Journal of Chromatography B, 958 (2014) 29–35

Contents lists available at ScienceDirect

Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb

Liquid chromatography–tandem mass spectrometry method for simultaneous quantification of bisoprolol, ramiprilat, propranolol and midazolam in rat dried blood spots Katja Cvan Trobec a , Jurij Trontelj b , Jochen Springer c,d,e , Mitja Lainscak c,f , Mojca Kerec Kos b,∗ a

University Clinic of Respiratory and Allergic Diseases Golnk, Pharmacy Department, Golnik 36, 4204 Golnik, Slovenia University of Ljubljana, Faculty of Pharmacy, Askerceva 7, 1000 Ljubljana, Slovenia c Applied Cachexia Research, Department of Cardiology, Charité Medical School, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany d Center for Cardiovascular Research, Charité Medical School, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany e Norwich Medical School, University of East Anglia, Norwich NR4 7TJ, UK f University Clinic of Respiratory and Allergic Diseases Golnik, Division of Cardiology, Golnik 36, 4204 Golnik, Slovenia b

a r t i c l e

i n f o

Article history: Received 20 November 2013 Accepted 9 March 2014 Available online 15 March 2014 Keywords: Dried blood sample Bisoprolol Ramiprilat Propranolol Midazolam

a b s t r a c t Dried blood spot (DBS) sampling represents a suitable method for pharmacokinetic studies in rats, particularly if serial sampling is needed. To study the pharmacokinetics of drugs in a rat heart failure (HF) model, we developed and validated a method for the simultaneous determination of bisoprolol, ramiprilat, propranolol and midazolam in DBS samples. Bisoprolol and ramipril are widely used in the treatment of HF, and midazolam and propranolol are markers of hepatic metabolism, which can be altered in HF. A 20 ␮L sample of rat blood was pipetted onto Whatman 903 Protein Saver Card and allowed to dry. The whole spot was excised and 300 ␮L of solvent (methanol with 10% ultrapure water and 0.1% formic acid) was added. After mixing and incubating the sample in an ultrasonic bath, a mixture of isotopically labeled internal standards was added. After centrifugation, the extracts were cleaned on an OstroTM plate and analyzed using liquid chromatography–tandem mass spectroscopy. The method was successfully validated. No significant interference was observed in the retention times of analytes or internal standards. The intraday and interday accuracy and precision were within a ±15% interval. The method was linear in the range 5–250 ␮g/L and the lower limit of quantification was 5 ␮g/L for all four analytes. The absolute matrix effect ranged from 98.7% for midazolam to 121% for ramiprilat. The recovery was lowest for ramiprilat and highest for propranolol. Samples were stable at all tested temperatures. The method has been used successfully in a real-time pharmacokinetic study in rats. © 2014 Elsevier B.V. All rights reserved.

1. Introduction

Abbreviations: DBS, dried blood spot; EMA, European Medicines Agency; FDA, Food and Drug Administration; LLOQ, lower limit of quantification; ME, matrix effect; RSD, relative standard deviation; QC, quality control sample; QCl , quality control sample of 15 ␮g/L (low concentration); QCm , quality control sample of 75 ␮g/L (medium concentration); QCh , quality control sample of 200 ␮g/L (high concentration). ∗ Corresponding author at: University of Ljubljana, Faculty of Pharmacy, Chair of Biopharmaceutics and Pharmacokinetics, Askerceva 7, SI-1000 Ljubljana, Slovenia. Tel.: +386 1 47 69 500; fax: +386 1 42 58 031. E-mail addresses: [email protected] (K. Cvan Trobec), [email protected] (J. Trontelj), [email protected] (J. Springer), [email protected] (M. Lainscak), [email protected] (M. Kerec Kos). http://dx.doi.org/10.1016/j.jchromb.2014.03.009 1570-0232/© 2014 Elsevier B.V. All rights reserved.

