Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

Contents lists available at ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba

Simultaneous determination of omeprazole and their main metabolites in human urine samples by capillary electrophoresis using electrospray ionization-mass spectrometry detection夽 a ˜ ˜ Juan José Berzas Nevado a , Gregorio Castaneda Penalvo , Rosa María Rodríguez Dorado a , b,∗ Virginia Rodríguez Robledo a b

Department of Analytical Chemistry and Food Technology, Faculty of Chemistry, University of Castilla-La Mancha, Ciudad Real, Spain Department of Analytical Chemistry and Food Technology, Faculty of Pharmacy University of Castilla-La Mancha, Albacete, Spain

a r t i c l e

i n f o

Article history: Received 26 September 2013 Received in revised form 13 December 2013 Accepted 18 December 2013 Available online 17 January 2014 Keywords: Capillary electrophoresis-mass spectrometry Human urine Metabolites Omeprazole Pharmacokinetics

a b s t r a c t We report a novel method for the simultaneous determination of omeprazole and their main metabolites (omeprazole sulphide, omeprazole sulphone and 5-hydroxy omeprazole) in human urine samples. For this purpose, two new capillary electrophoresis (CE) methods were developed for the simultaneous determination of target compounds, using initially diode-array for optical detection and electrospray ionization-mass spectrometry (ESI-MS) for metabolites identification and identity confirmation. A new metabolite (5-hydroxysulphide omeprazole) was identified by electrospray ionization multi-stage mass spectrometry (ESI-MS2) fragment which was then used to support the proposed chemical structure. Pharmacokinetic results using CE method were compared with those obtained when a HPLC method was used. Equivalent pharmacokinetics profiles resulted when any analytical methods were carried out. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Omeprazole (OME), which is the pharmaceutical name for 5-methoxy-2[[(4-methoxy-3,5-di-methyl-2-pyridinyl)methyl] sulphinyl]-1H-benzimidazole, is a substituted benzimidazole which reacts with a cysteine group in H+/K+ ATPase under the acid conditions of the stomach and diminishes the ability of parietal cells to produce gastric acid as a result [1]. OME is typically used to treat acid-induced inflammation and ulcers of the stomach and duodenum, gastroesophageal reflux disease, erosive esophagitis and Zollinger–Ellison syndrome, as well as to prevent upper gastrointestinal bleeding in ill patients. The drug is completely metabolized mainly in the liver, to several primary metabolites including omeprazole sulphone, 5-hydroxy omeprazole and omeprazole sulphide, which have no significant effect on acid secretion [2].

夽 In memoriam of professor Juan José Berzas Nevado. ∗ Corresponding author at: Departamento de Química Analítica y Tecnología de los Alimentos, Universidad de Castilla-La Mancha, 02071 Albacete, Spain. Tel.: +34 967599200x8240. E-mail address: [email protected] (V.R. Robledo). 0731-7085/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jpba.2013.12.020

Omeprazole has so far been determined in a number of formulations and biological fluids by using a variety of analytical techniques including spectrophotometry, high-performance liquid chromatography with ultraviolet detection and liquid chromatography coupled with tandem mass spectrometry [3]. Various chromatographic methods for isolating omeprazole and their metabolites have been reported mostly in human plasma samples [4]. The HPLC technique, in combination with UV [5–8] and mass spectrometry [9–13] mainly, has to date been the most common choice for this purpose. Rambla et al. [14] used micellar liquid chromatography technique to determine omeprazole and their two major metabolites in physiological samples such as human urine and serum samples. Due to omeprazole and its metabolites are into the same dissolution together with other compounds of matrix, a highly selective and sensible analytical technique should be to choice, in order to carry out the separation and identification of each one of targets compounds. In recent years, capillary electrophoresis (CE) has proven to be a fast, valid and reliable technique that allows separate complex mixtures of organic compound requiring only small amounts of sample and buffer. For example the target drug and their metabolites have been resolved by micellar electrokinetic capillary chromatography with UV detection [15] just in plasma samples. Capillary electrophoresis coupling mass spectrometry (CE–MS) and

212

J.J.B. Nevado et al. / Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

Fig. 1. Chemical structures of omeprazole, 5-hydroxy omeprazole, omeprazole sulphide and omeprazole sulphone.

