Sequential hollow-fiber liquid phase microextraction for the determination of rosiglitazone and metformin hydrochloride (anti-diabetic drugs) in biological fluids

Talanta 131 (2015) 590–596

Contents lists available at ScienceDirect

Talanta journal homepage: www.elsevier.com/locate/talanta

Sequential hollow-fiber liquid phase microextraction for the determination of rosiglitazone and metformin hydrochloride (anti-diabetic drugs) in biological fluids Gazala Mohamed Ben-Hander, Ahmad Makahleh n, Bahruddin Saad nn, Muhammad Idiris Saleh, Kek Wan Cheng School of Chemical Sciences, Universiti Sains Malaysia, 11800 Minden, Penang, Malaysia

art ic l e i nf o

a b s t r a c t

Article history: Received 1 July 2014 Received in revised form 9 August 2014 Accepted 12 August 2014 Available online 20 August 2014

A new analytical method for the simultaneous determination of the antidiabetic drugs rosiglitazone (ROS) and metformin hydrochloride (MH) with marked differences in their affinity towards organic solvents (log P of 2.4 and
1.43, respectively) was developed. Prior to the HPLC separation, the drugs were subjected to a sequential hollow fiber liquid phase microextraction (HF-LPME) procedure. Two sequential HF-LPME approaches were considered, the preferred one involves the use of two vials containing solution mixtures for the extraction of ROS (vial 1) and MH (vial 2), respectively, but using the same fiber and acceptor phase. Important parameters that affect the extraction efficiency such as extracting solvent, donor phase conditions, HCl concentration, agitation, extraction time, addition of salt, etc. were studied. Under the optimum conditions, good enrichment factors (EF, 471 and 86.6 for ROS and MH, respectively) were achieved. Calibration curves were linear over the range 1–500 (r2 ¼0.998) and 5–2500 ng mL
1 (r2 ¼ 0.999) for ROS and MH, respectively. The relative standard deviation values (RSD%) for six replicates were below 8.4%. Detection and quantitation limits based on S/N ratio of 3 and 10 were 0.12, 1.0 and 0.36, 3.0 ng mL
1 for ROS and MH, respectively. The proposed method is simple, sensitive and opens up new opportunities for the microextraction of analytes with contrasting properties. & 2014 Elsevier B.V. All rights reserved.

Keywords: Sequential HF-LPME HPLC–UV Metformin hydrochloride Rosiglitazone Biological fluids

1. Introduction Treatment of diabetes mellitus using monotherapy with an oral anti-diabetic agent is insufficient to reach the target glycaemic goals in many patients, thus multi-drugs are necessary to achieve adequate control and satisfactory blood glucose levels. The combination of biguanides and thiazolidinedione derivatives is commonly used in clinical practice [1]. Metformin hydrochloride (MH) (biguanide class, Fig. 1), chemically [1,1-dimethylbiguanidehydrochloride], is an oral biguanide antihyperglycemic drug which is used to improve the insulin sensitivity, inhibits hepatic gluconeogenesis and reduces hepatic glucose production in patients that suffer from type 2 diabetes mellitus (T2DM) [2,3]. Rosiglitazone (ROS), (thiazolidinedione class, Fig. 1), chemically [[( 7)-5[4-[2-[N-methyl-N (2-pyridyl) amino] ethoxy] benzyl]-2,4-dione thiazolidine], is a drug for the treatment of T2DM which works by n

Corresponding author. Tel.: þ 60 4 653 6018; fax: þ60 4 657 4857. Corresponding author. Tel.: þ 60 4 653 2544; fax: þ 60 4 656 9869. E-mail addresses: [email protected] (A. Makahleh), [email protected] (B. Saad). nn

http://dx.doi.org/10.1016/j.talanta.2014.08.037 0039-9140/& 2014 Elsevier B.V. All rights reserved.

increasing the insulin sensitivity in the target tissues, as well as decreasing hepatic gluconeogenesis [4,5]. A combination of MH and ROS was found to be better in the treatment of T2DM compared to single-agent therapy alone due to its high effect on lowering blood glucose [6,7] and improving beta-cell function [8]. Furthermore, the combination tablet formulation is advantageous in terms of its convenience and patient compliance [9]. Several methods for the determination of ROS and MH either individually or simultaneously have been reported. High performance liquid chromatography with ultraviolet detection (HPLC– UV) [6–17] is the most commonly used method for the analysis of ROS and MH. However, high limit of quantitation (LOQ Z 20 ng mL
1) was observed when UV detection was used [8,10, 11,17]. Alternatively, HPLC with fluorescence detection (FL) [18,19] or tandem mass spectrometry (MS/MS) [2–4,20–27] was used. Although FL gives better sensitivity compared to UV detection, but the separation was rather long (Z15 min) [18,19]. LC–MS/MS is an efficient analysis tool providing low quantitation limits (Z1 ng mL
1) [20–23], short run time, improved sensitivity and selectivity, but it is costly and the equipment is not always available in clinical laboratories. The use of gas chromatography

G.M. Ben-Hander et al. / Talanta 131 (2015) 590–596

NH

NH

C H2 N

C N H

N

CH3

. HCl

CH3

O CH3 N

NH .

