Phase II

Article pubs.acs.org/ac

Modular Microfluidic System for Emulation of Human Phase I/Phase II Metabolism Thomas Kampe,† Anna König,† Hendrik Schroeder,†,§ Jan G. Hengstler,∥ and Christof M. Niemeyer*,†,‡ †

Technische Universität Dortmund, Fakultät Chemie, Biologisch-Chemische Mikrostrukturtechnik, Otto Hahn Str. 6, 44227 Dortmund, Germany ‡ Karlsruhe Institute of Technology (KIT), Institute for Biological Interfaces (IBG 1), Hermann-von-Helmholtz-Platz, D-76344 Eggenstein-Leopoldshafen, Germany § Chimera Biotec GmbH, Emil-Figge-Str. 76 A, D-44227 Dortmund, Germany ∥ Leibniz-Institut für Arbeitsforschung (IfADo), Ardeystr. 67, 44139 Dortmund, Germany S Supporting Information *

ABSTRACT: We present a microfluidic device for coupled phase I/phase II metabolic reactions in vitro. The chip consists of microchannels, which are used as packed bed reactor compartments, filled with superparamagnetic microparticles bearing recombinant microsomal phase I cytochrome P450 or phase II conjugating enzymes (UDP-glucuronosyltransferase). Online coupling of the microfluidic device with LC/MS enabled the quantitative assessment of coupled phase I/phase II transformations, as demonstrated for two different substrates, 7-benzyloxy-4-trifluoromethylcoumarin (BFC) and dextromethorphan (DEX). In contrast, conventional sequential one-pot incubations did not generate measurable amounts of phase II metabolites. Because the microfluidic device is readily assembled from standard parts and can be equipped with a variety of recombinant enzymes, it provides a modular platform to emulate and investigate hepatic metabolism processes, with particular potential for targeted small-scale synthesis and identification of metabolites formed by sequential action of specific enzymes.

■

T

he analysis of hepatic phase I and phase II metabolism is of paramount importance for pharmaceutical research and drug development.1,2 Although human metabolic processes can often be emulated by animal testing, ethical concerns have led to increasing efforts for the establishment of in vitro test methods to refine, reduce, and replace animal studies.3−5 Microfluidic systems have been explored to emulate human metabolism on the basis of embedded human tissue samples,6−8 hepatocytes,9−12 or rat or human liver microsomes.13−18 This approach, however, does not permit investigation of distinctive transformations conducted by either phase I enzymes, such as cytochrome P450 isoforms (CYP), or phase II conjugating enzymes, such as UDP-glucuronosyltransferases (UGTs). In contrast, individual phase I transformations have been conducted in vitro using isolated enzymes immobilized in microfluidic systems.19 To the best of our knowledge, no microfluidic system is yet available for the coupled phase I/phase II transformation of selected substrates in vitro. We here present a modular microreactor system, which contains magnetic microbead-immobilized CYP and UGT isoforms and which is coupled online to an HPLC-MS instrument to facilitate quantitative assessment of metabolic phase I/phase II transformations by the sequential action of specific enzymes. © 2014 American Chemical Society

EXPERIMENTAL DETAILS

Enzyme Immobilization on Magnetic Beads. For immobilization of Supersomes (BD Biosciences) on streptavidin-coated 1−3 μm diameter superparamagnetic polyvinyl alcohol beads (M-PVA SAV2, Chemagen), 126 μL of the beads was washed twice with 800 μL of H2O using the aid of magnetic separation. The beads were resuspended in 126 μL of TBS buffer (0.1 M Tris, 0.15 M NaCl, pH 7.4 @ 37 °C). For the BFC assay, 90 μL of CYP3A4 (90 pmol) and 36 μL of UGT2B7 (0.18 mg) Supersomes (BFC assay (Figure 3)), and for the DEX assay (Figure 4), 90 μL of CYP2D6 (90 pmol) and 90 μL of UGT2B7 (0.45 mg) were mixed with the beads, and TBS buffer was added to a final volume of 750 μL. The mixture was incubated for 1 h at room temperature, washed twice with 750 μL of TBS using magnetic separation, and the final volume was adjusted with TBS to 450 μL. Enzyme Assays. Enzyme activities in suspension were conducted in TBS buffer (0.1 M Tris, 0.15 M NaCl, pH 7.4 at 37 °C). Typical samples contained supersomes, substrate, Received: December 14, 2013 Accepted: February 25, 2014 Published: March 5, 2014 3068

dx.doi.org/10.1021/ac404128k | Anal. Chem. 2014, 86, 3068−3074

Analytical Chemistry

Article

Figure 1. Microfluidic chip for coupled phase I/phase II metabolic transformations. The principle of the compartmentalized microreactor is sketched in (A), and the schematic drawing of the four-channel PMMA chip is shown in (B). In this work, only two of the four channels were used as compartments for enzyme-modified microbeads. The dimensions of one channel are 58.5 × 1.0 × 0.2 mm. Side (C) and top (D) view of a chip, in which two channels are connected by a bridging tube (orange arrow), while the tubings on the right are used as inlet (green arrow) from the syringe pump and outlet (red arrow) to the HPLC-MS instrument. Rectangular Nd magnets are arranged underneath the chip on an iron plate holder to enable immobilization of superparamagnetic microbeads, and four small magnets are placed on top to fix the assembled chip. For clarity of demonstration, only the lower channel is filled with particles on its right side. The microscope images in (E) show immobilized particles at two different magnifications.