tandem MS/MS, which combine the high efficiency and resolution of CE with the intrinsically high selectivity and sensitivity of MS, provide a highly attractive method for analytical determinations of mostly unknowing compounds [16–19]. So, capillary electrophoresis with mass spectrometric detection is an effective alternative to LC–MS in this context in order to identify and confirm the present of some metabolites. The aim of this work was to develop and validate a simple and rapid capillary zone electrophoresis (CZE) method for the simultaneous separation and determination of omeprazole and its metabolites such as omeprazole sulphide, omeprazole sulphone and 5-hydroxy omeprazole (see Fig. 1) in human urine samples. To this end, a new CE–UV method was developed, optimized and validated in terms of precision, linearity, accuracy, robustness, and detection and quantitation limits. The effects of the main chemical and instrumental variables influencing the analyte separation efficiency were studied. The proposed method was applied to real urine samples in order to study the pharmacokinetics of 5-hydroxy omeprazole previous isolation of target analytes by solid-phase extraction procedures. The presence of several unknown peaks such as compounds come from matrix, interfering analytes or even other types of metabolites, led us to develop a new CE–ESI-MSn method in order to identify

and confirm the chemical structures of studied metabolites. Omeprazole metabolites were determinate and new metabolite, 5-hydroxysulphide omeprazole never before reported, was also identified.

2. Material and methods 2.1. Chemicals and reagents Omeprazole was supplied by Sigma (Madrid, Spain). 5-Hydroxy omeprazole, omeprazole sulphone and omeprazole sulphide were obtained from AstraZeneca (Södertälje, Sweden). The buffer reagents [viz. sodium dihydrogen phosphate, ammonium acetate, analytical-grade ammonium hydroxide and sodium hydroxide (which were used to both adjust pH and rinse the capillary)] were purchased from Panreac (Barcelona, Spain). The organic solvents, acetonitrile and methanol were obtained from Sigma; and 2-propanol and ethanol from Panreac. Ultrapure water from a Milli-Q apparatus (Millipore, Milford, MA) was used throughout. Nitrogen gas for the LCQTM and helium damping gas for the iontrap were both supplied by Air Liquide (Madrid, Spain).

J.J.B. Nevado et al. / Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

213

2.2. Instrumentation

2.4. Operating conditions

A P/ACE MDQ capillary electrophoresis system from Beckman (Fullerton, CA) equipped with a diode array detector (DAD) and software also from Beckman was used. The capillary cartridge was thermostated at 16 ◦ C, and a detection window of 800 ␮m × 100 ␮m used. The CE system was coupled to an LCQ DECA XP Plus mass spectrometer from ThermoFinnigan (San Jose, CA) that was equipped with a tricoaxial pneumatically assisted electrospray ionization (ESI) source designed for coupling CE to MS, and also with an ion trap (IT). Mass spectrometry data were acquired and processed with the aid of the software Xcalibur 1.4. HPLC analyses were performed on a model LC-10A chromatograph from Shimadzu (Tokyo, Japan) equipped with a model SPD-M10A DAD detector and furnished with a double pump system intended to facilitate gradient operation. A Rheodyne (Cotati, CA) model 7725 injector with a 20 ␮L sample loop was also used. All measurements were made and data processed by using a Silicon computer running the software CLASS-LC 10. Phosphate buffer solutions and water samples were filtered through a 0.45 ␮m MFTM-membrane filters, and methanol was filtered through a 0.5 ␮m FluoroporeTM membrane filters. Both membrane filters were purchased from Millipore (Milford, MA, USA). A model 2002 pH-meter from Crison Instruments (Alella, Barcelona, Spain) was used to adjust the pH of the separation buffers. Samples were mixed and agitated in a vortex shaker for tubes (OVAN, Barcelona, Spain) and urine cleared on a model S240 centrifuge from Selecta (Barcelona, Spain). Extraction and preconcentration of the samples were accomplished by using a custom-made assembly consisting of a water manifold (Supelco VisiprepTM Sep-Pack system, Madrid, Spain) coupled to a Millipore XF 54 23050 vacuum pump. Solid-phase extraction (SPE) was done in Sep-Pack Plus C18 500 mg cartridges from Waters (Milford, CA). The extracts obtained from human urine samples were evaporated by dry block heating in Pierce Reactive-ThermTM heating modules which provided precise, uniform temperature control and N2 flow rate.