CO2H

HC

S

N

HC

CO2H

O O

Analyte

pKa

Log P

MH

12.4

-1.43

ROS

6.1, 6.8

2.4

Fig. 1. Chemical structures, pKa and log P values of (a) metformin hydrochloride (MH) and (b) rosiglitazone (ROS).

coupled with nitrogen [28], flame ionization [29] or mass spectrometry (MS) detectors [30] for the analysis of MH was described. Capillary zone electrophoresis (CE) with UV detection [1,31–33] and MS [34] have also been reported. Liquid–liquid extraction (LLE) [6,10,19,20,27], solid-phase extraction (SPE) [2,4,9,13,14,18,33] and protein precipitation [3,17, 21,22,24–26,34] are the most widely used sample preparation technique for the analysis of ROS and MH in biological fluids. However, these techniques have many disadvantages as they usually require large volumes of high-purity solvents, multi-step and long extraction time which lead to analyte losses. To overcome these problems, microextraction techniques such as hollow fiber liquid phase microextraction (HF-LPME) has been used for the individual analysis of ROS [11,12] or MH [16] in biological fluids. The main advantages of the HF-LPME technique are fast, simple, inexpensive, low consumption of organic toxic solvents (only microliter volumes), no carry over due to the single use of the fiber and high enrichment factor. Also, the clean-up and preconcentration of the analytes are done in a single step due to the small pore size of the hollow fiber membrane which act as a microfilter that eliminates interfering macromolecules and produce clean extracts that are suitable for direct instrumental analyses. The simultaneous microextraction of ROS and MH is analytically challenging, if not impossible, due to the marked differences in their physical properties. Pertaining to extraction are the pKa and log P values of these drugs. MH is readily soluble in water, highly polar (log P ¼
1.43) and is strongly basic (pKa ¼12.4) while log P and pKa of ROS are 2.4 and 6.1, 6.8, respectively. Furthermore, MH is non-chromophoric. It is rationalized that in the absence of a simultaneous method for the microextraction, a sequential microextraction approach that extracts one drug, followed by the next one would also be worth considering. Since these two drugs had been individually extracted using the HF-LPME technique, the present studies is aimed at modifying and integrating these work for the sequential approach, that will eventually lead to the simultaneous HPLC determination.

2. Experimental 2.1. Chemicals and reagents Metformin HCl (MH) and rosiglitazone maleate (ROS) reference standards were kindly donated by Hikma Pharmaceuticals (Amman,