In any pharmacokinetic study, the pharmacokinetic profile of a drug is assessed by sampling at multiple time points. In such studies in humans, several milliliters of blood can be withdrawn on any single occasion without affecting the patient’s health [1], but in rats the blood volume that can be withdrawn is considerably smaller [2]. An average rat weighing 200 g has approximately 13 mL of blood in circulation [3], thus is not possible to obtain 1 mL of blood at each time point for determination of the drug concentration in plasma. In the past, one animal or even a group of animals frequently had to be killed at each time point; DBS sampling, on the other hand, requires only 10–20 ␮L of blood at each time point, thus pharmacokinetic profiles can be determined using multiple blood samples from an individual rat. The use of rat DBS samples, instead

30

K. Cvan Trobec et al. / J. Chromatogr. B 958 (2014) 29–35

of regular plasma samples, significantly reduces the number of animals needed for experiments and allows repeated determination of pharmacokinetics in the same animals (e.g. in a disease model) [4]. The main objective of this work was to develop and validate an analytical method that would enable simultaneous determination of four different drugs (bisoprolol, ramiprilat, midazolam and propranolol) in rat DBS samples. The method was developed to study the pharmacokinetics of these drugs in a rat heart failure model, using a multiple-sampling-times strategy in a single rat. The study protocol was designed to minimize the number of animals needed to achieve the study goals. Bisoprolol, a beta blocker, and ramipril, an angiotensin-converting-enzyme inhibitor, are widely used in the treatment of heart failure [5]. They have an important role in improving survival and reducing the number of hospitalizations in patients with chronic heart failure [6]. In this condition, blood flow to the liver is frequently reduced [7] or the liver is congested [8], thus hepatic metabolism is affected in a way that modifies drug pharmacokinetics. Propranolol is a marker of a hepatic blood flow and midazolam is a marker of intrinsic hepatic metabolism and activity of the cytochrome enzyme CYP3A4 [9,10]. Together, their pharmacokinetics provide information on the hepatic metabolism of drugs with high and low extraction ratios and thus help to explain possible changes in the metabolism of other drugs. The technique of DBS sampling has gained popularity in recent years, and methods for measurement of bisoprolol, ramipril, propranolol or midazolam concentrations in human DBS have been developed [11–16]. Some of the methods measure combinations of drugs, but no method has yet been published for the specific combination of midazolam, propranolol, bisoprolol and ramiprilat. Moreover, with the exception of propranolol, no method has been established for measurement of the concentrations of these drugs in rat DBS samples [17]. We present the development and validation of an analytical method for simultaneous quantification of four marker drugs in rat DBS: midazolam, propranolol, bisoprolol and ramiprilat.

the Republic of Slovenia and the Veterinary Administration of the Republic of Slovenia. 2.2. Working solutions Stock solutions of bisoprolol, ramiprilat and propranolol were prepared by diluting an accurately weighed amount of each drug in methanol to yield drug concentrations of 1 mg/mL. Midazolam stock solution was purchased in the form of a 1 mg/mL methanol solution. The primary solution was prepared by mixing 25 ␮L of each stock solution and further diluting it to a final volume of 5 mL to obtain a mixture of all four drugs at a concentration of 5 mg/L for each one. Working solutions were prepared by diluting appropriate volumes of the primary solutions with a mixture of 25% methanol and 75% 0.1% formic acid. Stock solutions of internal standards (bisoprolol-D5, ramiprilatD5, racemic propranolol-D7 and [13 C6 ]-midazolam) were prepared by dissolving weighed amounts in methanol to yield concentrations of 1 mg/mL. An appropriate volume of each solution was diluted with methanol to yield a mixture of internal standards: bisoprololD5 500 ␮g/L, ramiprilat-D5 1000 ␮g/L, propranolol-D7 1000 ␮g/L and [13 C6 ]-midazolam 400 ␮g/L. All solutions were stored at −20 ◦ C and brought to room temperature before use. 2.3. Preparation of calibration standards and quality control samples Calibration standards and quality control (QC) samples were prepared by adding 20 ␮L of an appropriate working solution to 380 ␮L of blank rat blood, which was a mixture of blood obtained from three untreated rats. Samples were stirred for 15 min at 1300 rpm and 37 ◦ C, and then 20 ␮L of each blood sample was spotted directly onto Whatman protein saver cards using a volumetric pipette (Eppendorf Multipette Plus). The cards were allowed to dry for at least 2 h at room temperature before processing or storage. For each drug, the concentrations of the calibration standards were 0, 5, 10, 25, 50, 100, 175 and 250 ␮g/L; the QC concentrations were 15 (QCl ), 75 (QCm ), and 200 ␮g/L (QCh ).