2.4.1. CE–UV method Separation was done in a 30 cm long (20 cm to detector) × 75 ␮m ID fused-silica capillary housed in a cartridge with a 800 ␮m × 100 ␮m detector window. Prior to first use, each capillary was conditioned with 0.5 M NaOH, water and electrophoretic buffer for 10 min each. Further flushing with 0.5 M NaOH for 0.5 min and running electrolyte for 1 min were applied at the start of each sequence of analysis and between runs in order to ensure the obtainment of well-defined peaks and reproducible migration data, and also to avoid contamination with interferences potentially present in urine. The composition of background electrolyte was 60% phosphate buffer (pH 12, 25 mM) and 40% acetonitrile. Samples were injected hydrodynamically at 0.5 psi for 3 s each and refrigerated at 20 ◦ C inside the equipment. An applied voltage of 20 kV and a ramp of 33.3 kV/min were used. All electropherograms were recorded at 301 nm.

2.3. Samples and solutions Stock solutions containing 1000 mg L−1 concentration of omeprazole and each of their metabolites were prepared by dissolving the required amount of pure solid compound in ethanol. These solutions were stored at 4 ◦ C in the dark. Working standard solutions were daily prepared by appropriate dilution of the stock solutions with de-ionized (Milli-Q) water. Working standard solutions at 10 mg L−1 concentration each were prepared on a daily basis by diluting appropriate aliquots of the previous standard stock solutions in Milli-Q water. When CE–ESI-MS was used, appropriate tuning standard solution containing a 20 mg L−1 concentration of omeprazole and their metabolites were prepared and diluted in running buffer. Sheath liquid consisted of 70% 2-propanol and 5 mM running buffer, and was passed at a flow rate of 3 ␮L min−1 . This solution was also prepared freshly each day and degassed by sonication for 5 min prior to use in order to ensure accurate, reproducible ionization.

2.4.2. CE–ESI-MS and CID-MS/MS detection using the MRM mode Electrophoretic separations were done on 80 cm fused-silica capillaries (75 ␮m i.d., 375 ␮m o.d.), using 10 mM ammonium acetate/ammonium hydroxide buffer at pH 9.0 and previously degassed by sonication as running electrolyte. A separation voltage of 30 kV with an initial ramp of 0.2 min was used for electrophoretic separation. Samples were injected at a pressure of 0.5 psi for 10 s. Prior to first use, each capillary was conditioned by consecutive flushing with 1.0 M NaOH at 20 psi for 10 min and water at 20 psi for 10 min, followed by electroconditioning with BGE at 20 psi for 5 min. The sheath gas flow was set at 50 on the scale of arbitrary units provided by the instrument software and the auxiliary gas flow at 0. The heated capillary temperature was held at 200 ◦ C and the electrospray needle set at 4.00 kV. About 5 cm of the poly-imide coating was removed from the fused-silica capillary in order to avoid dissolution by the sheath liquid and minimize contamination of the electrospray source as a result. CE–MS data were acquired throughout the m/z range 190–400 m/z (135–350), using the centroid mode, negative polarity and an injection time of 300 ms; scans were done at 3 ␮s intervals. The MS detector was calibrated with the tuning standard solution that was infused through the capillary to the detector at an injection pressure of 0.5 psi under a constant CE voltage of 3 kV. In both CE detection systems, different electrolyte vials were used for rinsing and separation in order to maintain a constant electrolyte level on the anode side. Each set of separation vials was changed after 2 separation runs. Prior to overnight storage, the capillary was flushed with 0.5 M NaOH and water for consecutive periods of 10 min, followed by drying with air for 3 min. 2.4.3. HPLC–UV method A binary gradient elution program containing a mobile phase (A) methanol and (B) phosphate buffer (50 mM, pH 7.2) was selected. An elution gradient (0–10 min, 15% A → 60% A) using a constant flow rate of 1 mL min−1 and an injection volume of 20 ␮L were used for the analytical separation. The analytical column was a Kromasil (Bohus, Sweden) C18 model (150 mm long, 4.6 mm i.d., 5 ␮m particle size). All chromatograms were recorded at 301 nm. 2.4.4. Treatment of urine samples by solid-phase extraction procedure Fresh urine samples were collected, buffered at pH 8.0 with sodium carbonate and stored at −20 ◦ C prior to analysis. Before electrophoretic separation, the samples were subjected to

214

J.J.B. Nevado et al. / Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

Table 1 Precision obtained in the determination of omeprazole and its metabolites at nominal concentration of 35 ␮g L−1 each. RSD (%) Sulphide