NADP+, 3.3 mM G6P, 0.4 U/mL G6P-DH, 10 mM MgCl2 as well as the substrate (50 μM DEX or 250 μM BFC). A flow rate of 3 μL/min was used. LC-MS2 Analysis. All HPLC-MS2 analyses were performed on an Agilent 1200 series HPLC equipped with a 10-port-2position switching valve and a Thermo LCQ Advantage Max mass spectrometer with electrospray ionization. Briefly, a volume of either 5 or 20 μL sample was injected by the autosampler by switching the valve in position B (Figure 2). The sample was flushed onto the SPE cartridge (MercuryMS Strata-X SPE, 2.0 × 20 mm; 25 μm, Phenomenex, Aschaffenburg, Germany) by the loading pump with a flow rate of 2 mL/min loading buffer (95% water, 5% acetonitrile, formic acid, pH 3.1). After 12 s, the valve switched to position A, and the analytes were flushed backward from the SPE cartridge onto the analytical column (100 × 2.1 mm Kinetex PFP column; Phenomenex). Gradient elution was applied using solvents A (water containing 0.1% formic acid) and B (acetonitrile) with a 0.25 mL/min linear gradient of 5%−95% B in 42 s (BFC) or 30 s (DEX) For 5.5 min, 95% B was kept, and the condition was switched back to 5% for another 2 min. The mass spectrometer was operated in negative ion mode for BFC analytics, and data were collected in full scan mode. Data were visualized with extracted ion chromatograms of the following m/z values: CHC 187.0 ± 0.5, BFC 319 ± 0.5, HFC 229 ± 0.5, and HFCGlu 405 ± 0.5. The mass spectrometer was operated in positive ion mode for DEX analytics, and data were collected in full scan mode. Data were visualized with extracted ion chromatograms of following m/z values: DORGlu 434.0 ± 0.5, HOMGlu 420 ± 0.5, DOR 258 ± 0.5, HOM 244 ± 0.5, DEX 272 ± 0.5, and MOM 258 ± 0.5. The retention times of HFCGlu, CHC, HFC, and BFC were 3.8, 4.0, 4.4, and 5.3 min, respectively. The retention times of HOMGlu, DORGlu,

MgCl2 (10 mM), as well as cofactors. In the case of phase I reactions, a cofactor regenerating system comprising 0.1 mM NADP+, 3.3 mM glucose-6-phosphate, and 0.4 U/mL glucose6-phosphate dehydrogenase was used. In the case of phase II reactions, 0.25 μg/μL alamethicin and 2 mM UDP-glucuronic acid were used, and for coupled phase I/phase II reactions, all cofactors were present. Substrates were dissolved in methanol, and the final methanol concentration was adjusted to 2% for all reactions. Enzymes, substrates, and alamethicin (for phase II reactions) were mixed and preincubated for 10 min at 37 °C. The cofactor mix was preincubated separately for the same time period. Reactions were started by combining the enzyme/ substrate solution with the cofactor mix. Reactions were stopped by the addition of 1/10 of the reaction volume of 60% perchloric acid containing the internal standard, as detailed below. The samples were centrifuged at 2250g for 20 min, and the supernatant was transferred in a microplate and analyzed by HPLC-MS. Microfluidic Metabolism Experiments. Supersomemodified beads were manually injected into individual compartments of a four-channel PMMA chip (microfluidic chipshop, Jena, Germany) using a standard plastic syringe. Successful loading of the reactor was monitored by naked-eye inspection. Documentation by light microscopy was carried out with an Olympus IX81 microscope equipped with an UPLSAPO 10× objective and a Hamamatsu Orca R2 CCD camera (Figure 1E). Filled channels were connected with a short tubing (Agilent) using MiniLuer plugs (microfluidic chipshop). The same tubing and plugs were used to connect the inlet of the assembled chip with a syringe pump holding the cofactor/substrate solution and the outlet with the HPLC-MS2 instrument. The syringe pump was filled with 2 mL of the following cofactor/substrate solution: 2 mM UDPGA, 100 μM 3069



Article

Figure 2. Schematic representation of the online coupling of microfluidic chip and HPLC-MS instrumentation (A, B) and representative extracted ion chromatograms obtained from the substrate BFC and its metabolites HFCGlu and HFC (C) or DEX and its metabolites HOMGlu, DORGlu, DOR, MOM, and HOM (D). The metabolic transformations of BFC and DEX are illustrated in Figures 3 and 4, respectively (NL: normalization level). Note that the internal quantification standards CHC and LEV are injected by the autosampler during stage (B). Parent compounds DEX and BFC were not quantified because their concentration was outside the dynamic range of MS detection.

