Journal of Chromatography A, 1344 (2014) 83–90
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Evaluation of 13 C- and 2 H-labeled internal standards for the determination of amphetamines in biological samples, by reversed-phase ultra-high performance liquid chromatography–tandem mass spectrometry Thomas Berg a,∗ , Morten Karlsen b , Åse Marit Leere Øiestad a , Jon Eigill Johansen b , Huiling Liu b , Dag Helge Strand a a b
Division of Forensic Medicine and Drug Abuse Research, Norwegian Institute of Public Health, Oslo, Norway Chiron AS, Trondheim, Norway
a r t i c l e
i n f o
Article history: Received 13 January 2014 Received in revised form 4 April 2014 Accepted 7 April 2014 Available online 18 April 2014 Keywords: 13 C Stable isotope-labeled internal standard LC–MS/MS Isotope effect Deuterium Amphetamine
a b s t r a c t Stable isotope-labeled internal standards (SIL-ISs) are often used when applying liquid chromatography–tandem mass spectrometry (LC–MS/MS) to analyze for legal and illegal drugs. ISs labeled with 13 C, 15 N, and 18 O are expected to behave more closely to their corresponding unlabeled analytes, compared with that of the more classically used 2 H-labeled ISs. This study has investigated the behavior of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine, during sample preparation by liquid–liquid extraction and LC–MS/MS analyses. None or only minor differences in liquid–liquid extraction recoveries of amphetamine and the SIL-ISs were observed. The chromatographic resolution between amphetamine and the 2 H-labeled amphetamines increased with the number of 2 H-substitutes. For chromatographic studies we also included seven additional 13 C6 -amphetamines and their analytes. All the 13 C6 -labeled ISs were co-eluting with their analytes, both when a basic and when an acidic mobile phase were used. MS/MS analyses of amphetamine and its SIL-ISs showed that the ISs with the highest number of 2 H-substitutes required more energy for fragmentation in the collision cell compared with that of the ISs with a lower number. The findings, in this study, support those of previous studies, showing that 13 C-labeled ISs are superior to 2 H-labeled ISs, for analytical purposes. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Liquid chromatography–tandem mass spectrometry (LC–MS/MS) is commonly used for qualitative and quantitative analyses of drugs in human biological samples [1,2]. SIL-ISs, which are compounds where the atoms in a molecule are replaced by their stable isotopes, such as 2 H, 13 C, 15 N, and 18 O, are often used to improve these drug detections [3–5]. To avoid for the natural occurrence of analyte isotopes interfering with a SIL-IS it should be labeled with three or more isotopes [6]. Ideally, the SIL-IS behaves exactly as the drug of interest during the sample pre-treatment, the chromatographic separation, and the LC–MS/MS detection; however, due to isotope effects (IEs), their behaviors are often
∗ Corresponding author. Tel.: +47 92 60 21 52; fax: +47 22 38 32 33. E-mail address: [email protected] (T. Berg). http://dx.doi.org/10.1016/j.chroma.2014.04.020 0021-9673/© 2014 Elsevier B.V. All rights reserved.
not identical. IEs are defined as the effects on the rate (kinetic IEs), or on the equilibrium constants (thermodynamic IEs), of two systems that differ only in the isotopic composition of one or more of their otherwise chemically identical components [7]. Primary IEs arise when a bond to the isotope is formed or cleaved, while secondary IEs arise when the bond remains intact. The main cause of IEs is the difference in nuclear mass [8–10]. This difference is greater between hydrogen isotopes than between isotopes of other elements, and thereby explains why IEs usually are greater among the 2 H-labeled ISs than for the 13 C, 15 N, and 18 O-labeled ones [3,6,11–18]. For deuterated compounds, the 2 H C bond is shorter, stronger, and less polarizable than the 1 H C bond. The slightly smaller volumes of deuterated compounds compared with the volumes of their protonium isotopomers favors the former in polar phases [19,20]. In addition, the less polarizable 2 H C bond presumably generates weaker London dispersion forces [21]. El Tayar et al. found that when comparing the n-octanol/water partition
84
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
coefficient (P) of deuterated and non-deuterated aromates, the deuterated aromates had the lower P-values (P ≤ 0.05) [22]. Kovach and Quinn found that P-values of deuterated carbonyl compounds were reduced by 1–2% in the organic phase [23]. Other studies have shown how IEs may contribute to changes in pKa values [24–26]. Filer’s paper from 1999 is a thorough review on IEs influence on different chromatographic techniques [27]. When performing LC–MS/MS analyses of drugs in biological samples, it is important for the IS to co-elute with the compound of interest, to be able to correct for possible ion suppression/enhancement effects, and thereby to avoid false results [28–31]. However, for reversed-phase (RP) LC separations, it is well known that the 2 H-labeled ISs often elute slightly earlier than their corresponding analytes [5,15,18,20,32,33]. This problem has become even more relevant during recent years, due to the highly efficient chromatographic separations obtained by ultra-high performance liquid chromatography (UHPLC) instruments [34–36]. Generally, the number of 2 H-substitutes in the IS increases the potential for chromatographic resolution between the IS and its analog, but also factors like molecule size and structure, location of the 2 H-substitutes, retention mechanism, retention time, and choice of mobile and stationary phase may influence the resolution [5,8,37–39]. One important factor is the location of the 2 H-isotopes [40]. Zhang et al. have previously proven that grouping the 2 Hisotopes around polar functional groups, showing little affinity to the non-polar stationary phase, compared with grouping the 2 H-isotopes to non-polar parts of the molecule, may dramatically reduce the IEs in RP chromatography [39]. Deuteration may also affect MS/MS analysis. Ottinger, Vestal, and Futrell have all previously presented the isotope effects on the electron ionization MS-spectra of deuterated propanes [41,42]. For LC–MS/MS analyses of drugs, in biological samples, 2 Hlabeled ISs are much more frequently used than ISs labeled with heavier compounds [3,6,11,13,15–18,43], most probably due to the fact that few 13 C-, 15 N-, or 18 O-labeled ISs are commercially available. Chavez-Eng et al. have, however, investigated both the use of 13 C7 -rofecoxib and 13 C2 H3 -rofecoxib as ISs, for the determination of rofecoxib in human plasma, by LC–MS/MS [44]. They found 13 C7 -rofecoxib best suited as IS due to stability problems with 13 C2 H3 -rofecoxib. Gonzalez-Antuna et al. recently showed that they were able to correct for matrix effects by using singly 13 C-labeled compounds for the determination of beta-agonists, also by LC–MS/MS [14]. In a previous study, using UHPLC–MS/MS, we found that 13 C6 -amphetamine and 13 C6 -methamphetamine coeluted with their analytes perfectly, and improved the ability to correct for ion suppression effects compared with using 2 H-labeled ISs [11]. In this study, we investigated further the use of 13 C6 labeled and 2 H-labeled amphetamines, and their behavior during sample preparation by liquid–liquid extraction (LLE) and during UHPLC–MS/MS analyses. Fig. 1 shows the analytes and the ISs used in this study.