Within-day (n = 8) Between-day

Hydroxy

Omeprazole

Sulphone

MT

CPA

H

MT

CPA

H

MT

CPA

H

MT

CPA

H

0.68 2.78

2.25 2.24

1.67 0.36

1.25 2.75

2.90 1.02

2.46 0.49

1.35 2.92

2.45 3.67

6.11 2.64

1.37 2.96

2.81 0.36

3.24 1.04

solid-phase extraction procedure in order to remove potential interferences. The SPE procedure was carried out using reversed phase 500 mg Sep-Pak Plus C18 cartridges from Waters (Milford, MA). The sorbent was previously conditioned with 5 mL of methanol and 5 mL of 10 mM phosphate buffer at pH 7. Then, 9.0 mL of urine sample was slowly loaded onto the cartridge and subsequently washed with 2 mL of 100 mM phosphate buffer at pH 7, followed by 7.5 mL of 20:80 methanol/water (v/v). Finally, the target analytes were eluted with 4 mL of acetonitrile and the eluates were dried with N2 heating at 40 ◦ C. The final residue was re-dissolved in 100 ␮L of 50:50 methanol/water (v/v) and injected into the CE and HPLC instruments. 3. Results and discussion 3.1. CE–UV method 3.1.1. Optimization of electrophoretic method The effect of pH on the separation of omeprazole and their metabolites was examined over the pH range 9–13, using 20 mM phosphate buffer for adjustment. Between pH 9 and 11.5, omeprazole, 5-hydroxy omeprazole and omeprazole sulphone occur in anionic form, whereas omeprazole sulphide is in neutral form and migrates with EOF. All analytes are anionic over the pH range 11.5–13, and, pH 12 was thus selected to ensure optimal separation between analyte peaks. Various organic additives (including methanol, ethanol, acetonitrile and isopropanol) were studied to improving resolution. The solvents were added in proportions from 10 to 45% to the background electrolyte (20 mM phosphate buffer at pH 12). Based on the results, acetonitrile was the best candidate. A proportion of 40% acetonitrile was used as the best compromise between analysis time and peak resolution. The influence of the phosphate buffer (pH 12) concentration, the effect of the applied voltage, the voltage ramp and the capillary temperature on separation were examined over the range of 10–30 mM, 5–25 kV, 0.1–0.8 min (200–25 kV/min) and 15–30 ◦ C respectively. Optimum values of 25 mM concentration, 20 kV and 0.6 min (33.3 kV/min) and 16 ◦ C, respectively, were selected in order to obtain a good resolution and short analysis time. With goal to lower detection limits and increase analytical sensitivity, the effect of the injection time was examined over the range 3–10 s, using an injection pressure of 0.5 psi. Increasing the injection time decreased resolution between analytes by increasing peak areas. Injection time of 3 s was therefore adopted. 3.1.2. Validation of CE method In order to assess the reliability of CE–UV proposed method, precision, limits of detection and quantitation, linearity and recovery were evaluated. The validation procedure was carried out using electrophoresis separation and diode-array detection (CE–DAD) for the quantification. The proposed method was validated by using

an extract from urine samples previously spiked with omeprazole and their metabolites. Mentioned extracts were obtained from solid phase extraction procedure. Previously every stage of SPE procedure (organic and aqueous solvents volume for washing stages, samples volume, eluent volume and extracts final volume) was evaluated in an effort to achieve a complete extraction of selected compounds. The best recoveries for each compound were achieved using optimized SPE procedure already described in Section 2.4. As shown in the validation results which are discussed below, the optimized SPE procedure provided recoveries near 100%. The limits of detection (LOD) and quantitation (LOQ) were calculated from baseline noise. Thus, LOD and LOQ for each analyte were taken to be the concentrations giving peak three and ten times, respectively, as high as baseline noise [20]. LODs and LOQs among 8.7–26.1 ␮g L−1 and 29.0–86.9 ␮g L−1 respectively were obtained. The proposed method was also assessed in terms of repeatability and reproducibility by determining the precision in the migration times (MT), the corrected peak areas (CPA) and peak high (H). Appropriate tests on extracts from urine samples, previously spiked with the analytes and subjected to the SPE procedure (final concentration 300 ␮g L−1 ), were used to this end. The relative standard deviations (RSD) thus obtained are shown in Table 1. Linearity was evaluated by using spiked urine samples subjected to the SPE procedure. The results, in the form of CPAs, were obtained by triplicate injection of each analyte at five different concentration levels. The detector response was linearly dependent on the sample concentration over the range 0.03/0.08–20 mg L−1 . All calibration curves exhibited acceptable regression coefficients (R > 0.99). Table 2 shows the statistical figures of merit for each analyte. The accuracy of the proposed method was assessed by analysing several human urine samples previously spiked at three different levels (low, medium and high concentrations) with omeprazole and their metabolites. All samples were analyzed under the above-described optimum isolation and preconcentration procedure using SPE and CE conditions. The results were calculated by comparison with spiked blank urine extracts used as external standards at final concentration of 100, 300 and 500 (␮g L−1 ). Averaged recoveries among 96.2 and 99.5% were obtained for all studied compounds at each level.