2. Experimental

2.4. Extraction of dried blood samples

2.1. Chemicals and materials

Whole DBS were cut out of the Whatman cards with scissors and transferred to 1.5 mL Eppendorf microtubes. A 300 ␮L volume of solvent (methanol with10% ultrapure water and 0.1% formic acid) was added and vortex mixed for 1 min before the microtubes were put in an ultrasonic bath for 10 min. The tubes were then shaken on an orbital shaker for 45 min at 100 rpm (room temperature) before the addition of 35 ␮L of the internal standards mixture to each sample. Following centrifugation (10 min, 15,000 × g, 5 ◦ C), portions of supernatant (280 ␮L) were transferred to an Ostro plate and filtered to a 96-well collection plate for liquid chromatography–tandem mass spectrometry analysis.

Bisoprolol fumarate and ramiprilat were purchased from Sequoia Research Products Ltd., Pangbourne, UK. Propranolol hydrochloride was obtained from Fluka–Sigma–Aldrich, Buchs, Switzerland, and midazolam from Cerilliant, Round Rock, TX, USA. For in vivo experiments, the same substances were used, except for midazolam (Midazolam Torrex parenteral solution 1 mg/mL, Chiesi Pharmaceuticals GmbH, Austria). The internal standards bisoprolol-D5, ramiprilat-D5 and racemic propranololD7 were purchased from Santa Cruz Biotechnology, CA, USA. [13 C6 ]-midazolam was provided by Alsachim, Illkirch Graffenstaden, France. Ultrapure water was obtained using the Milli-Q Advantage A10 Ultrapure Water Purification System (Millipore Corporation, Bedford, MA, USA). Methanol Chromasol V® was purchased from Sigma–Aldrich, Steinheim, Germany. Formic acid 98–100% Suprapur® was obtained from Merck KGAa, Darmstadt, Germany. Whatman 903 Protein Saver Cards (Whatman, UK) were used for spotting the blood samples. Spots were manually excised with scissors to ensure that the whole spot was included in the analysis. OstroTM (Waters, USA) 96-well plates were used for the final removal of phospholipids from samples. The samples of whole blood of Wistar Han rats with K2-EDTA anticoagulant were obtained in accordance with the law on the protection of animals in

2.5. Chromatographic and mass spectrometry conditions After extraction, centrifugation and purification, the DBS samples were analyzed using an Agilent 1290 Infinity liquid chromatography system coupled to a 6460 triple quadrupole mass spectrometer (Agilent Technologies, Santa Clara, USA). The injection volume was 1 ␮L and the sample needle was washed with 80% methanol for 10 s after each injection. Chromatographic separation was achieved on a Kinetex C18 50 mm × 2.1 mm column with 2.6 ␮m particles, guarded by a C18 guard column (Phenomenex, Torrance, USA) at 50 ◦ C. The mobile phase consisted of 0.1% formic acid in Milli-Q water (A) and 100% acetonitrile (B); the flow rate was set at 0.5 mL/min with the following gradient (time, % of mobile

K. Cvan Trobec et al. / J. Chromatogr. B 958 (2014) 29–35

31

Table 1 Multiple reaction monitoring parameters for quantification of analytes in DBS extracts. Analyte