591

Jordan). Acetonitrile (HPLC-grade; 99.99%) was purchased from Fisher Scientific (Milwaukee, WI, USA). Methanol (HPLC-grade; Z99.96%), hydrochloric acid (37%, w/w) were purchased from Merck (Darmstadt, Germany). n-Decane (99.0%) and n-tridecane (99.0%) were obtained from Acros Organics (Geel, Belgium). Pentafluorobenzoyl chloride (99.0%), Sodium hydroxide (Z98.0%), dihexyl ether (97.0%), n-heptane (99.0%), n-hexadecane (99.0%) and nitrobenzene (Z99.0) were purchased from Sigma-Aldrich (St. Louis, MO, USA). 1-Heptanol (Z99.9%) and 1-octanol (Z99.5%) were purchased from Fluka (Buchs, Switzerland). Phosphoric acid (85%) was purchased from Univar (Ingleburn, Australia). Ultrapure water (resistivity, 18.2 MΩ cm
1) was produced by a Milli-Q system (Millipore, MA, USA). Blank plasma sample was kindly donated by Centre for Drug Research, Universiti Sains Malaysia, Penang. Human urine sample was obtained from a healthy student volunteer. Derivatizing solution was prepared by dilution 10 mg of Pentafluorobenzoyl chloride (PFBC) in 1 mL acetonitrile and stored at 4 1C until used. 2.2. Instrumentation Chromatographic analyses were performed using a Hitachi LC6200 intelligent pump (Tokyo, Japan) equipped with a HewlettPackard 1050 UV detector (Waldbronn, Germany). Sample injection was performed via a Rheodyne 7125 injection valve (Cotati, CA, USA) with a 5 mL loop. A PowerChrom data acquisition was obtained from eDAQ (Denistone East, Australia) and performed with PowerChrom software (version 2.7.2) to record and analyze the chromatographic data. The separation was obtained using a ODS Hypersil C18 column (250  4.6 mm, 5 mm). The mobile phase composition was a mixture of acetonitrile and 10 mM sodium phosphate buffer (pH 4.0) (60:40, v/v). The elution was performed under isocratic mode at a flow rate of 1.0 mL min
1. The UV detection wavelength was set at 230 nm. Prior to the analysis, the mobile phase was filtered through nylon membrane filter (0.45 mm) from Agilent Technologies (Waldbronn, Germany) and degassed by ultrasonic bath for 15 min. For UVscanning purpose, a Lambda 35 UV/vis system from Perkin Elmer (Waltham, MA, USA) was used. The extraction was performed using a 25 mL micro-syringe with a blunt needle tip (model 702SNR) and it was purchased from Hamilton (Reno, NV, USA). A multi-hotplate stirrer from DAIHAN scientific (Seoul, South Korea) was used for the stirring through the extraction process. 2.3. Preparation of stock standard solutions ROS stock solution (1000 mg mL
1) was prepared by dissolving the desired amount in acetonitrile, while MH stock solution (2000 mg mL
1) was prepared in water. A mixture solution of ROS and MH (200 and 1000 mg mL
1, respectively) was prepared by a proper dilution of the stock solutions in water and stored at 4 1C until use. Working standard solution was prepared daily by diluting the standard mixture in water as described in Section 2.5. 2.4. Minimizing the matrix effect of plasma and urine In order to reduce the matrix effect of plasma sample, the following pretreatment steps have been conducted. 200 mL HCl (0.05 M) was added to the plasma sample (2 mL) spiked with standard mixture at the desired concentration. The sample mixture was vortex-mixed thoroughly for 30 s. The protein precipitation was accomplished by addition of acetonitrile (3 mL) and then the mixture was centrifuged at (1900 rpm) for 15 min. An aliquot of supernatant was collected and evaporated to dryness at 40 1C under gentle nitrogen stream. The dried residue was reconstituted with water as described in Section 2.5 for sequential HF-LPME analysis. In order to reduce the matrix effects (e.g., albumins,

592

G.M. Ben-Hander et al. / Talanta 131 (2015) 590–596

sugars, urea, etc.) in the urine samples, urine sample spiked with standard mixture was diluted 1:1 with water.

2.5. Sequential HF-LPME procedure Two sequential HF-LPME procedures were considered. In the first procedure (Fig. 2(a)), two different vials but sharing the same HF segment (4 cm) and acceptor phase (10 mL of 0.1 M HCl) were used. Each vial (12 mL) contains 10 mL of working standard or sample solutions. The first vial was dedicated for the extraction of ROS, it contained samples that had been diluted to 10 mL with water (pH adjusted to 9.0 using 0.05 M NaOH solution), while the second vial for the extraction of MH also contained samples that had been mixed with 600 mL NaOH (4.0 M) and 100 mL derivatizing solution (10 mg mL
1) and diluted to 10 mL using water. A magnetic stirring bar (15  5 mm2) was placed in each vial. Acceptor phase (10 mL) was withdrawn using a micro-syringe and the syringe needle was inserted into the HF segment. The HF was bent to a U-shape and the assembly was immersed 10 s in dihexyl ether (as organic solvent) to impregnate the solvent into the porous wall of the fiber. It was next soaked in water for 5 s in order to wash the extra organic solvent. Subsequently, the HF was

placed immediately in the first vial. The acceptor phase (AP) was completely injected into the lumen of the HF and the sample solution was agitated at 300 rpm at room temperature. After 30 min, the fiber was placed in the second vial that was heated at 70 1C and agitated at the same speed as that for the first vial (300 rpm) for 30 min. At the end of the extraction time, the extract (5 mL) was carefully withdrawn into the syringe and the HF was discarded. Finally, the extract was directly injected into the HPLC system. In the second procedure (Fig. 2(b)), a vial that contained a mixtures of ROS and MH (10 mL) was used. The pH of the solution mixture (donor phase) was adjusted to 9.0 and the HF-LPME extraction for ROS was performed as the first approach. After 30 min, PFCB and NaOH (4.0 M) were added to the same vial and the mixture was heated at 70 1C for 30 min. Finally, the acceptor phase was withdrawn from the lumen of the HF and injected into the HPLC unit.

3. Results and discussion Initial chromatographic conditions used were adopted from our previous work [16], i.e., hypersil ODS column with mobile phase

Fig. 2. First (a) and second (b) procedure of sequential HF-LPME. Where: AP, acceptor phase; HF, hollow fiber; V, vial; and RT, room temperature.