Table 2 Calibration curves and regression coefficients for omeprazole and its metabolites under optimized CE conditions: electrolyte composition (ammonium buffer pH 9.0, 10 mM), separation voltage and temperature of 30 kV and 16 ◦ C respectively and injection time of 10 s. Analytes

Equation

R2

Omeprazole Omeprazole sulphide 5-Hydroxy omeprazole Omeprazole sulphone

Y = (245.40 ± 3.32) X + (−26.85 ± 34.08) Y = (414.75 ± 4.86) X + (−49.56 ± 49.91) Y = (177.74 ± 2.44) X + (−24.08 ± 25.09) Y = (263.99 ± 4.24) X + (−11.94 ± 43.49)

0.99958 0.99945 0.99943 0.99923

J.J.B. Nevado et al. / Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

215

Table 3 External and internal factors governing robustness and ruggedness, and levels used for their assessment. Factor

Type

Optimal

Level (–)

Level (+)

M

A: buffer pH B: buffer concentration C: % acetonitrile D: voltage E: temperature F: injection time G: urine volume H: elution volume I: different days J: capillaries K: SPE cartridge lots

Internal Internal Internal Internal Internal Internal Internal Internal External External External

12.0 25 mM 40 20 kV 16 ◦ C 3s 9 mL 3 mL – – –

11.8 24 39 19 15 2.8 8.8 2.8 I I I

12.2 26 41 21 17 3.2 9.2 3.2 II II II

−1.59 −1.06 −0.73 −0.46 −0.22 0.00 0.22 0.46 0.73 1.06 1.59

Table 4 Optimum separation conditions for CE–ESI-MS developed.

3.2. Analysis of real urine samples

Variable

Studied range

Optimum conditions

Electrolyte composition

pH: 6.5–11.0 10–50 mM ammonium buffer 5–30 kV 15–30 ◦ C 3–10 s 40–90% (2-propanol) 5–20 mM buffer 2–7 ␮L min−1 (flow rate) 10–70 arbitrary units 2–5 kV 150–400 ◦ C 0–50 mm

pH 9.0 10 mM ammonium buffer 30 kV 16 ◦ C 10 s 40% 2propanol + 5 mM running buffer 3 ␮L min−1

1.00–1.25 cm

1.00 cm

0–400 ms

200 ms

Separation voltage Separation temperature Injection time Sheath liquid

Sheath gas Spray voltage Capillary temperature Length of CE capillary protruding from ES needle Distance between CE capillary and MS heated capillary Injection time for IT

50 arbitrary units 4 kV 200 ◦ C 0 mm

3.1.3. Evaluation of integral robustness and ruggedness In this work, we evaluated the potential effects of internal (robustness) and external variables (ruggedness) of the method on performance [21]. To this end, we used the Plackett–Burman factorial model, which is based on balanced incomplete blocks. For statistical reasons, a design with fewer than eight experiments is considered ineffective and one with more than 24 impractical [22]. Plackett–Burman design for evaluating robustness and ruggedness simultaneously by using 11 factors and 12 experiments (N = 12) was used. Table 3 shows the external (ruggedness) and internal (robustness) factors (A–K) used to construct the model; the plus and minus signs in it denote the upper and lower limit, respectively, of the optimum level for each variable. The effects of changing the levels of the most critical electrophoretic responses of the method were examined in terms of CPAs, peak heights and migration times for the analytes. M was taken to be constant for each specific given design and to represent the means of the order statistics (Validation of Compendia Methods) [23] for a sample size of eleven (N = 11). The ranked effects of the 12 factors (on the X-axis) were plotted against M (on the Y-axis) for each critical electrophoretic response. Robustness and ruggedness were evaluated with triplicate injections of urine samples spiked with a 1.5 mg L−1 concentration of each analyte. The test revealed that our off line SPE/CE–UV method is both robust and rugged throughout the variation ranges studied for each critical electrophoretic response.