2. Experimental 2.1. Reagents and standards Ammonia (25%), methyl tert-butyl ether (MTBE), natriumhydroxid (NaOH), and nitric acid (HNO3 ) were purchased from Merck (Darmstadt, Germany). Cyclohexane was purchased from Rathburn Chemicals Ltd. (Walkerburn, Scotland). Acetonitrile (ACN) and methanol (MeOH) were purchased from LabScan (Dublin, Ireland). Ammonium formate was purchased from BDH (Pole, England) and formic acid from BDH Prolabo (Briare, France). Type 1 water (18.2 M cm) was obtained from an in-house Milli-Q Biocel, from Millipore, with an Ultrapore
Quantum Organex cartridge. Amphetamine was purchased from Sigma–Aldrich (St. Louis, MO, USA). Methamphetamine, methylenedioxyamphetamine (MDA), methylenedioxymethamphetamine (MDMA), 2 H11 -amphetamine, 2 H8 -amphetamine, 2 H6 amphetamine, 2 H5 -amphetamine were purchased from Cerilliant (Round Rock, TX, USA). Methylenedioxyethylamphetamine (MDEA) was purchased from Alltech (Deerfield, IL, USA). Parametoxyamphetamine (PMA), parametoxymethamphetamine (PMMA), 2,5-dimethoxy-4-bromophenethylamine (2C-B), and 2 H -amphetamine were purchased from Lipomed (Arlesheim, 3 Switzerland). 13 C6 -amphetamine, 13 C6 -methamphetamine, 13 C6 MDA, 13 C6 -MDMA, 13 C6 -MDEA, 13 C6 -PMA, 13 C6 -PMMA, and 13 C -2C-B were purchased from Chiron AS (Trondheim, Norway). 6 2.2. Biological samples Biological samples are received for analysis at the Norwegian Institute of Public Health (NIPH), representing many different case categories, such as medico-legal autopsies, medical cases, driving under the influence of drugs [45], drug abuse among inmates suspected by prison services, social medicine, and workplace drug testing. Urine containers without preservatives were purchased from Sterilin (Staffordshire, UK) and Greiner Bio-One (Kremsmûnster, Austria). Urine samples found to be negative for amphetamines by an immunological screening (EMIT) method were used to determine the recovery of amphetamine and ISs in urine samples. Blank whole blood containing 2 g sodium fluoride, 6 mL heparin, and 10 mL water per 450 mL blood, from the blood bank at Ullevaal University Hospital (Oslo, Norway), was used to determine the recovery of amphetamine and ISs in whole blood. 2.3. Preparation of solutions Stock solutions for each analyte and ISs, diluted in MeOH, were prepared in glass volumetric flasks. Working solutions were made in Type 1 water by appropriate dilution of the stock solutions. Standard samples, used to determine recovery, were prepared by appropriate dilution of working solutions in urine and in whole blood. Methamphetamine, dissolved in MeOH, was used as the IS when the recoveries of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine, were determined. 2.4. Sample preparation for the recovery studies Sample preparation was performed by adding 0.10 mL of blank urine or whole blood into plastic vials. Before LLE was performed, 100 L of 10% NaOH was added to all samples. Then 0.700 mL of organic solvent (cyclohexane or MTBE) was added, before the samples were tilted for 7 min and then centrifuged at 3500 rpm for 5 min using an Allegra X-15R centrifuge from Beckman Coulter (Fullerton, CA, USA). From the top layer of each sample, 0.450 mL was transferred into 5 mL glass vials. 50 L of a working solution, containing 1493 ng/mL (10 M) of methamphetamine, was added to all samples into the 5 mL glass vials. 30 L of 0.1% HNO3 , in MeOH, was added to each vial before evaporation at 45–60 ◦ C, using N2 at five PSI on a turbovap evaporator. The samples were reconstituted in 60 L of a solution containing MeOH/5 mM of NH4 -formate at pH 10.2 (5:95), and transferred into autosampler vials. Five microliters of the extracted samples were analyzed by UHPLC–MS/MS. For spiked samples, 50 L of a working solution was added, either after adding biological sample to the plastic vials or into the 5 mL glass vials after LLE. Recovery was calculated by comparing the LC–MS/MS concentrations of standard samples spiked with the analytes before extraction with the LC–MS/MS
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
85
Fig. 1. Molecule structures of analytes and SIL-ISs.
concentrations of standard samples spiked with the analytes after extraction. Only 0.450 mL of the 0.700 mL organic phase was transferred after LLE and before evaporation, and was compensated for when the recoveries were determined.
2.5. Instrumentation An Acquity UHPLC with a sample manager and a binary solvent manager, coupled to a Quattro Premier Xe MS/MS, all from
86
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
Table 1 Transition ions of the analytes and ISs, with mass spectrometric parameters (cone voltage, collision energy, and dwell time). Cone voltagea (V)
Collision energya (V)
Dwell time (ms)
91.05 119.09
15 15
12 12
5 5
> >
97.07 125.11
15 15
12 12
5 5
139.13 139.13
> >
92.06 122.10
15 15
12 12
5 5
H5 -amphetamine
141.14 141.14
> >
96.09 124.12
15 15
12 12
5 5
2
H6 -amphetamine
142.15 142.15
> >
93.07 125.12
15 15
12 12
5 5
2
H8 -amphetamine
144.16 144.16
> >
97.09 127.14
15 15
12 12
5 5
2
H11 -amphetamine
147.18 147.18
> >
98.10 130.15
15 15
12 12
5 5
Methamphetamine 13 C6 -methampetamine
150.13 156.15
> >
91.05 97.07
15 15
14 14
5 5
MDA 13 C6 -MDA
180.10 186.12
> >
163.08 169.10
15 15
10 10
5 5
MDMA 13 C6 -MDMA
194.12 200.14
> >
163.08 169.10
20 20
14 14
5 5
MDEA 13 C6 -MDEA
208.13 214.15
> >
163.08 169.10
20 20
14 14
5 5
PMA 13 C6 -PMA
166.12 172.14
> >
149.10 155.12
15 15
14 14
5 5
PMMA 13 C6 -PMMA
180.14 186.16
> >
149.10 155.12
20 20
10 10
5 5
2
262.03 268.05
> >
245.00 251.02
20 20
15 15
5 5
Compound
Transition ion
Amphetamine
136.11 136.11
> >
13
C6 -ampetamine
142.13 142.13
2
H3 -amphetamine
2
C-B C6 -2C-B
13 a
Values used unless otherwise specified in the experiments.
Table 2 Relative MRM ion ratios at different collision energies (CEs).e Compound
CE (V)
nd
[M+H]+a Relative response (%)
RSD (%) – – – – – – –
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
5 5 5 5 5 5 5
6 6 6 6 6 6 6
100 100 100 100 100 100 100
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
10 10 10 10 10 10 10
6 6 6 6 6 6 6
18 22 20 24 23 26 30
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
15 15 15 15 15 15 15
6 6 6 6 6 6 6
a
[F1+H]+ b
0.40 0.52 0.56 0.73 0.68 1.01 0.88
1.2 0.8 0.8 1.1 1.3 1.0 1.4 6.6 4.5 5.5 3.9 6.9 2.8 1.3
Relative response (%) 32 26 27 23 23 20 17 100 100 100 100 100 100 100 7.9 9.1 10.9 12.0 12.3 16.8 13.5
[F2+H]+ c RSD (%) 1.2 0.7 0.9 1.0 1.1 0.5 0.8
Relative response (%) 4.5 3.7 3.3 3.0 3.1 2.2 2.3
RSD (%) 1.6 1.1 1.8 2.3 3.8 2.0 1.3
– – – – – – –
121 104 89 90 88 65 73
0.9 0.7 0.8 0.8 1.0 0.6 0.8
0.7 1.4 1.0 1.1 1.0 1.2 0.5
100 100 100 100 100 100 100
– – – – – – –
[M+H]+ ion corresponds to the MRM ion for amphetamine with m/z 136 > 136. Fragment 1, [F1+H]+ ion corresponds to the MRM ion for amphetamine with m/z 136 > 119. c Fragment 2, [F2+H]+ ion corresponds to the MRM ion for amphetamine with m/z 136 > 91. d UHPLC–MS/MS analyses were performed with six injections from the working solutions of each compound. The injection order was organized to avoid for instrument drifting to potentially affect the results. e The gradient profile used for this experiment, performed with the basic mobile phase, was; 5% B in 0.0–0.15 min, 5–60% B in 0.15–3.50 min, 60–98% B in 3.50–4.00 min, 98% B in 4.00–5.20 min, 98–5% B in 5.20–5.40 min, 5% B in 5.40–5.80 min. b
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
87
Fig. 2. Chromatographic separation of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine. UHPLC–MS/MS analyses of working solutions were performed using a basic mobile phase on a Waters Acquity BEH C18 -column (a), and an acidic mobile phase on a Waters Acquity HSS T3-column (b). The gradient profile used on the BEH C18 -column was 5% B in 0.00–0.15 min, 5–10% B in 0.15–0.30 min, 10–40% B in 0.30–16.00 min, 40–98% B in 16.00–16.50 min, 98% B in 16.50–17.00 min, 98–5% B in 17.00–17.50 min, and 5% B in 17.50–18.00. The gradient profile used on the HSS T3-column was 0% B in 0.00–0.15 min, 0–25% B in 0.15–15.00 min, 25–98% B in 15.00–15.50 min, 98% B in 15.50–17.50 min, 98–0% B in 17.50–17.70 min, and 0% B in 17.70–18.50 min. Abbreviations: A (amphetamine), MA (methamphetamine).