Precursor ion [m/z]

MS1 resolution [amu]

Product ion [m/z]

MS2 resolution [amu]

Fragmentor [V]

Collision energy [eV]

Bisoprolol Bisoprolol-D5 Midazolam [13 C6 ]-midazolam Propranolol Propranolol-D7 Ramiprilat Ramiprilat-D5

326.2 331.2 326.1 332.1 260.2 267.2 389.2 394.2

1.2 1.2 2.5 2.5 2.5 2.5 2.5 2.5

116.1 121.1 291.1 297.1 116.1 123.1 206.1 211.3

1.2 1.2 2.5 2.5 2.5 2.5 2.5 2.5

150 150 170 170 124 124 110 110

13 13 25 25 13 13 13 13

phase B): (0.25 min, 3%), (0.5 min, 9%), (1 min, 30%), (1.25 min, 50%), (1.7 min, 90%), (1.75 min, 3%). The total run time was 2.5 min. Analytes were ionized in JetStreamTM ESI operated in positive mode with the following parameters: drying gas temperature 275 ◦ C, drying gas flow 5 L/min, nebulizer 45 psi, sheath gas temperature 320 ◦ C, sheath gas flow 11 L/min, capillary entrance voltage 4000 V, nozzle voltage 1000 V. Column effluent was directed to mass spectrometry from 1.4 min through 2.0 min. Multiple reaction monitoring parameters are shown in Table 1. The dwell time for each analyte was 20 ms and the delta electron multiplier voltage was 200 V. MassHunter Workstation software B.04.01 (Agilent Technologies, Santa Clara, USA) was used for instrument control, data acquisition and quantification. 2.6. Validation of method The recommendations published by the Food and Drug Administration (FDA) [18] and European Medicines Agency (EMA) [19] were used as the basis for validation of the method. 2.6.1. Selectivity The selectivity of the developed method was determined by analyzing blank DBS samples from five untreated rats. The samples were extracted according to the procedure described in Section 2.4. Blank sample chromatograms were compared with chromatograms of the lower limit of quantification (LLOQ). The average response of LLOQ samples at the retention time of each analyte was required to be at least five times greater than the average blank sample response. 2.6.2. Accuracy and precision For each compound, the intraday and interday accuracy and precision were determined by analyzing QCl , QCm and QCh samples. The QCs were assayed in seven replicates on three consecutive days. Intraday accuracy and precision were estimated on a single day using seven replicates at each QC level; interday accuracy and precision were calculated using seven replicates at each QC concentration over three consecutive days. The accuracy was expressed as the bias or relative error and precision as the relative standard error (RSD). A relative error and RSD ≤ 15% at all concentrations (with the exception of LLOQ) was considered acceptable according to FDA and EMA guidelines. 2.6.3. Linearity and lower limit of quantification Calibration standards were prepared in replicate (N = 7). A calibration curve for each target analyte was obtained by plotting the ratio of analyte/internal standard peak area against nominal concentration. A 1/x2 weighted linear regression was applied. The LLOQs were determined for all four compounds; LLOQ was defined as the concentration where bias and RSD was ≤20%. The backcalculated concentrations of the calibration standards other than LLOQ should not have exceeded a bias of ±15% from their nominal concentration.

2.6.4. Absolute matrix effect and recovery Absolute matrix effect (ME) was determined according to Matuszewski et al. [20]. For the evaluation of ME and recovery, blank blood from five untreated rats was used. The following QCl , QCm and QCh samples were prepared in triplicate: (A) neat solution QC samples prepared in mobile phase, (B) blank QC samples spiked post-extraction with both the analyte and internal standards, and (C) matrix blanks spiked pre-extraction with both the analyte and the internal standards. The absolute ME was calculated as the peak area ratio of sample A and B: ME (%) = response B/response A × 100%. The RSD of determined absolute ME from five individual animals was required to be