3.2.1. Off-line SPE/CE–UV method The applicability of the proposed off line SPE/CE–UV detection method was confirmed by application to urine samples from several patients under medical treatment with omeprazole. Corrected peak areas (CPAs) (viz. area/migration time ratios) were used for quantitative analysis. Samples were analyzed by using the standard addition method. Urine from volunteers under treatment with omeprazole at 40 mg per day was analyzed at different times after administration in order to extract useful pharmacokinetic information. In all cases, a main peak at migration time corresponding 5hydroxy omeprazole was detected. The pharmacokinetic study of 5-hydroxy omeprazole is shown in Fig. 2. All determinations were carried out in triplicate using both CE and HPLC techniques. The results of our study showed that the highest concentration of 5-hydroxy omeprazole was found 145 min later after the dosing (30 mg L−1 ). The presence of this metabolite in the urine could be quantified up to 720 min after administration at level of 0.1 mg L−1 . As can be see, both CE and HPLC method showed a good concordance in the values obtained at different times. The pharmacokinetics showed that the proposed off SPE/CE–UV method is effective toward the determination of target analytes with acceptable accuracy and could be useful to obtain relevant and complex clinical information related to bioactivity, t½, excretion, etc., for this drug and its main metabolite. Other peak at migration time and spectrum very similar than omeprazole, was also detected. Finally this peak was submitted to standard addition with omeprazole standard solution, resulting different compound than target drug. Therefore, anew CE method using mass spectrometry detector was developed and optimized in order to obtain structural information about the unknowing peak.

Fig. 2. CE–ESI-MS electropherograms for a spiked extract of urine containing a mixture 10 ␮g L−1 concentration of omeprazole and its metabolites as recorded under optimum conditions.

216

J.J.B. Nevado et al. / Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

Fig. 3. MS–MS spectra corresponding to new metabolite (m/z 344 to 179) for the urine sample.

Table 5 MSn characterization data for omeprazole and its studied metabolites (collision energy (%) MS1 to MS2 and MS2 to MS3 was 23 and 30 respectively). Analytes

Molecular ions MS1 [M−H]− m/z

Product ions MS2 mode m/z

Product ions MS3 mode m/z

Omeprazole 5-Hydroxy omeprazole Omeprazole sulphide Omeprazole sulphone Unknown metabolite: proposed chemical structure as 5-Hydroxy sulphide omeprazole

344 360 328 360 344

194 194 179 210 179

179 179 164 147 164

3.2.2. Optimization of CE–ESI-MS method In the same way that CE–UV method, optimal electrophoretic conditions for the simultaneous determination of omeprazole and their metabolites, compatible to MS detector, were studied. For that, blank urine extracts previously spiked with the analytes were used. A series of preliminary tests was conducted in order to improving performance in the coupled CE–MS system. Non-volatile CE buffer components were replaced by volatile compounds (ammonia, acetate, formiate) in order to make them compatible with mass spectrometry system. The effects of the variables potentially affecting both electrophoretic separation and mass spectrometer detector were examined in order to ensure good resolution between peaks, short migration times and an acceptable signal/noise ratio.

The optimum CE–ESI-MS separation conditions and studied range are summarized in Table 4. Fig. 3 shows an electropherogram of spiked extract of urine containing omeprazole and their metabolites under selected optimum CE–MS conditions. 3.2.3. CE–MS detection and confirmation of structural assignments The presence of several unknown peaks such as compounds come from matrix, interferences analytes or even other types of metabolites, led us to develop a new CE–ESI-MS method in order to confirm the chemical structures of studied metabolites. Human urine samples were also injected into the CE–ESI-MS system. In the same way than CE–UV, two main peaks were

J.J.B. Nevado et al. / Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

217

Fig. 4. CE–MS electropherogram for a human urine sample obtained 3 h after administration of 40 mg of omeprazole. The transitions of m/z 344 to 179 and 179 to 164 were used for the identification of new metabolite and the transitions of m/z 360 to 194 and 194 to 179 were used for 5-OH-omeprazole.

detected one of them at 344 m/z and other at 360 m/z, respectively (see Fig. 4). In a first approach, the compounds could be identified by rapport to migration times, standard addition and MS data of each compound. But the structural confirmation of 5-hydroxy omeprazole and the unambiguous identification of unknown analyte required a selective fragmentation in order to obtain MSn data. Supplementary structural information was obtained using electrospray ionization multi-stage mass spectrometry (ESI-MSn ) detection in order to confirm the chemical structure of analytes present in the samples. Table 5 summarized MSn data for omeprazole, its main metabolites and unknown metabolite. As it was expected, the metabolite 5-hydroxy omeprazole with a characteristic transition of m/z 360 to 194 was identified. In addition

in this table we have proposed a chemical structure for the new metabolite taking into account the MSn information (see also Fig. 1). Firstly in order to elucidate the chemical structure of unknown peak detected at m/z 344 (same ion [M−H]− value than omeprazole) a MS/MS spectrum was carried out. As can be seen in Fig. 5 and Table 5, the transition m/z 344 to 179 is totally different than omeprazole (m/z 344 to 194). Nevertheless the compounds with m/z 344 have got the same transition than omeprazole sulphide and provides the same lost than 5-Hydroxy sulphide omeprazole in the first transitions (360 → 194 and 344 → 179 respectively). Therefore, chemical structure for the new metabolite, has been proposed having in mind the mass spectrometry behavior of omeprazole,