G.M. Ben-Hander et al. / Talanta 131 (2015) 590–596

593

composition of acetonitrile: 10 mM phosphate buffer (pH 4.0) (60:40; v/v) at flow rate of 1 mL min
1. Under these conditions, peak broadening of ROS was observed. Different compositions of mobile phase were investigated by varying the acetonitrile ratio (55–65%). An improvement in the ROS peak shape was observed as the ratio of acetonitrile was decreased. As a compromise between peak shape, resolution and run time, 60% acetonitrile was chosen. The effect of buffer pH (3.5–5.0) was also explored. Although low pH enhanced the peak shapes but a decrease in resolution between the peaks was observed. The best peak shape and resolution was observed at pH 4.0. Therefore, a mixture of 40% of 10 mM sodium phosphate buffer (pH 4.0) and 60% acetonitrile at flow rate 1.0 mL min
1 was selected as the optimum mobile phase composition. Fig. 4. Typical chromatograms of MH and ROS after subjecting to sequential HFLPME using (a) approach 1, and (b) approach 2. Please refer to text for experimental details.

1.20 1.00 ROS in water

Absorbance

0.80

3M KOH

0.60

4M KOH 5M KOH

0.40

3M NaOH

0.20

4M NaOH

0.00 200 220 240 260 280 300 320 340 360 380 400 -0.20

5M NaOH

Wavelength (nm) Fig. 3. UV scans of ROS prepared in different concentrations of NaOH and KOH.

120 100 Enrichment factor

ROS and MH are basic compounds with pKa values 6.8 and 12.1, respectively. However, their affinity towards the organic solvents is markedly different (log P, 2.4 and
1.43, respectively) [12,16]. In order to enhance the extraction of MH towards organic solvents, derivatization is required. We have previously reported an in-situ HF-LPME method for the extraction of MH in biological fluids where MH was derivatized using PFBC in the presence of high concentration of NaOH (4.0 M) [16]. However, the use of too concentrated base in the present work may hinder the extraction of ROS [12]. Thus, preliminary studies were conducted using UV–vis spectroscopy to study the stability of ROS in concentrated bases. Thus ROS in different concentrations of NaOH and KOH (3.0–5.0 M) was prepared, and their spectrum was recorded (Fig. 3). A big shift in the maximum wavelength of ROS that was prepared in concentrated NaOH or KOH was found. This is probably due to the decomposition of the diimide group under strongly basic conditions [35,12]. Further confirmation was carried out by mixing the standard mixture with PFBC and NaOH (4.0 M), then subjected to the previously reported in-situ HF-LPME conditions [16]. Only MH peak was observed. Sequential HF-LPME involving first the extraction of ROS for 30 min, followed by the addition of PFBC and NaOH (4.0 M) to the sample for the extraction of MH in another 30 min was performed. It was observed that the two analytes were extracted when relatively large volumes of NaOH ( Z2 mL) were used. Good extraction for ROS was observed but not for MH (EF, 7.6), probably due to the dilution effect of the donor phase. As an alternative, sequential HF-LPME procedure was performed using the same HF and acceptor phase to extract ROS and MH in two different vials, i.e., first vial with pH adjusted to 9.0 for extraction of ROS, while the second contains PFBC and NaOH (4.0 M) for the extraction of MH. This procedure was preferred as better extraction was found (EF, 86.6 and 100 for MH and ROS, respectively) (Fig. 4). Thus,

80 60 MH

40

ROS

20 0

Organic solvent

500

400 Enrichment factor

3.1. Optimization of sequential HF-LPME parameters

300 ROS

200

MH

100

0 0.03

0.05

0.08

0.10

HCl (M)

Fig. 5. Effect of organic solvent (A), concentration of acceptor phase (HCl) (B) on the enrichment factors (n¼ 3).

several extraction parameters that influence the extraction efficiency (i.e., selection of organic solvent, donor phase conditions, acceptor phase (HCl) concentration, stirring speed, temperature, time and salt addition) of this procedure were investigated and optimized. The selection of extraction organic solvent is important in HF-LPME technique [36]. The organic solvent should be insoluble in water, have good affinity for the fiber and analytes, and have low volatility to prevent solvent losses during the extraction process [37,38]. In view of these factors, eight organic solvents, i.e., dihexyl ether, nitrobenzene, 1-heptanol, 1-octanol, n-heptane, n-decane, n-tridecane and n-hexadecane were tested. Dihexyl ether produced the highest extraction for both ROS and MH (Fig. 5(A)). The selected organic solvent was in agreement with