Waters (Milford, MA, USA), was applied. A basic mobile phase was used for the recovery studies and the MS studies, while the chromatographic separation studies were performed with the basic and with an acidic mobile phase. Chromatographic separations using the basic mobile phase was performed at 60 ◦ C, using an Acquity UHPLC BEH C18 -column (2.1 mm ID × 50 mm, 1.7 m particles) and an Acquity UHPLC BEH C18 -precolumn (2.1 mm ID × 5 mm, 1.7 m particles), both from Waters (Wexford, Ireland). The mobile phase flow rate was 0.400 mL/min, and consisted of 5 mM ammonium format, pH 10.2 (solvent A) and MeOH (solvent B). The gradient profiles used with the basic mobile phase are specified in figure legends (Figs. 2 and 3) and in tables (Tables 2 and 3). Chromatographic separation using the acidic mobile phase was performed at 60 ◦ C on the Acquity UHPLC HSS T3-column (2.1 mm ID × 100 mm, 1.8 m particles) with an Acquity UHPLC HSS T3-precolumn (2.1 mm ID × 5 mm, 1.8 m particles), both from Waters (Wexford, Ireland). The mobile phase flow rate was 0.400 mL/min and the mobile phase consisted of 0.2% HCOOH (pH ∼ 2.7) in Type 1 water (solvent A) and in MeOH (solvent B). The gradient profile used for the chromatographic separations with the acidic mobile phase is specified in Figs. 2 and 3 legends. Positive electrospray ionization (ESI+ ) MS/MS-detection was performed in multiple reaction monitoring (MRM) mode. Desolvation gas temperature was 500 ◦ C and desolvation gas flow was 900 L/h. Capillary voltage was 1 kV. Table 1 shows the analyte and IS-transition ions, with associated mass spectrometric parameters (cone voltage, collision energy, and dwell time).
3. Results and discussion 3.1. Chromatographic separation The chromatographic separation of amphetamines and different SIL-ISs were investigated by using UHPLC–MS/MS, analyzing the working solutions. Fig. 2 shows the chromatographic separation of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine, while Fig. 3 shows the chromatographic separation of eight 13 C6 -labeled amphetamines and their corresponding analytes. Fig. 2a and b shows that when the number of 2 H-substitutes in the IS are increased, the chromatographic resolution between the 2 H-labeled amphetamines and amphetamine increases. However, there is one exception; the chromatographic separation between 2 H5 -amphetamine and amphetamine was greater than between 2 H6 -amphetamine and amphetamine, in the experiment performed using an acidic mobile phase (Fig. 2b). This finding may be explained by the 2 H-isotopes of 2 H6 -amphetamine being located closer to the amine group than for that of 2 H5 -amphetamine (Fig. 1). Under acidic conditions, the amine group of the amphetamines (pKa ∼ 10) is partially ionized and therefore has a low affinity to the non-polar stationary phase. This observation is also consistent with findings in a former study by Zhang et al., which showed that grouping the 2 H-isotopes around polar functional groups could dramatically reduce the IEs in RP chromatography [39].
88
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
Fig. 3. Chromatographic separation of eight 13 C6 -labeled amphetamines and their corresponding analytes. UHPLC–MS/MS analyses of working solutions were performed using a basic mobile phase on a Waters Acquity BEH C18 -column (a), and an acidic mobile phase on a Waters Acquity HSS T3-column (b). The gradient profiles were the same as described in Fig. 2 legend. Abbreviations: A (amphetamine), MA (methamphetamine).
Fig. 3a and b shows that all of the eight 13 C-labeled ISs co-eluted with their analytes, both when using a basic- and an acidic mobile phase. In a previous study, we showed the importance of using ISs that co-elute with the analytes to correct for ion suppression effects, when amphetamine and methamphetamine were analyzed by UHPLC–MS/MS [11]. 3.2. Cone voltage The cone voltages giving maximum [M+H]+ -response for amphetamine, 2 H3 -, 2 H5 -, 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine were determined by UHPLC–MS/MS analyses of working solutions, using a basic mobile phase. Maximum responses
were observed at cone voltages 14–15 V, for all compounds (data not presented). 3.3. Relative responses of [M+H] + and fragment ions The relative responses for the [M+H]+ -ions and the two fragment ions, of amphetamine and the different SIL-ISs, were investigated by applying UHPLC–MS/MS for the analyses at three different collision energies (Table 2). Table 2 shows that with a collision-energy of 5 V, the relative responses of the fragment ions are lowest among the ISs with the highest number of 2 H-substitutes. When the collision energies were increased to 10 V and 15 V, amphetamine had the lowest
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
89
Table 3 Extraction recovery of amphetamine and isotop labeled amphetamines.a , b Compound
Medium
Organic solvent (LLE)
n
Theoretical concentration (M)
Theoretical concentration (ng/mL)
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
Whole blood Whole blood Whole blood Whole blood Whole blood Whole blood Whole blood
Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane
6 6 6 6 6 6 6
0.20 0.20 0.20 0.20 0.20 0.20 0.20
27.0 28.2 27.6 28.0 28.2 28.6 29.2
Recovery (%) 72 76 75 71 72 72 71
RSD 4.8 8.8 8.8 2.5 3.5 3.5 3.2
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
Whole blood Whole blood Whole blood Whole blood Whole blood Whole blood Whole blood
MTBE MTBE MTBE MTBE MTBE MTBE MTBE
6 6 6 6 6 6 6
0.20 0.20 0.20 0.20 0.20 0.20 0.20
27.0 28.2 27.6 28.0 28.2 28.6 29.2
96 96 94 95 95 96 93
3.3 6.6 5.3 1.8 3.4 6.6 1.6
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
Urine Urine Urine Urine Urine Urine Urine
Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane
6 6 6 6 6 6 6
0.20 0.20 0.20 0.20 0.20 0.20 0.20
27.0 28.2 27.6 28.0 28.2 28.6 29.2
91 91 86 89 89 88 88
8.4 4.0 2.6 5.4 3.3 5.2 4.3
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
Urine Urine Urine Urine Urine Urine Urine
MTBE MTBE MTBE MTBE MTBE MTBE MTBE
6 6 6 6 6 6 6
0.20 0.20 0.20 0.20 0.20 0.20 0.20
27.0 28.2 27.6 28.0 28.2 28.6 29.2
102 101 99 100 105 105 101
6.1 5.5 8.5 2.6 5.0 4.9 3.9
a
Methamphetamine was used as the IS for the seven components investigated. The gradient profile used for this experiment, performed with the basic mobile phase, was; 5% B in 0.0–0.15 min, 5–30% B in 0.15–0.30 min, 30–40% B in 0.30–2.00 min, 40–50% B in 2.00–3.00 min, 50–98% B in 3.00–4.20 min, 98% B in 4.20–5.20, 98–5% B in 5.20–5.40 min, 5% B in 5.40–5.80 min. b
relative response of the [M+H]+ -ion, while the 2 H8 - and 2 H11 amphetamines had the highest relative [M+H]+ -ion responses. These observations indicate that it is easier to cause fragmentation of amphetamine than what it is to fragment the SIL-ISs, and that the ISs with the highest number of 2 H-substitutes require more energy for fragmentation. 3.4. Extraction recovery of amphetamine and standard isotope-labeled amphetamines A series of experiments were conducted to determine the recovery of amphetamine, 2 H3 -, 2 H5 -, 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine in urine and in whole blood samples, prepared by LLE (Table 3). Table 3 shows that there were none or only minor differences between the LLE recoveries of amphetamine and the different SILISs. This observation agrees with other studies that show only minor IEs on factors that may influence on LLE recovery of a compound, like P-values, pKa and liphophilicity [21–25]. Table 3 also shows that the recoveries were greater when MTBE was used as the organic solvent, compared with the recoveries observed with the use of cyclohexane. 4. Conclusion This study investigated the behavior of amphetamine, 2 H3 , 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine, during sample preparation by LLE and during LC–MS/MS analyses. None or only minor differences were observed in the LLE recoveries of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine. UHPLC–MS/MS analyses show that the chromatographic separation of amphetamine and the 2 H-labeled ISs increased with the number of 2 H-substitutes in the ISs, both when using a basic- and an acidic mobile phase. There was, however,
one exception; applying an acidic mobile phase, the separation of 2 H -amphetamine and amphetamine was observed greater than 5 for that of 2 H6 -amphetamine and amphetamine, most probably explained by the 2 H-isotopes of 2 H5 -amphetamine being located closer to the partially ionized functional group than for that of 2 H6 amphetamine. For the chromatographic analyses, seven additional 13 C -labeled amphetamines and their corresponding analytes were 6 also included, with all of the 13 C6 -labeled ISs co-eluting with their analytes. The LC–MS/MS analyses of amphetamine and its SIL-ISs indicate that the fragmentation of amphetamine occurs more easily than for the ISs, and that the ISs with the highest number of 2 Hsubstitutes require more energy for fragmentation. The findings, in this study, support those of previous studies that have found 13 Clabeled ISs to be superior to 2 H-labeled ISs, for analytical purposes. Using 2 H-labeled ISs, one should always consider choosing the ISs with few, but still three or more 2 H-substitutes.