218

J.J.B. Nevado et al. / Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

Fig. 5. Pharmacokinetics profile of 5-hydroxy-omeprazole in urine samples of patient under oral treatment with 40 mg of omeprazole by CE and HPLC methods.

5-hydroxy omeprazole and omeprazole sulphide This new metabolite was never previously found in urine samples and it has been named 5-hydroxysulphide omeprazole.

4. Conclusions In this work, we optimized the separation of omeprazole and their metabolites (omeprazole sulphide, omeprazole sulphone and 5-hydroxy omeprazole) by capillary zone electrophoresis for their UV–Vis detection. The analytes were extracted from urine by solid phase extraction (SPE) since potential interferences of the urine matrix are suppressed, increasing sensitivity and improving detection limits as a result. The proposed method is expeditious (a whole run takes only 7 min to complete) and provides limits of detection from 8.7 to 26.1 ␮g L−1 for target analytes. The method, which involves extraction, preconcentration and electrophoretic separation of the analytes, was validated by application to urine samples from patients under medical treatment with omeprazole. Using CE and HPLC as analytical techniques and SPE as extraction procedure, allowed us to derive pharmacokinetic information about 5-hydroxy

omeprazole. Neither omeprazole sulphide nor omeprazole sulphone were detected in the urine samples, where they typically occur in very small amounts. However, an additional metabolite released concomitantly with the studied analytes was detected. So this proposes a simple, rapid and selective method for the simultaneous determination of omeprazole and their main metabolites by CE–MS was developed. Optimally coupling the CE and MS systems required examining the influence of some variables affecting electrophoretic separation and or electrospray stability. Using mass spectrometry allowed a previously undetected omeprazole metabolite (viz. 5-hydroxysulphide omeprazole) to be identified and the presence of the known analyte 5-hydroxy omeprazole confirmed.

Acknowledgments The authors would like to thank to Astrazeneca for providing research samples of omeprazole and their metabolites. We are indebted to Ministerio de Ciencia e Innovación of Spain for the financial support (Project CTQ 2008-02126/BQU).

J.J.B. Nevado et al. / Journal of Pharmaceutical and Biomedical Analysis 92 (2014) 211–219

References [1] H.D. Langtry, M.I. Wilde, Omeprazole. A review of its use in Helicobacter pylori infection, gastro-oesophageal reflux disease and peptic ulcers induced by nonsteroidal anti-inflammatory drugs, Drugs 56 (3) (1998) 447–486. [2] H. Kanazawa, A. Okada, M. Higaki, H. Yokota, F. Mashige, K. Nakahara, Stereospecific analysis of omeprazole in human plasma as a probe for CYP2C19 phenotype, J. Pharm. Biomed. Anal. 30 (6) (2003) 1817–1824. [3] M. Espinosa, A.J. Ruiz, F. Sánchez, C. Bosch, Analytical methodologies for the determination of omeprazole: an overview, J. Pharm. Biomed. Anal. 44 (4) (2007) 831–844. [4] J. Macek, J. Klíma, P. Ptácek, Rapid determination of omeprazole in human plasma by protein precipitation and liquid chromatography-tandem mass spectrometry, J. Chromatogr. B 852 (1–2) (2007) 282–287. [5] N.L. Rezk, K.C. Brown, A.D.M. Kashuba, A simple and sensitive bioanalytical assay for simultaneous determination of omeprazole and its three major metabolites in human blood plasma using RP-HPLC after a simple liquid–liquid extraction procedure, J. Chromatogr. B 844 (2) (2006) 314–321. [6] M. Shimizu, T. Unoa, T. Niioka, N. Yaui-Furukori, T. Takahata, K. Sugawara, T. Tateishi, Sensitive determination of omeprazole and its two main metabolites in human plasma by column-switching high-performance liquid chromatography: application to pharmacokinetic study in relation to CYP2C19 genotypes, J. Chromatogr. B 832 (2) (2006) 241–248. [7] R.M. Orlando, P.S. Bonato, Simple and efficient method for enantioselective determination of omeprazole in human plasma, J. Chromatogr. B 795 (2) (2003) 227–235. [8] H.M. González, E.M. Romero, T.J. Chávez, A. Aaron, V. Quezada, C. Hoyo-Vadillo, Phenotype of CYP2C19 and CYP3A4 by determination of omeprazole and its two main metabolites in plasma using liquid chromatography with liquid–liquid extraction, J. Chromatogr. B 780 (2) (2002) 459–465. [9] J. Martens-Lobenhoffer, I. Reiche, U. Tröger, K. Mönkemüller, P. Malfertheiner, S.M. Bode-Böger, Enantioselective quantification of omeprazole and its main metabolites in human serum by chiral HPLC-atmospheric pressure photoionization tandem mass spectrometry, J. Chromatogr. B 857 (2) (2007) 301–307. [10] Q. Song, W. Naidong, Analysis of omeprazole and 5-OH omeprazole in human plasma using hydrophilic interaction chromatography with tandem mass spectrometry (HILIC-MS/MS) eliminating evaporation and reconstitution steps in 96-well liquid/liquid extraction, J. Chromatogr. B 830 (1) (2006) 135–142.

