision of this method for MEL, 5-HMEL, and 5-CMEL. The intra- and inter-day accuracy was within ±19.0% for the LLOQ and within ±6.3% for the other concentrations. The precision of the LLOQ was less than 17.0%; the precision was less than 8.0% for the other concentrations. These accuracy and precision values were within standard guidelines.
| Theoretical concentration (ng/mL) | Intra-day (n=5) | Inter-day (n=3) | ||||
|---|---|---|---|---|---|---|
| Mean concentration (ng/mL) | Accuracy (%) | Precision (%) | Mean concentration (ng/mL) | Accuracy (%) | Precision (%) | |
| MEL | ||||||
| 40 | 46.0 | 115.1 | 6.3 | 41.4 | 103.6 | 14.1 |
| 100 | 101.5 | 101.5 | 2.4 | 100.7 | 100.7 | 3.0 |
| 200 | 201.5 | 100.8 | 5.0 | 200.5 | 100.3 | 1.0 |
| 400 | 384.7 | 96.1 | 3.2 | 397.3 | 99.3 | 2.8 |
| 1000 | 1003.6 | 100.4 | 5.7 | 1002.7 | 100.3 | 0.2 |
| 5-HMEL | ||||||
| 40 | 47.6 | 119.0 | 2.2 | 41.9 | 104.7 | 12.7 |
| 100 | 102.8 | 102.8 | 2.0 | 101.8 | 101.8 | 3.2 |
| 200 | 199.8 | 99.9 | 2.2 | 201.2 | 100.6 | 0.6 |
| 400 | 384.8 | 96.2 | 2.2 | 395.1 | 98.8 | 2.3 |
| 1000 | 1006.8 | 100.7 | 8.0 | 1004.5 | 100.5 | 0.5 |
| 5-CMEL | ||||||
| 40 | 46.5 | 116.3 | 1.8 | 40.8 | 101.9 | 17.0 |
| 100 | 106.3 | 106.3 | 1.9 | 102.7 | 102.7 | 4.4 |
| 200 | 201.1 | 100.6 | 4.1 | 200.3 | 100.1 | 0.7 |
| 400 | 380.2 | 95.0 | 3.0 | 393.4 | 98.4 | 2.9 |
| 1000 | 1009.0 | 100.9 | 7.0 | 1001.0 | 100.1 | 0.8 |
The concentration of rat plasma was measured at 0, 0.16, 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, and 10 h after oral administration of MEL. The plasma concentration/time profiles after administration of MEL (2.0 mg/kg) in rats are shown in Fig. 3. The PK parameters of MEL in rats are listed in Table 4. These parameters could not be compared with previous studies because the animal experiments were not conducted under the same conditions used in this paper. PK parameters vary by administration method, dosage form, and animal species. The concentration of MEL in the plasma samples was detectable at each time point. However, the concentrations of 5-HMEL and 5-CMEL, the metabolites of MEL, in rat plasma were lower than the LLOQ at several sampling times (a few sampling point concentrations were greater than the LLOQ; see Fig. 3(B)). These results were considered to be preliminary because the data in Fig. 3(B) were not exact profiles. Therefore, an experiment involving intravenous administration of 5-HMEL was performed. The plasma concentration/time profiles after this administration of 5-HMEL (20 mg/kg) in rats are shown in Fig. 4. The PK parameters of 5-HMEL and 5-CMEL in rats are listed in Table 5. The results showed clearly that 5-HMEL was metabolized to 5-CMEL in vivo in rats.
Each point represents the mean values with vertical bars showing the S.D. of 1 to 3 experiments.
| PK parameter | Mean±S.D. |
|---|---|
| Cmax (µg/mL) | 1.6±0.4 |
| tmax (h) | 2.7±1.5 |
| t1/2 (h) | 2.6±0.4 |
| AUCinf (µg h/mL) | 8.1±1.6 |
Each point represents the mean values with vertical bars showing the S.D. of 1 to 3 experiments.
| PK parameter | Mean±S.D. |
|---|---|
| 5-HMEL | |
| C0 (µg/mL) | 46.2±138.4 |
| t1/2 (h) | 0.4±0.1 |
| AUCinf (µg h/mL) | 18.4±8.5 |
| 5-CMEL | |
| Cmax (µg/mL) | 8.0±0.9 |
| tmax (h) | 0.2±0.0 |
| t1/2 (h) | 0.3±0.1 |
| AUCinf (µg h/mL) | 6.0±3.2 |
Rat plasma concentration/time profiles after oral administration of MEL have been reported4–9); however, the profiles for 5-HMEL and 5-CMEL have not been investigated. We have developed a system to measure the concentration of these metabolites in plasma.
A detection method for MEL, 5-HMEL, and 5-CMEL in rat plasma using a readily accessible HPLC-UV technique was developed, validated, and applied to the analysis of plasma concentration profiles after administration of MEL and 5-HMEL. Therefore, the transition from MEL to metabolites and the enterohepatic circulation of MEL can be clarified by the plasma concentration/time profiles of MEL, 5-HMEL, and 5-CMEL. This method is also applicable for analysis of the metabolism of MEL by CYPs and the genetic polymorphism of CYPs in vitro.
To the best of our knowledge, this is the first report of quantitative and qualitative measurement of MEL and its metabolites using a HPLC-UV method. This analytical method is applicable for PK studies in other species.
This study was supported by a research Grant for the Research on Regulatory Science of Pharmaceuticals and Medical Devices from the Japan Agency for Medical Research and Development (AMED). The authors thank Yuki Ota, BSc, Rei Kimura, BSc, Naoya Kuroda, BSc, Juri Konno, BSc, Keisuke Namatame, BSc, Keiko Miyaki, BSc, Mizuho Mori, BSc, Mari Yamanaka, BSc, and Yuki Sunagawa, BSc for their assistance.
The authors declare no conflict of interest.