Acknowledgements The authors would like to thank Stine Marie Havig and Vigdis Vindenes for their valuable comments and for critical reading of the manuscript. The authors would also like to thank Anja Valen for drawing the molecular structures presented in Fig. 1, and finally Elisabeth Øiestad and Vigdis Vindenes for their support and encouragements during the writing process.
References [1] [2] [3] [4]
L. Novakova, H. Vlckova, Anal. Chim. Acta 656 (2009) 8. F.T. Peters, Clin. Biochem. 44 (2011) 54. L. Novakova, J. Chromatogr. A 1292 (2013) 25. A. Tan, K. Awaie, W. in Li, J. Zhang, F.L.S. Tse (Eds.), Handbook of LC–MS bioanalysis, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2013, p. 217. [5] J. Wieling, Chromatographia 55 (2002) 107.
90
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
[6] E. Stokvis, H. Rosing, J.H. Beijnen, Rapid Commun. Mass Spectrom. 19 (2005) 401. [7] P. Krumbiegel, Isotopes Environ. Health Stud. 47 (2011) 1. [8] M. Turowski, N. Yamakawa, J. Meller, K. Kimata, T. Ikegami, K. Hosoya, N. Tanaka, E.R. Thornton, J. Am. Chem. Soc. 125 (2003) 13836. [9] F.H. Westheimer, Chem. Rev. 61 (1961) 265. [10] K.B. Wiberg, Chem. Rev. 55 (1955) 713. [11] T. Berg, D.H. Strand, J. Chromatogr. A 1218 (2011) 9366. [12] C.J. Briscoe, M.R. Stiles, D.S. Hage, J. Pharm. Biomed. Anal. 44 (2007) 484. [13] L. Couchman, P.E. Morgan, Biomed. Chromatogr. 25 (2011) 100. [14] A. Gonzalez-Antuna, J.C. Dominguez-Romero, J.F. Garcia-Reyes, P. RodriguezGonzalez, G. Centineo, J.I. Garcia Alonso, A. Molina-Diaz, J. Chromatogr. A 1288 (2013) 40. [15] K. Guo, C. Ji, L. Li, Anal. Chem. 79 (2007) 8631. [16] A.K. Hewavitharana, J. Chromatogr. A 1218 (2011) 359. [17] E.A. Van, K. Lanckmans, S. Sarre, I. Smolders, Y. Michotte, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 877 (2009) 2198. [18] S. Wang, M. Cyronak, E. Yang, J. Pharm. Biomed. Anal. 43 (2007) 701. [19] L.S. Bartell, R.R. Roskos, J. Chem. Phys. 44 (1965) 457. [20] D. Wade, Chem. Biol. Interact. 117 (1999) 191. [21] A. Bechalany, N. El Tayar, P.A. Carrupt, B. Testa, J.B. Falconnet, Y. Cherrah, Y. Benchekroun, J.L. Brazier, Helv. Chim. Acta 72 (1989) 472. [22] N. El Tayar, H. van de Waterbeemd, M. Gryllaki, B. Testa, W.F. Trager, J. Pharmaceut. 19 (1984) 271. [23] I.M. Kovach, D.M. Quinn, J. Am. Chem. Soc. 105 (1983) 1947. [24] T. Pehk, E. Kiirend, E. Lippamaa, U. Ragnarsson, L. Grehn, J. Chem. Soc. Perkin Trans. 2 (1997) 445. [25] C.L. Perrin, B.K. Ohta, J. Kuperman, J. Am. Chem. Soc. 125 (2003) 15008.
[26] A. Streitwieser, H.S. Klein, J. Am. Chem. Soc. (1963) 2759. [27] C.N. Filer, J. Labelled Compd. Radiopharm. 42 (1999) 169. [28] T.M. Annesley, Clin. Chem. 49 (2003) 1041. [29] A. Furey, M. Moriarty, V. Bane, B. Kinsella, M. Lehane, Talanta 115 (2013) 104. [30] D. Remane, M.R. Meyer, D.K. Wissenbach, H.H. Maurer, Rapid Commun. Mass Spectrom. 24 (2010) 3103. [31] D. Remane, D.K. Wissenbach, M.R. Meyer, H.H. Maurer, Rapid Commun. Mass Spectrom. 24 (2010) 859. [32] G.P. Cartoni, I. Ferretti, J. Chromatogr. 122 (1976) 287. [33] N. Tanaka, E.R. Thornton, J. Am. Chem. Soc. (1977) 7300. [34] M.I. Churchwell, N.C. Twaddle, L.R. Meeker, D.R. Doerge, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 825 (2005) 134. [35] L. Novakova, L. Matysova, P. Solich, Talanta 68 (2006) 908. [36] M.E. Swartz, J. Liq. Chrom. Relat. Technol. 28 (2005) 1253. [37] S.S. Iyer, Z.P. Zhang, G.E. Kellogg, H.T. Karnes, J. Chromatogr. Sci. 42 (2004) 383. [38] N. Lindegardh, A. Annerberg, N.J. White, N.P. Day, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 862 (2008) 227. [39] R. Zhang, C.S. Sioma, R.A. Thompson, L. Xiong, F.E. Regnier, Anal. Chem. 74 (2002) 3662. [40] J.R. Heys, J. Chromatogr. 407 (1987) 37. [41] C. Ottinger, J. Chem. Phys. 47 (1967) 1452. [42] M. Vestal, J.H. Futrell, J. Chem. Phys. 52 (1970) 978. [43] R. Zhang, F.E. Regnier, J. Proteome Res. 1 (2002) 139. [44] C.M. Chavez-Eng, M.L. Constanzer, B.K. Matuszewski, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 767 (2002) 117. [45] V. Vindenes, D. Jordbru, A.B. Knapskog, E. Kvan, G. Mathisrud, L. Slordal, J. Morland, Forensic Sci. Int. 219 (2012) 1.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Evaluation of 13 C- and 2 H-labeled internal standards for the determination of amphetamines in biological samples, by reversed-phase ultra-high performance liquid chromatography–tandem mass spectrometry Thomas Berg a,∗ , Morten Karlsen b , Åse Marit Leere Øiestad a , Jon Eigill Johansen b , Huiling Liu b , Dag Helge Strand a a b
Division of Forensic Medicine and Drug Abuse Research, Norwegian Institute of Public Health, Oslo, Norway Chiron AS, Trondheim, Norway
a r t i c l e
i n f o
Article history: Received 13 January 2014 Received in revised form 4 April 2014 Accepted 7 April 2014 Available online 18 April 2014 Keywords: 13 C Stable isotope-labeled internal standard LC–MS/MS Isotope effect Deuterium Amphetamine
a b s t r a c t Stable isotope-labeled internal standards (SIL-ISs) are often used when applying liquid chromatography–tandem mass spectrometry (LC–MS/MS) to analyze for legal and illegal drugs. ISs labeled with 13 C, 15 N, and 18 O are expected to behave more closely to their corresponding unlabeled analytes, compared with that of the more classically used 2 H-labeled ISs. This study has investigated the behavior of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine, during sample preparation by liquid–liquid extraction and LC–MS/MS analyses. None or only minor differences in liquid–liquid extraction recoveries of amphetamine and the SIL-ISs were observed. The chromatographic resolution between amphetamine and the 2 H-labeled amphetamines increased with the number of 2 H-substitutes. For chromatographic studies we also included seven additional 13 C6 -amphetamines and their analytes. All the 13 C6 -labeled ISs were co-eluting with their analytes, both when a basic and when an acidic mobile phase were used. MS/MS analyses of amphetamine and its SIL-ISs showed that the ISs with the highest number of 2 H-substitutes required more energy for fragmentation in the collision cell compared with that of the ISs with a lower number. The findings, in this study, support those of previous studies, showing that 13 C-labeled ISs are superior to 2 H-labeled ISs, for analytical purposes. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Liquid chromatography–tandem mass spectrometry (LC–MS/MS) is commonly used for qualitative and quantitative analyses of drugs in human biological samples [1,2]. SIL-ISs, which are compounds where the atoms in a molecule are replaced by their stable isotopes, such as 2 H, 13 C, 15 N, and 18 O, are often used to improve these drug detections [3–5]. To avoid for the natural occurrence of analyte isotopes interfering with a SIL-IS it should be labeled with three or more isotopes [6]. Ideally, the SIL-IS behaves exactly as the drug of interest during the sample pre-treatment, the chromatographic separation, and the LC–MS/MS detection; however, due to isotope effects (IEs), their behaviors are often
∗ Corresponding author. Tel.: +47 92 60 21 52; fax: +47 22 38 32 33. E-mail address: [email protected] (T. Berg). http://dx.doi.org/10.1016/j.chroma.2014.04.020 0021-9673/© 2014 Elsevier B.V. All rights reserved.
not identical. IEs are defined as the effects on the rate (kinetic IEs), or on the equilibrium constants (thermodynamic IEs), of two systems that differ only in the isotopic composition of one or more of their otherwise chemically identical components [7]. Primary IEs arise when a bond to the isotope is formed or cleaved, while secondary IEs arise when the bond remains intact. The main cause of IEs is the difference in nuclear mass [8–10]. This difference is greater between hydrogen isotopes than between isotopes of other elements, and thereby explains why IEs usually are greater among the 2 H-labeled ISs than for the 13 C, 15 N, and 18 O-labeled ones [3,6,11–18]. For deuterated compounds, the 2 H C bond is shorter, stronger, and less polarizable than the 1 H C bond. The slightly smaller volumes of deuterated compounds compared with the volumes of their protonium isotopomers favors the former in polar phases [19,20]. In addition, the less polarizable 2 H C bond presumably generates weaker London dispersion forces [21]. El Tayar et al. found that when comparing the n-octanol/water partition
84
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
coefficient (P) of deuterated and non-deuterated aromates, the deuterated aromates had the lower P-values (P ≤ 0.05) [22]. Kovach and Quinn found that P-values of deuterated carbonyl compounds were reduced by 1–2% in the organic phase [23]. Other studies have shown how IEs may contribute to changes in pKa values [24–26]. Filer’s paper from 1999 is a thorough review on IEs influence on different chromatographic techniques [27]. When performing LC–MS/MS analyses of drugs in biological samples, it is important for the IS to co-elute with the compound of interest, to be able to correct for possible ion suppression/enhancement effects, and thereby to avoid false results [28–31]. However, for reversed-phase (RP) LC separations, it is well known that the 2 H-labeled ISs often elute slightly earlier than their corresponding analytes [5,15,18,20,32,33]. This problem has become even more relevant during recent years, due to the highly efficient chromatographic separations obtained by ultra-high performance liquid chromatography (UHPLC) instruments [34–36]. Generally, the number of 2 H-substitutes in the IS increases the potential for chromatographic resolution between the IS and its analog, but also factors like molecule size and structure, location of the 2 H-substitutes, retention mechanism, retention time, and choice of mobile and stationary phase may influence the resolution [5,8,37–39]. One important factor is the location of the 2 H-isotopes [40]. Zhang et al. have previously proven that grouping the 2 Hisotopes around polar functional groups, showing little affinity to the non-polar stationary phase, compared with grouping the 2 H-isotopes to non-polar parts of the molecule, may dramatically reduce the IEs in RP chromatography [39]. Deuteration may also affect MS/MS analysis. Ottinger, Vestal, and Futrell have all previously presented the isotope effects on the electron ionization MS-spectra of deuterated propanes [41,42]. For LC–MS/MS analyses of drugs, in biological samples, 2 Hlabeled ISs are much more frequently used than ISs labeled with heavier compounds [3,6,11,13,15–18,43], most probably due to the fact that few 13 C-, 15 N-, or 18 O-labeled ISs are commercially available. Chavez-Eng et al. have, however, investigated both the use of 13 C7 -rofecoxib and 13 C2 H3 -rofecoxib as ISs, for the determination of rofecoxib in human plasma, by LC–MS/MS [44]. They found 13 C7 -rofecoxib best suited as IS due to stability problems with 13 C2 H3 -rofecoxib. Gonzalez-Antuna et al. recently showed that they were able to correct for matrix effects by using singly 13 C-labeled compounds for the determination of beta-agonists, also by LC–MS/MS [14]. In a previous study, using UHPLC–MS/MS, we found that 13 C6 -amphetamine and 13 C6 -methamphetamine coeluted with their analytes perfectly, and improved the ability to correct for ion suppression effects compared with using 2 H-labeled ISs [11]. In this study, we investigated further the use of 13 C6 labeled and 2 H-labeled amphetamines, and their behavior during sample preparation by liquid–liquid extraction (LLE) and during UHPLC–MS/MS analyses. Fig. 1 shows the analytes and the ISs used in this study.