594

G.M. Ben-Hander et al. / Talanta 131 (2015) 590–596

our previous reports for the HF-LPME extraction of ROS and MH [12,16]. Thus, it was selected for subsequent studies. The effect of the donor phase conditions (e.g., pH (for vial 1) and NaOH, PFBC volume and concentration (for vial 2)) on the extraction efficiency were investigated. The influence of pH (8.0–10.0) on the extraction of ROS (first vial) was found to increase as the pH is increased from 8.0 to 9.0, but drops abruptly thereafter. The significant drop in enrichment factor as pH is increased is probably due to deprotonation, decomposition and/or precipitation of ROS at high pH [12]. The optimum volume and concentration of NaOH and PFBC for the extraction of MH (vial 2) were 4.0 M NaOH (600 mL) and 10 mg mL
1 PFBC (100 mL) and these were in agreement with our previous report [16]. Due to the basic properties for both analytes, acidic acceptor phase is required to ionize them to enhance their solubilities and prevent their back-extraction to the organic phase [12,16]. The effect of HCl concentration (0.025–0.100 M) on the extraction efficiency for ROS and MH was studied. It was observed that the enrichment factor increased as the concentration of HCl increased (Fig. 5(B)). Higher concentration of HCl was avoided in order to prevent the possible corrosion of the injector and other HPLC parts [12]. Therefore, 0.1 M HCl was selected as the optimum acceptor phase for the subsequent experiments.

Table 1 Optimum conditions of the sequential HF-LPME method. Extraction parameters

ROS (vial 1)

MH (vial 2)

Organic solvent Acceptor phase Donor phase

Dihexyl ether 0.1 M HCl (10 μL) pH 9.00

Stirring speed Extraction temperature Extraction time (min) Salt addition Enrichment factor

300 rpm Ambient 30 — 471

Agitation of the sample solution increases the mass transfer of the target compounds and reduces the extraction time required to reach thermodynamic equilibrium [36,39]. Therefore, the extraction can be accelerated by stirring the sample solution. Different stirring speeds (200–700 rpm) were tested to determine the optimum stirring speed for the extractions of ROS and MH. The highest enrichment factors (EF) were obtained at stirring speeds of 600 and 300 rpm for ROS and MH, respectively. Further increasing of stirring speed may limit the mass transfer due to the dissolution of the membrane liquid and formation of air bubbles in the donor phase that is attached to the fiber surface, especially when heating is used [40–42]. As there is no significant difference in the EF of ROS over the range 300–600 rpm, 300 rpm was selected. In order to balance the extraction efficiency of both analytes, the influence of extraction time was examined. The extraction time of each vial was varied from 20 to 35 min, thus total extraction time ranged from 50 to 65 min. It was observed that the extraction efficiency increased as the extraction time increased. Total extraction time of 60 min provided the highest extraction efficiency for both analytes. Longer extraction time was avoided in order to prevent the loss of organic solvent. The extraction of ROS was found to increase as the extraction time increased. 35 min resulted in the highest extraction. However, the longest extraction time for ROS gave the lowest extraction efficiency of MH. Thus, 30 min extraction time was chosen. Similarly, the EF for MH (vial 2) was found to increase as the extraction time was increased up to 30 min, and decreased thereafter. This is probably due to the possible formation of air bubbles on the fiber and evaporation of the organic solvent especially under heating conditions [16]. Therefore, 30 min was used for both analytes.

600 mL NaOH (4 M) 100 μL PFBC (10 mg mL
1) 70 1C 30 86.6

Table 2 Repeatability (%RSD) for ROS and MH after subjected to the sequential HF-LPME– HPLC–UV method (n¼6). %RSD Spiked level (ng mL
1)

ROS MH

50

500

1000

4.1 7.7

7.3 8.4

8.1 7.5

Table 3 Recoveries obtained by spiking plasma and urine samples with standard mixture and subjected to sequential HF-LPME–HPLC–UV method (n¼ 6). ROS

MH

Cadded Cfound Recovery (ng mL
1) (ng mL
1) (% 7 SD)

Cadded Cfound Recovery (ng mL
1) (ng mL
1) (% 7 SD)

Plasma 200 100 10

200 98.1 9.96

1007 4.3 98.17 8.5 99.6 7 7.5

1000 500 50

978 481 48.5

97.8 7 11 96.2 7 13 97.0 7 4.5

Urine

202 108 9.91

1017 11 1087 8.3 99.17 7.3

1000 500 50

941 491 49.1

94.17 5.0 98.2 7 10 98.2 7 3.5

200 100 10

Fig. 6. Typical chromatograms of plasma (A) and urine (B) samples after subjected to the sequential HF-LPME procedure. Spiked concentration; 100 and 500 ng mL
1 for ROS and MH, respectively.