HOM, DOR, LEV, MOM, and DEX were 2.0, 2.0, 3.6, 3.6, 4.0, 4.2, and 4.3 min, respectively. To enable absolute quantification, calibration was conducted by aid of external standard samples containing known concentrations of the analytes (e.g., BFC, HFC, and HFCGlu) as well as a constant amount of an internal standard (3-cyano-7hydroxycoumarin (CHC) in the case of BFC, or levallorphan (LEV) in the case of DEX analytics). The calibration samples were analyzed by the same protocols and instrumentation parameters as described above. For online analysis of microfluidic fractions, the outlet of the microfluidic chip was connected to the sample loop through the switching valve, as schematically illustrated in Figure 2. During LC-MS analysis of a previous fraction, the microfluidic flow was passed through the sample loop into waste. At the beginning of a new cycle (valve is switched to position B), the fraction aliquot inside the sample loop is flushed on the SPE cartridge for further analysis.

In that case, the autosampler injected an internal standard solution containing the same concentration as used in the calibration standards. Additional information on data analysis, accuracy, and precision are given in the Supporting Information.

■

RESULTS AND DISCUSSION The prototype of our system consists of a poly(methyl methacrylate) (PMMA) chip containing four microchannels each having a volume of about 12 μL (Figure 1). By means of conventional tubing, the channels can be connected with external pumps and read-out instrumentation or else with each other to generate individual compartments of a packed bed reactor. The reaction compartments are filled with enzymes taking advantage of paramagnetic microbeads, which can be immobilized within the channels by an arrangement of rectangular Nd magnets. 3070



Article

Due to the commercial availability of a wide variety of recombinant CYP and UGT isoforms in the form of microsomes derived from baculovirus-infected insect cells (supersomes; BD Biosciences), we chose this enzyme formulation as catalyst for our system. It is noteworthy, however, that enzyme immobilization had proven extremely difficult. On the one hand, recombinant microsomes display a limited half-life of 1−3 h (Figure S1, Supporting Information), which puts severe limitations on the immobilization procedure. On the other hand, CYP and UGT have a directed orientation in the microsomal membrane, and thus, immobilization on a solid surface can reduce accessibility of the enzyme’s active site due to inappropriate orientation with respect to the solid support. Experimental evaluation of previously reported protocols for immobilization of metabolic enzymes in hydrogels like alginate,20,21 sol−gel materials,22 or Matrigel8 indicated that these are not suited for application in microfluidic systems. Mixing of the supersomes with hydrogel precursors and the concomitant dilution of the enzymes increased the bulk phase volumes and reduced specific activity, thereby leading to largely inactive packed-bed formulations (data not shown). We could eventually identify a suitable protocol, which is based on the physisorption of supersome membrane fragments on streptavidin-coated 1−3 μm diameter superparamagnetic polyvinyl alcohol beads (M-PVA SAV2, Chemagen). This procedure allowed for rapid catalyst immobilization along with acceptable loss in activity of about 40% (Figure S2, Supporting Information). Functional packed bed reactors were readily prepared by injecting 450 μL of enzyme-functionalized beads into the microchannels. Subsequent to filling, the chip was connected to a syringe pump, and a buffered solution containing substrate and all necessary cofactors was flushed through the reactor with a flow rate of 3 μL/min. The outlet of the reactor was connected to a conventional HPLC-MS2 system, equipped with a 10-port-2-position switching valve (Figure 2A). In the collection mode, the chip is in line with a 20 μL sample loop connected to the valve (red flow in Figure 2A), whereas in the meantime, a previously collected sample is analyzed by HPLCMS (green flow). At the beginning of a new injection step, the valve switches to flush the content of the sample loop to a solidphase extraction (SPE) precolumn, where hydrophobic analytes are trapped while buffer salts elute into waste (yellow flow, in Figure 2B). After 18 s, the valve is switched back to position (a), and analytes are backflush eluted from the precolumn onto the analytical column by gradient elution and detected by mass spectrometry. In parallel, the next fraction is collected. Injection of an internal standard by the autosampler during stage (b) allowed us to quantify metabolites by aid of calibration curves. Representative extracted ion chromatograms are shown in Figure 2C,D. For an initial test of our microfluidic system, we chose the model substrate BFC (7-benzyloxy-4-trifluoromethylcoumarin), which is debenzylated by CYP3A4 to generate HFC (7-hydroxy-4-trifluoromethylcoumarin). The latter is glucuronylated by UGT isoform 2B7 to form HFCGlu (4trifluoromethylumbelliferyl glucuronide), as schematically illustrated in Figure 3A.23,24 BFC is well suited to investigate coupled phase I/phase II reactions because of the extraordinary high turnover of the phase II reaction, which is required for a high coupling efficiency (Keff, see also Figure S3, Supporting Information). Both phase I (CYP3A4) and phase II (UGT2B7) supersomes were immobilized on magnetic microbeads, which