2. Experimental 2.1. Reagents and standards Ammonia (25%), methyl tert-butyl ether (MTBE), natriumhydroxid (NaOH), and nitric acid (HNO3 ) were purchased from Merck (Darmstadt, Germany). Cyclohexane was purchased from Rathburn Chemicals Ltd. (Walkerburn, Scotland). Acetonitrile (ACN) and methanol (MeOH) were purchased from LabScan (Dublin, Ireland). Ammonium formate was purchased from BDH (Pole, England) and formic acid from BDH Prolabo (Briare, France). Type 1 water (18.2 M cm) was obtained from an in-house Milli-Q Biocel, from Millipore, with an Ultrapore
Quantum Organex cartridge. Amphetamine was purchased from Sigma–Aldrich (St. Louis, MO, USA). Methamphetamine, methylenedioxyamphetamine (MDA), methylenedioxymethamphetamine (MDMA), 2 H11 -amphetamine, 2 H8 -amphetamine, 2 H6 amphetamine, 2 H5 -amphetamine were purchased from Cerilliant (Round Rock, TX, USA). Methylenedioxyethylamphetamine (MDEA) was purchased from Alltech (Deerfield, IL, USA). Parametoxyamphetamine (PMA), parametoxymethamphetamine (PMMA), 2,5-dimethoxy-4-bromophenethylamine (2C-B), and 2 H -amphetamine were purchased from Lipomed (Arlesheim, 3 Switzerland). 13 C6 -amphetamine, 13 C6 -methamphetamine, 13 C6 MDA, 13 C6 -MDMA, 13 C6 -MDEA, 13 C6 -PMA, 13 C6 -PMMA, and 13 C -2C-B were purchased from Chiron AS (Trondheim, Norway). 6 2.2. Biological samples Biological samples are received for analysis at the Norwegian Institute of Public Health (NIPH), representing many different case categories, such as medico-legal autopsies, medical cases, driving under the influence of drugs [45], drug abuse among inmates suspected by prison services, social medicine, and workplace drug testing. Urine containers without preservatives were purchased from Sterilin (Staffordshire, UK) and Greiner Bio-One (Kremsmûnster, Austria). Urine samples found to be negative for amphetamines by an immunological screening (EMIT) method were used to determine the recovery of amphetamine and ISs in urine samples. Blank whole blood containing 2 g sodium fluoride, 6 mL heparin, and 10 mL water per 450 mL blood, from the blood bank at Ullevaal University Hospital (Oslo, Norway), was used to determine the recovery of amphetamine and ISs in whole blood. 2.3. Preparation of solutions Stock solutions for each analyte and ISs, diluted in MeOH, were prepared in glass volumetric flasks. Working solutions were made in Type 1 water by appropriate dilution of the stock solutions. Standard samples, used to determine recovery, were prepared by appropriate dilution of working solutions in urine and in whole blood. Methamphetamine, dissolved in MeOH, was used as the IS when the recoveries of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine, were determined. 2.4. Sample preparation for the recovery studies Sample preparation was performed by adding 0.10 mL of blank urine or whole blood into plastic vials. Before LLE was performed, 100 L of 10% NaOH was added to all samples. Then 0.700 mL of organic solvent (cyclohexane or MTBE) was added, before the samples were tilted for 7 min and then centrifuged at 3500 rpm for 5 min using an Allegra X-15R centrifuge from Beckman Coulter (Fullerton, CA, USA). From the top layer of each sample, 0.450 mL was transferred into 5 mL glass vials. 50 L of a working solution, containing 1493 ng/mL (10 M) of methamphetamine, was added to all samples into the 5 mL glass vials. 30 L of 0.1% HNO3 , in MeOH, was added to each vial before evaporation at 45–60 ◦ C, using N2 at five PSI on a turbovap evaporator. The samples were reconstituted in 60 L of a solution containing MeOH/5 mM of NH4 -formate at pH 10.2 (5:95), and transferred into autosampler vials. Five microliters of the extracted samples were analyzed by UHPLC–MS/MS. For spiked samples, 50 L of a working solution was added, either after adding biological sample to the plastic vials or into the 5 mL glass vials after LLE. Recovery was calculated by comparing the LC–MS/MS concentrations of standard samples spiked with the analytes before extraction with the LC–MS/MS
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
85
Fig. 1. Molecule structures of analytes and SIL-ISs.
concentrations of standard samples spiked with the analytes after extraction. Only 0.450 mL of the 0.700 mL organic phase was transferred after LLE and before evaporation, and was compensated for when the recoveries were determined.
2.5. Instrumentation An Acquity UHPLC with a sample manager and a binary solvent manager, coupled to a Quattro Premier Xe MS/MS, all from
86
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
Table 1 Transition ions of the analytes and ISs, with mass spectrometric parameters (cone voltage, collision energy, and dwell time). Cone voltagea (V)
Collision energya (V)
Dwell time (ms)
91.05 119.09
15 15
12 12
5 5
> >
97.07 125.11
15 15
12 12
5 5
139.13 139.13
> >
92.06 122.10
15 15
12 12
5 5
H5 -amphetamine
141.14 141.14
> >
96.09 124.12
15 15
12 12
5 5
2
H6 -amphetamine
142.15 142.15
> >
93.07 125.12
15 15
12 12
5 5
2
H8 -amphetamine
144.16 144.16
> >
97.09 127.14
15 15
12 12
5 5
2
H11 -amphetamine
147.18 147.18
> >
98.10 130.15
15 15
12 12
5 5
Methamphetamine 13 C6 -methampetamine
150.13 156.15
> >
91.05 97.07
15 15
14 14
5 5
MDA 13 C6 -MDA
180.10 186.12
> >
163.08 169.10
15 15
10 10
5 5
MDMA 13 C6 -MDMA
194.12 200.14
> >
163.08 169.10
20 20
14 14
5 5
MDEA 13 C6 -MDEA
208.13 214.15
> >
163.08 169.10
20 20
14 14
5 5
PMA 13 C6 -PMA
166.12 172.14
> >
149.10 155.12
15 15
14 14
5 5
PMMA 13 C6 -PMMA
180.14 186.16
> >
149.10 155.12
20 20
10 10
5 5
2
262.03 268.05
> >
245.00 251.02
20 20
15 15
5 5
Compound
Transition ion
Amphetamine
136.11 136.11
> >
13
C6 -ampetamine
142.13 142.13
2
H3 -amphetamine
2
C-B C6 -2C-B
13 a
Values used unless otherwise specified in the experiments.
Table 2 Relative MRM ion ratios at different collision energies (CEs).e Compound
CE (V)
nd
[M+H]+a Relative response (%)
RSD (%) – – – – – – –
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
5 5 5 5 5 5 5
6 6 6 6 6 6 6
100 100 100 100 100 100 100
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
10 10 10 10 10 10 10
6 6 6 6 6 6 6
18 22 20 24 23 26 30
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
15 15 15 15 15 15 15
6 6 6 6 6 6 6
a
[F1+H]+ b
0.40 0.52 0.56 0.73 0.68 1.01 0.88
1.2 0.8 0.8 1.1 1.3 1.0 1.4 6.6 4.5 5.5 3.9 6.9 2.8 1.3
Relative response (%) 32 26 27 23 23 20 17 100 100 100 100 100 100 100 7.9 9.1 10.9 12.0 12.3 16.8 13.5
[F2+H]+ c RSD (%) 1.2 0.7 0.9 1.0 1.1 0.5 0.8
Relative response (%) 4.5 3.7 3.3 3.0 3.1 2.2 2.3
RSD (%) 1.6 1.1 1.8 2.3 3.8 2.0 1.3
– – – – – – –
121 104 89 90 88 65 73
0.9 0.7 0.8 0.8 1.0 0.6 0.8
0.7 1.4 1.0 1.1 1.0 1.2 0.5
100 100 100 100 100 100 100
– – – – – – –
[M+H]+ ion corresponds to the MRM ion for amphetamine with m/z 136 > 136. Fragment 1, [F1+H]+ ion corresponds to the MRM ion for amphetamine with m/z 136 > 119. c Fragment 2, [F2+H]+ ion corresponds to the MRM ion for amphetamine with m/z 136 > 91. d UHPLC–MS/MS analyses were performed with six injections from the working solutions of each compound. The injection order was organized to avoid for instrument drifting to potentially affect the results. e The gradient profile used for this experiment, performed with the basic mobile phase, was; 5% B in 0.0–0.15 min, 5–60% B in 0.15–3.50 min, 60–98% B in 3.50–4.00 min, 98% B in 4.00–5.20 min, 98–5% B in 5.20–5.40 min, 5% B in 5.40–5.80 min. b
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
87
Fig. 2. Chromatographic separation of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine. UHPLC–MS/MS analyses of working solutions were performed using a basic mobile phase on a Waters Acquity BEH C18 -column (a), and an acidic mobile phase on a Waters Acquity HSS T3-column (b). The gradient profile used on the BEH C18 -column was 5% B in 0.00–0.15 min, 5–10% B in 0.15–0.30 min, 10–40% B in 0.30–16.00 min, 40–98% B in 16.00–16.50 min, 98% B in 16.50–17.00 min, 98–5% B in 17.00–17.50 min, and 5% B in 17.50–18.00. The gradient profile used on the HSS T3-column was 0% B in 0.00–0.15 min, 0–25% B in 0.15–15.00 min, 25–98% B in 15.00–15.50 min, 98% B in 15.50–17.50 min, 98–0% B in 17.50–17.70 min, and 0% B in 17.70–18.50 min. Abbreviations: A (amphetamine), MA (methamphetamine).