219

[11] V.A. Frerichs, C. Zaranek, C.E. Haas, Analysis of omeprazole, midazolam and hydroxy-metabolites in plasma using liquid chromatography coupled to tandem mass spectrometry, J. Chromatogr. B 824 (1–2) (2005) 71–80. [12] E.J. Woolf, B.K. Matuszewski, Simultaneous determination of omeprazole and 5 -hydroxyomeprazole in human plasma by liquid chromatography-tandem mass spectrometry, J. Chromatogr. A 828 (1998) 229–238. [13] U. Hofmann, M. Schwab, G. Treiber, U. Klotz, Sensitive quantification of omeprazole and its metabolites in human plasma by liquid chromatography-mass spectrometry, J. Chromatogr. B 831 (1–2) (2006) 85–90. [14] M. Rambla, J. Esteve, S. Carda, Analysis of omeprazole and its main metabolites by liquid chromatography using hybrid micellar mobile phases, Anal. Chim. Acta 633 (2) (2009) 250–256. [15] T. Pérez-Ruiz, C. Martínez-Lozano, A. Sanz, E. Bravo, R. Galera, Determination of omeprazole, hydroxyomeprazole and omeprazole sulfone using automated solid phase extraction and micellar electrokinetic capillary chromatography, J. Pharm. Biomed. Anal. 42 (1) (2006) 100–106. [16] J.L. Josephs, M. Sanders, P. Shipkova, Detection and Characterization of Pharmaceutical Metabolites, Degradants and Impurities by the Application of MS/MS Software Algorithms, in: Technical Program, February 25, 2007. [17] A.C. Servais, J. Crommen, M. Fillet, Capillary electrophoresis-mass spectrometry, an attractive tool for drug bioanalysis and biomarker discovery, Electrophoresis 27 (13) (2006) 2616–2629. ˜ [18] J.J. Berzas, G. Castaneda, V. Rodríguez, G. Vargas, New CE–ESI-MS analytical method for the separation, identification and quantification of seven phenolic acids including three isomer compounds in virgin olive oil, Talanta 79 (5) (2009) 1238–1246. ˜ [19] J.J. Berzas, G. Castaneda, V. Rodríguez, Advantages of using a modified orthogonal sampling configuration originally designed for LC-ESI-MS to couple CE and MS for the determination of antioxidant phenolic compounds found in virgin olive oil, Talanta 82 (2) (2010) 548–554. [20] K.D. Altria, D. Elder, Overview of the status and applications of capillary electrophoresis to the analysis of small molecules, J. Chromatogr. A 1023 (1) (2004) 1–14. [21] National Formulary 18, The United States Pharmacopeia, 23rd edn., Rockville, MD, 1995. [22] R.L. Plakett, J.P. Burman, The design of optimum multifactorial experiments, Biometrika 33 (1946) 305–325. [23] Validation of Compendia Methods, USP (1994) 23, 1225, USPC, Rockville, MD.

Simultaneous determination of omeprazole and their main metabolites in human urine samples by capillary electrophoresis using electrospray ionization-mass spectrometry detection.

We report a novel method for the simultaneous determination of omeprazole and their main metabolites (omeprazole sulphide, omeprazole sulphone and 5-h...
1MB Sizes 0 Downloads 0 Views