G.M. Ben-Hander et al. / Talanta 131 (2015) 590–596

It has been reported that the addition of salt to the sample solution will increase the ionic strength of the DP and decrease the solubility of the organic analytes in the aqueous solution [43,44]. In order to investigate the influence of ionic strength on the extraction, experiments were performed by adding different amounts of NaCl (0–15%, w/v) into the sample solutions. As the concentration of NaCl decreased, the extraction efficiencies of the analytes decreased. This is probably due to the increase of the viscosity of the DP solutions which reduce the diffusion rate of the target analytes from DP solutions to the extraction solvent [12]. Therefore, no NaCl was used in this study. The optimum conditions of the selected sequential HF-LPME procedure for the extraction of ROS and MH are summarized in Table 1. Under these conditions, EFs of 471 and 86.6 were obtained for ROS and MH, respectively. The EF for ROS was higher than those previously reported [11,12] but it was lower for MH.

3.2. Validation of the method Under the optimized conditions, linearity of the proposed sequential HF-LPME method was investigated over the range 1–500 and 5–2500 ng mL
1 for ROS and MH, respectively. The method was found to be linear over the studied range with regression equations and correlation coefficients (r2) of y¼ 1.4772x (r2 ¼0.998) and y¼0.029x (r2 ¼0.999) for ROS and MH, respectively. The limits of detection (LODs) and quantitation (LOQs) were calculated based on the signal-to-noise ratio (S/N) of 3 and 10, respectively. The LOD was

595

0.12 and 1.0 ng mL
1, while the LOQ was 0.36 and 3.0 ng mL
1 for ROS and MH, respectively. The obtained LODs and LOQs results were better than the reported methods using LLE with LC–MS/MS, HPLC–FL or HPLC–UV [6,19,27], SPE with LC–MS/MS, HPLC–UV or HPLC–FL [2,14,18], protein precipitation (PP) followed by HPLC–UV, LC–MS/MS or CE–UV [17,24,31], solid–liquid extraction (SLE) with LC–MS/MS [23] and HF-LPME with HPLC–UV or CE–UV [11,12]. However, the results were comparable with the reported method using SPE with LC–MS/ MS [4] but higher than the reported in situ HF-LPME with HPLC–UV [16]. The relative standard deviation (%RSD) values for six replicate extractions at three different concentration levels (10, 100, 200 ng mL
1 for ROS and 50, 500, 1000 ng mL
1 for MH) were below 8.4% (Table 2), indicating the good precision of the method.

3.3. Recovery in biological fluid samples The recovery of the developed method was studied by the determination of ROS and MH in plasma and urine samples that were spiked at three different concentration levels (10, 100, 200 and 50, 500, 1000 ng mL
1 for ROS and MH, respectively). Prior to the sequential HF-LPME step, the spiked samples were subjected to a pre-treatment procedure (Section 2.4) in order to minimize the protein binding of ROS and reduce possible matrix effect [12]. Good recoveries (94.1–108%) were obtained for both analytes (ROS and MH) in plasma and urine samples (Table 3). These results were better than the previous reported recoveries using PP with LC–MS/MS [22,24] CE with ESI-MS [34] or CE–UV [31], SPE

Table 4 Comparison between the proposed sequential HF-LPME–HPLC–UV method with previously reported methods. Analyte

Instrumentation

Detection

Sample preparation

Type of sample

Linearity (ng mL
1)

LOD (ng m
1)

LOQ (ng mL
1)

Recovery (%)

Reference

ROS ROS ROS

LCa HPLCd HPLC

MS/MSb UVe UV

SPEc LLEf HF-LPMEg

0.1–2000 5–1250 50–6000

0.1 – –

0.3 5.0 49

91.2–99.3 76.0–81.0 63.2–81.3

4 6 11

ROS

HPLC

UV

HF-LPME

1–500

0.18

0.56

ROS

CEh

UV

HF-LPME

ROS ROS

HPLC

FLi

SPE

HPLC

FL

LLE

ROS MH MH MH

LC LC LC HPLC

MS/MS MS/MS MS/MS UV

SLEj PPk SPE SPE

MH

HPLC

UV

MH MH ROS MH ROS MH ROS MH

LC CE

MS/MS UV

LLE PP

Urine Plasma Rat liver Urine Plasma Urine Plasma Plasma Sheep plasma Amniotic fluid Plasma Plasma Plasma Plasma Plasma Urine Plasma Plasma

HPLC

UV

PP

Plasma

LC

MS/MS

PP

Plasma

CE

ESI-MS

In situ HF-LPME

PP

Serum

Sequential HF-LPME

Plasma Urine Plasma Urine

ROS HPLC MH a

UV

Liquid chromatography. Tandem mass spectrometry. c Solid-phase extraction. d High performance Liquid Chromatography. e Ultraviolet detection. f Liquid–liquid extraction. g Hollow-fiber liquid phase microextraction. h Capillary electrophoresis. i Fluorescence detecon. j Supported liquid extraction. k Protein precipitation. b