Waters (Milford, MA, USA), was applied. A basic mobile phase was used for the recovery studies and the MS studies, while the chromatographic separation studies were performed with the basic and with an acidic mobile phase. Chromatographic separations using the basic mobile phase was performed at 60 ◦ C, using an Acquity UHPLC BEH C18 -column (2.1 mm ID × 50 mm, 1.7 m particles) and an Acquity UHPLC BEH C18 -precolumn (2.1 mm ID × 5 mm, 1.7 m particles), both from Waters (Wexford, Ireland). The mobile phase flow rate was 0.400 mL/min, and consisted of 5 mM ammonium format, pH 10.2 (solvent A) and MeOH (solvent B). The gradient profiles used with the basic mobile phase are specified in figure legends (Figs. 2 and 3) and in tables (Tables 2 and 3). Chromatographic separation using the acidic mobile phase was performed at 60 ◦ C on the Acquity UHPLC HSS T3-column (2.1 mm ID × 100 mm, 1.8 m particles) with an Acquity UHPLC HSS T3-precolumn (2.1 mm ID × 5 mm, 1.8 m particles), both from Waters (Wexford, Ireland). The mobile phase flow rate was 0.400 mL/min and the mobile phase consisted of 0.2% HCOOH (pH ∼ 2.7) in Type 1 water (solvent A) and in MeOH (solvent B). The gradient profile used for the chromatographic separations with the acidic mobile phase is specified in Figs. 2 and 3 legends. Positive electrospray ionization (ESI+ ) MS/MS-detection was performed in multiple reaction monitoring (MRM) mode. Desolvation gas temperature was 500 ◦ C and desolvation gas flow was 900 L/h. Capillary voltage was 1 kV. Table 1 shows the analyte and IS-transition ions, with associated mass spectrometric parameters (cone voltage, collision energy, and dwell time).
3. Results and discussion 3.1. Chromatographic separation The chromatographic separation of amphetamines and different SIL-ISs were investigated by using UHPLC–MS/MS, analyzing the working solutions. Fig. 2 shows the chromatographic separation of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine, while Fig. 3 shows the chromatographic separation of eight 13 C6 -labeled amphetamines and their corresponding analytes. Fig. 2a and b shows that when the number of 2 H-substitutes in the IS are increased, the chromatographic resolution between the 2 H-labeled amphetamines and amphetamine increases. However, there is one exception; the chromatographic separation between 2 H5 -amphetamine and amphetamine was greater than between 2 H6 -amphetamine and amphetamine, in the experiment performed using an acidic mobile phase (Fig. 2b). This finding may be explained by the 2 H-isotopes of 2 H6 -amphetamine being located closer to the amine group than for that of 2 H5 -amphetamine (Fig. 1). Under acidic conditions, the amine group of the amphetamines (pKa ∼ 10) is partially ionized and therefore has a low affinity to the non-polar stationary phase. This observation is also consistent with findings in a former study by Zhang et al., which showed that grouping the 2 H-isotopes around polar functional groups could dramatically reduce the IEs in RP chromatography [39].
88
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
Fig. 3. Chromatographic separation of eight 13 C6 -labeled amphetamines and their corresponding analytes. UHPLC–MS/MS analyses of working solutions were performed using a basic mobile phase on a Waters Acquity BEH C18 -column (a), and an acidic mobile phase on a Waters Acquity HSS T3-column (b). The gradient profiles were the same as described in Fig. 2 legend. Abbreviations: A (amphetamine), MA (methamphetamine).
Fig. 3a and b shows that all of the eight 13 C-labeled ISs co-eluted with their analytes, both when using a basic- and an acidic mobile phase. In a previous study, we showed the importance of using ISs that co-elute with the analytes to correct for ion suppression effects, when amphetamine and methamphetamine were analyzed by UHPLC–MS/MS [11]. 3.2. Cone voltage The cone voltages giving maximum [M+H]+ -response for amphetamine, 2 H3 -, 2 H5 -, 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine were determined by UHPLC–MS/MS analyses of working solutions, using a basic mobile phase. Maximum responses
were observed at cone voltages 14–15 V, for all compounds (data not presented). 3.3. Relative responses of [M+H] + and fragment ions The relative responses for the [M+H]+ -ions and the two fragment ions, of amphetamine and the different SIL-ISs, were investigated by applying UHPLC–MS/MS for the analyses at three different collision energies (Table 2). Table 2 shows that with a collision-energy of 5 V, the relative responses of the fragment ions are lowest among the ISs with the highest number of 2 H-substitutes. When the collision energies were increased to 10 V and 15 V, amphetamine had the lowest
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
89
Table 3 Extraction recovery of amphetamine and isotop labeled amphetamines.a , b Compound
Medium
Organic solvent (LLE)
n
Theoretical concentration (M)
Theoretical concentration (ng/mL)
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
Whole blood Whole blood Whole blood Whole blood Whole blood Whole blood Whole blood
Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane
6 6 6 6 6 6 6
0.20 0.20 0.20 0.20 0.20 0.20 0.20
27.0 28.2 27.6 28.0 28.2 28.6 29.2
Recovery (%) 72 76 75 71 72 72 71
RSD 4.8 8.8 8.8 2.5 3.5 3.5 3.2
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
Whole blood Whole blood Whole blood Whole blood Whole blood Whole blood Whole blood
MTBE MTBE MTBE MTBE MTBE MTBE MTBE
6 6 6 6 6 6 6
0.20 0.20 0.20 0.20 0.20 0.20 0.20
27.0 28.2 27.6 28.0 28.2 28.6 29.2
96 96 94 95 95 96 93
3.3 6.6 5.3 1.8 3.4 6.6 1.6
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
Urine Urine Urine Urine Urine Urine Urine
Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane Cyclohexane
6 6 6 6 6 6 6
0.20 0.20 0.20 0.20 0.20 0.20 0.20
27.0 28.2 27.6 28.0 28.2 28.6 29.2
91 91 86 89 89 88 88
8.4 4.0 2.6 5.4 3.3 5.2 4.3
Amphetamine 13 C6 -amphetamine 2 H3 -amphetamine 2 H5 -amphetamine 2 H6 -amphetamine 2 H8 -amphetamine 2 H11 -amphetamine
Urine Urine Urine Urine Urine Urine Urine
MTBE MTBE MTBE MTBE MTBE MTBE MTBE
6 6 6 6 6 6 6
0.20 0.20 0.20 0.20 0.20 0.20 0.20
27.0 28.2 27.6 28.0 28.2 28.6 29.2
102 101 99 100 105 105 101
6.1 5.5 8.5 2.6 5.0 4.9 3.9
a
Methamphetamine was used as the IS for the seven components investigated. The gradient profile used for this experiment, performed with the basic mobile phase, was; 5% B in 0.0–0.15 min, 5–30% B in 0.15–0.30 min, 30–40% B in 0.30–2.00 min, 40–50% B in 2.00–3.00 min, 50–98% B in 3.00–4.20 min, 98% B in 4.20–5.20, 98–5% B in 5.20–5.40 min, 5% B in 5.40–5.80 min. b
relative response of the [M+H]+ -ion, while the 2 H8 - and 2 H11 amphetamines had the highest relative [M+H]+ -ion responses. These observations indicate that it is easier to cause fragmentation of amphetamine than what it is to fragment the SIL-ISs, and that the ISs with the highest number of 2 H-substitutes require more energy for fragmentation. 3.4. Extraction recovery of amphetamine and standard isotope-labeled amphetamines A series of experiments were conducted to determine the recovery of amphetamine, 2 H3 -, 2 H5 -, 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine in urine and in whole blood samples, prepared by LLE (Table 3). Table 3 shows that there were none or only minor differences between the LLE recoveries of amphetamine and the different SILISs. This observation agrees with other studies that show only minor IEs on factors that may influence on LLE recovery of a compound, like P-values, pKa and liphophilicity [21–25]. Table 3 also shows that the recoveries were greater when MTBE was used as the organic solvent, compared with the recoveries observed with the use of cyclohexane. 4. Conclusion This study investigated the behavior of amphetamine, 2 H3 , 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine, during sample preparation by LLE and during LC–MS/MS analyses. None or only minor differences were observed in the LLE recoveries of amphetamine, 2 H3 -, 2 H5 , 2 H6 -, 2 H8 -, 2 H11 -, and 13 C6 -labeled amphetamine. UHPLC–MS/MS analyses show that the chromatographic separation of amphetamine and the 2 H-labeled ISs increased with the number of 2 H-substitutes in the ISs, both when using a basic- and an acidic mobile phase. There was, however,
one exception; applying an acidic mobile phase, the separation of 2 H -amphetamine and amphetamine was observed greater than 5 for that of 2 H6 -amphetamine and amphetamine, most probably explained by the 2 H-isotopes of 2 H5 -amphetamine being located closer to the partially ionized functional group than for that of 2 H6 amphetamine. For the chromatographic analyses, seven additional 13 C -labeled amphetamines and their corresponding analytes were 6 also included, with all of the 13 C6 -labeled ISs co-eluting with their analytes. The LC–MS/MS analyses of amphetamine and its SIL-ISs indicate that the fragmentation of amphetamine occurs more easily than for the ISs, and that the ISs with the highest number of 2 Hsubstitutes require more energy for fragmentation. The findings, in this study, support those of previous studies that have found 13 Clabeled ISs to be superior to 2 H-labeled ISs, for analytical purposes. Using 2 H-labeled ISs, one should always consider choosing the ISs with few, but still three or more 2 H-substitutes.