5–500

2.8

5.0

1–1000

0.25 –

1.0

1–500 20–2500 10–1000 50–2000

– – – 3.0

1.0 20 10 5.0

1–1000

0.56

1.7

10–1500 250–3500 100–2500 250–2500 1.05–263.5 4.04–5050 4–800 2–400

– 100 50 100 – 4.4 2.1

10 250 100 250 1.1 4.0 14.6 7.1

1–500

0.12

0.36

5–2500

1.0

3.0

2.5–250

2.5

83.0–98.0 87.0–105 81.0–101 84.0–102 94.0–101 86.6–92.4 85.4–105 78.6–80.7 69.3–71.2 67.0 97.9–101 82.7–99.0 97.0–105 Z 91.0 58.9–80.2 98.2–101 91.4–97.6 80.5–97.9 87.0–93.6 93.0 85.5 98.1–100 99.1–108 96.2–97.8 94.1–98.2

12 12 18 19 23 24 2 14 16 27 31 17 22 34

Present work

596

G.M. Ben-Hander et al. / Talanta 131 (2015) 590–596

with LC–MS/MS [2], LLE with HPLC–UV or HPLC–FL [6,19] and HF-LPME with LC–UV [11,12,16] or CE–UV [12] but lower or comparable with the reported methods using SPE with LC–MS/ MS, HPLC–FL or HPLC–UV [4,14,18]. The data obtained suggest that the proposed sequential HF-LPME procedure could be applied for therapeutic drug monitoring of ROS and MH. Typical chromatograms of the blank samples do not show any interfering peak at the retention time of ROS and MH (Fig. 6(A, B)). A comparison between the proposed sequential HF-LPME with the previously reported method is shown in Table 4. 4. Conclusions For the first time, a sequential HF-LPME of two drugs with contrasting log P values is demonstrated using MH and ROS as model compounds. Microextracts of plasma and urine samples were finally simultaneously determined using HPLC–UV. The developed method has various benefits such as simple, sensitive, low consumption of organic solvents, cost-effective, and exhibit excellent repeatability. In addition, good enrichment factors of 86.6 and 471 for MH and ROS, respectively, were achieved. The method was validated, can be used in therapeutic drug monitoring of patients undergoing treatment using a combination therapy of ROS and MH. The sequential HF-LPME approach can be extended for the extraction of analytes with different properties that require contrasting extraction conditions. Acknowledgments Financial support of the work by Universiti Sains Malaysia via Research University Grant (1001/PKIMIA/811201) and postdoctoral fellowship scheme to Ahmad Makahleh are gratefully acknowledged. One of us (Gazala Ben-Hander) thanked the Government of Libya for providing a postgraduate scholarship. References [1] C. Yardimcı, N. Özaltin, Anal. Chim. Acta 549 (2005) 88–95. [2] N. Koseki, H. Kawashita, M. Niina, Y. Nagae, N. Masuda, J. Pharm. Biomed. Anal. 36 (2005) 1063–1072. [3] M.A.S. Marques, A.D.S. Soares, O.W. Pinto, P.T.W. Barroso, D.P. Pinto, M. Ferreira-Filho, E. Werneck-Barroso, J. Chromatogr. B 852 (2007) 308–316. [4] C.-C. Chou, M.-R. Lee, F.-C. Cheng, D.-Y. Yang, J. Chromatogr. A 1097 (2005) 74–83. [5] P. Gomes, J. Sippel, A. Jablonski, M. Steppe, J. Pharm. Biomed. Anal. 36 (2004) 909–913.

[6] B.L. Kolte, B.B. Raut, A.A. Deo, M.A. Bagool, D.B. Shinde, J. Chromatogr. B 788 (2003) 37–44. [7] A.-M. Muxlow, S. Fowles, P. Russell, J. Chromatogr. B 752 (2001) 77–84. [8] C. Yardimci, N. Özaltin, Chromatographia 66 (2007) 589–593. [9] S. Aburuz, J. Millership, J. Mcelnay, J. Chromatogr. B 817 (2005) 277–286. [10] J.N. Jingar, S.J. Rajput, B. Dasandi, S. Rathnam, Chromatographia 67 (2008) 951–955. [11] L.A. Calixto, P.S. Bonato, J. Sep. Sci. 33 (2010) 2872–2880. [12] K.M. Al Azzam, A. Makahleh, B. Saad, S.M. Mansor, J. Chromatogr. A 1217 (2010) 3654–3659. [13] K. Tahara, A. Yonemoto, Y. Yoshiyama, T. Nakamura, M. Aizawa, Y. Fujita, T. Nishikawa, Biomed. Chromatogr. 20 (2006) 1200–1205. [14] S. Aburuz, J. Millership, J. Mcelnay, J. Chromatogr. B 798 (2003) 203–209. [15] B.L. Kolte, B.B. Raut, A.A. Deo, M.A. Bagool, D.B. Shinde, J. Chromatogr. Sci. 42 (2004) 70–73. [16] G.M. Ben-Hander, A. Makahleh, B. Saad, M.I. Saleh, J. Chromatogr. B 941 (2013) 123–130. [17] C. Yardimcı, N. Özaltın, A. Gürlek, Talanta 72 (2007) 1416–1422. [18] R.S. Pedersen, K. Brøsen, F. Nielsen, Chromatographia 62 (2005) 197–201. [19] M. Bazargan, A.K. Davey, B.S. Muhlhausler, J.L. Morrison, I.C. Mcmillen, D.J. R. Foster, J. Pharm. Biomed. Anal. 55 (2011) 360–365. [20] J. He, Y.F. Hu, L.F. Duan, Z.R. Tan, L.S. Wang, D. Wang, W. Zhang, Z. Li, J. Liu, J. H. Tu, Y.M. Yao, H.-H. Zhou, J. Pharm. Biomed. Anal. 43 (2007) 580–585. [21] L. Zhang, Y. Tian, Z. Zhang, Y. Chen, J. Chromatogr. B 854 (2007) 91–98. [22] L. Chen, Z. Zhou, M. Shen, A. Ma, J. Chromatogr. Sci. 49 (2011) 94–100. [23] G. Ơmaille, S.M. Pai, X. Tao, G.T.Douglas Jr., R.G. Jenkins, J. Pharm. Biomed. Anal. 48 (2008) 934–939. [24] H.N. Mistri, A.G. Jangid, P.S. Shrivastav, J. Pharm. Biomed. Anal. 45 (2007) 97–106. [25] W. Zhang, F. Han, H. Zhao, Z.J. Lin, Q.M. Huang, N. Weng, Biomed. Chromatogr. 26 (2012) 1163–1169. [26] C.-G. Ding, Z. Zhou, Q.-H. Ge, X.-J. Zhi, L.-L. Ma, Biomed. Chromatogr. 21 (2007) 132–138. [27] P. Sengupta, U. Bhaumik, A. Ghosh, A.K. Sarkar, B. Chatterjee, A. Bose, T.K. Pal, Chromatographia 69 (2009) 1243–1250. [28] J. Brohon, M. Noël, J. Chromatogr. 146 (1978) 148–151. [29] S.A. Majidano, M.Y. Khuhawar, Chromatographia 75 (2012) 1311–1317. [30] E. Uçaktürk, Anal. Methods 5 (2013) 4723–4730. [31] J.-Z. Song, H.-F. Chen, S.-J. Tian, Z.-P. Sun, J. Chromatogr. B 708 (1998) 277–283. [32] I.I. Hamdan, A.K.B. Jaber, A.M. Abushoffa, J. Pharm. Biomed. Anal. 53 (2010) 1254–1257. [33] E.P.C. Lai, S.Y. Feng, J. Chromatogr. B 843 (2006) 94–99. [34] J. Znaleziona, V. Maier, V. Ranc, J. Ševčík, J. Sep. Sci. 34 (2011) 1167–1173. [35] J. Mcmurry, Organic Chemistry, 6th ed., Thomson Learning Academic Resource Center, United States, 2004. [36] M. Mirzaei, H. Dinpanah, J. Chromatogr. B 879 (2011) 1870–1874. [37] J. Wang, Z. Du, W. Yu, S. Qu, J. Chromatogr. A 1247 (2012) 10–17. [38] H. Zhang, Z. Du, Y. Ji, M. Mei, Talanta 109 (2013) 177–184. [39] J. Zhou, P. Zeng, J.B. Sun, F.Q. Wang, Q. Zhang, J. Pharm. Biomed. Anal. 81–82 (2013) 27–33. [40] M. Hyder, J.A. Jönsson, J. Chromatogr. A 1249 (2012) 48–53. [41] L. Chimuka, T.A.M. Msagati, E. Cukrowska, H. Tutu, J. Chromatogr. A 1217 (2010) 2318–2325. [42] W. Liu, L. Zhang, L. Fan, Z. Lin, Y. Cai, Z. Wei, G. Chen, J. Chromatogr. A 1233 (2012) 1–7. [43] I.S. Román, M.L. Alonso, L. Bartolomé, R.M. Alonso, Talanta 100 (2012) 246–253. [44] S.-P. Huang, S.-D. Huang, J. Chromatogr. A 1135 (2006) 6–11.