Acknowledgements The authors would like to thank Stine Marie Havig and Vigdis Vindenes for their valuable comments and for critical reading of the manuscript. The authors would also like to thank Anja Valen for drawing the molecular structures presented in Fig. 1, and finally Elisabeth Øiestad and Vigdis Vindenes for their support and encouragements during the writing process.
References [1] [2] [3] [4]
L. Novakova, H. Vlckova, Anal. Chim. Acta 656 (2009) 8. F.T. Peters, Clin. Biochem. 44 (2011) 54. L. Novakova, J. Chromatogr. A 1292 (2013) 25. A. Tan, K. Awaie, W. in Li, J. Zhang, F.L.S. Tse (Eds.), Handbook of LC–MS bioanalysis, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2013, p. 217. [5] J. Wieling, Chromatographia 55 (2002) 107.
90
T. Berg et al. / J. Chromatogr. A 1344 (2014) 83–90
[6] E. Stokvis, H. Rosing, J.H. Beijnen, Rapid Commun. Mass Spectrom. 19 (2005) 401. [7] P. Krumbiegel, Isotopes Environ. Health Stud. 47 (2011) 1. [8] M. Turowski, N. Yamakawa, J. Meller, K. Kimata, T. Ikegami, K. Hosoya, N. Tanaka, E.R. Thornton, J. Am. Chem. Soc. 125 (2003) 13836. [9] F.H. Westheimer, Chem. Rev. 61 (1961) 265. [10] K.B. Wiberg, Chem. Rev. 55 (1955) 713. [11] T. Berg, D.H. Strand, J. Chromatogr. A 1218 (2011) 9366. [12] C.J. Briscoe, M.R. Stiles, D.S. Hage, J. Pharm. Biomed. Anal. 44 (2007) 484. [13] L. Couchman, P.E. Morgan, Biomed. Chromatogr. 25 (2011) 100. [14] A. Gonzalez-Antuna, J.C. Dominguez-Romero, J.F. Garcia-Reyes, P. RodriguezGonzalez, G. Centineo, J.I. Garcia Alonso, A. Molina-Diaz, J. Chromatogr. A 1288 (2013) 40. [15] K. Guo, C. Ji, L. Li, Anal. Chem. 79 (2007) 8631. [16] A.K. Hewavitharana, J. Chromatogr. A 1218 (2011) 359. [17] E.A. Van, K. Lanckmans, S. Sarre, I. Smolders, Y. Michotte, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 877 (2009) 2198. [18] S. Wang, M. Cyronak, E. Yang, J. Pharm. Biomed. Anal. 43 (2007) 701. [19] L.S. Bartell, R.R. Roskos, J. Chem. Phys. 44 (1965) 457. [20] D. Wade, Chem. Biol. Interact. 117 (1999) 191. [21] A. Bechalany, N. El Tayar, P.A. Carrupt, B. Testa, J.B. Falconnet, Y. Cherrah, Y. Benchekroun, J.L. Brazier, Helv. Chim. Acta 72 (1989) 472. [22] N. El Tayar, H. van de Waterbeemd, M. Gryllaki, B. Testa, W.F. Trager, J. Pharmaceut. 19 (1984) 271. [23] I.M. Kovach, D.M. Quinn, J. Am. Chem. Soc. 105 (1983) 1947. [24] T. Pehk, E. Kiirend, E. Lippamaa, U. Ragnarsson, L. Grehn, J. Chem. Soc. Perkin Trans. 2 (1997) 445. [25] C.L. Perrin, B.K. Ohta, J. Kuperman, J. Am. Chem. Soc. 125 (2003) 15008.
[26] A. Streitwieser, H.S. Klein, J. Am. Chem. Soc. (1963) 2759. [27] C.N. Filer, J. Labelled Compd. Radiopharm. 42 (1999) 169. [28] T.M. Annesley, Clin. Chem. 49 (2003) 1041. [29] A. Furey, M. Moriarty, V. Bane, B. Kinsella, M. Lehane, Talanta 115 (2013) 104. [30] D. Remane, M.R. Meyer, D.K. Wissenbach, H.H. Maurer, Rapid Commun. Mass Spectrom. 24 (2010) 3103. [31] D. Remane, D.K. Wissenbach, M.R. Meyer, H.H. Maurer, Rapid Commun. Mass Spectrom. 24 (2010) 859. [32] G.P. Cartoni, I. Ferretti, J. Chromatogr. 122 (1976) 287. [33] N. Tanaka, E.R. Thornton, J. Am. Chem. Soc. (1977) 7300. [34] M.I. Churchwell, N.C. Twaddle, L.R. Meeker, D.R. Doerge, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 825 (2005) 134. [35] L. Novakova, L. Matysova, P. Solich, Talanta 68 (2006) 908. [36] M.E. Swartz, J. Liq. Chrom. Relat. Technol. 28 (2005) 1253. [37] S.S. Iyer, Z.P. Zhang, G.E. Kellogg, H.T. Karnes, J. Chromatogr. Sci. 42 (2004) 383. [38] N. Lindegardh, A. Annerberg, N.J. White, N.P. Day, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 862 (2008) 227. [39] R. Zhang, C.S. Sioma, R.A. Thompson, L. Xiong, F.E. Regnier, Anal. Chem. 74 (2002) 3662. [40] J.R. Heys, J. Chromatogr. 407 (1987) 37. [41] C. Ottinger, J. Chem. Phys. 47 (1967) 1452. [42] M. Vestal, J.H. Futrell, J. Chem. Phys. 52 (1970) 978. [43] R. Zhang, F.E. Regnier, J. Proteome Res. 1 (2002) 139. [44] C.M. Chavez-Eng, M.L. Constanzer, B.K. Matuszewski, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 767 (2002) 117. [45] V. Vindenes, D. Jordbru, A.B. Knapskog, E. Kvan, G. Mathisrud, L. Slordal, J. Morland, Forensic Sci. Int. 219 (2012) 1.