Figure 3. Representative data of microfluidic coupling of CYP3A4and UGT2B7-mediated BFC metabolism. (A) Reaction scheme and experimental setup of forward (phase I → phase II, black) and reversed flow orientation (phase II → phase I, gray). The curves indicate product formation of phase I metabolite HFC (B) and phase II metabolite HFCGlu (C). Error bars indicate standard error of the mean (n = 7 and 4 for forward and reversed flow, respectively), and dashed lines indicate the level of quantification (LOQ), determined for the respective analytes. The insets depict the average normalized product formation rate for all 40 collected microfluidic fractions. Note that no phase II product is generated in the reversed-flow direction.

were then filled in separate but fluidically connected channels of the chip, and 200 μM BFC solution was delivered in either CYP3A4→UGT2B7 (black curves in Figure 3) or reverse flow direction (gray curves). Online quantification of phase I (HFC, Figure 3B) and phase II (HFCGlu, Figure 3C) indicated that metabolite formation occurred with characteristic profile revealing a maximum intensity peak after about 60 min with subsequent slow decrease, presumably due to loss of enzyme activity. Concentration of phase II metabolite HFCGlu was approximately 3 times higher than that of phase I product HFC, leading to a coupling efficiency Keff of around 80%. To 3071



Article

Figure 4. Representative data of microfluidic coupling of CYP2D6- and UGT2B7-mediated DEX metabolism. (A) Reaction scheme and experimental setup of forward (phase I → phase II, black curves) and reverse (phase II → phase I, gray curves) oriented flow. (B) Average normalized product formation rate for all 40 collected microfluidic fractions. Product formation of phase I metabolite DOR (C), MOM (D), HOM (E) and phase II metabolite DORGlu (F). Error bars indicate standard error of the mean (n = 5 and 2 for forward and reversed flow experiments, respectively) and dashed lines indicate the level of quantification (LOQ), determined for the respective analytes. Note that no phase II product is generated in the reversed flow direction.

demonstrate the necessity of phase I→phase II directionality, controls with reverse orientation were conducted (gray curves in Figure 3). As expected, formation of HFC occurred with a similar profile (Figure 3B). Slightly lower amounts of HFC were produced in reverse flow direction, presumably due to increased adsorption of the very hydrophobic substrate BFC to the microbeads. HFCGlu formation was not detectable in reverse flow direction (Figure 3C). Repetition of these

experiments indicated good reproducibility of microfluidic CYP3A4→UGT2B7 coupling. To demonstrate that our microfluidic system is useful for investigation of more complex and practically relevant metabolic reactions, we chose the antitussive drug dextromethorphan (DEX) as a substrate, which is a common highly cough-suppressing over-the-counter drug.25 DEX and its metabolite dextrorphan (DOR) are neuromodulatory and show analgesic effects, and because of the psychoactive effects 3072



Article

couple the most relevant phase I and phase II enzymes with each other to enable identification of combinations which generate metabolite conjugates. Such data on interacting metabolizing enzymes would be most valuable for preclinical drug development. Moreover, our microchip platform should be applicable for metabolite identification, enzyme kinetics or inhibitor studies, and in particular, for the targeted small-scale synthesis of, for instance, radiolabeled metabolites.

when administered at higher dosages, DEX has become a popular juvenile drug in recent years.26,27 Phase I metabolism of DEX is mediated by CYP2D6 and CYP3A4 and leads to formation of DOR as main metabolite and 3-hydroxymorphinan (HOM) and 3-methoxymorphinan (MOM) as minor metabolites (Figure 4A). Although it is known that the intermediates are glucuronidated during phase II metabolism,28 the responsible UGT isoforms have not been described yet. Hence, we initially screened 12 recombinant human UGTs for glucuronidation of DOR and HOM (Figure S4). Among the five capable isoforms, UGT2B7 and UGT2B4 showed the highest activity, and a detailed characterization by kinetic analyses (Figure S5, Table S4) also revealed that these phase II transformations are sensitive to the albumin effect.29 As judged from the kinetic parameters, the metabolism of DEX by CYP2D6 and UGT2B7 was best suited for the demonstration of microfluidic coupling. To this end, two compartments of the microfluidic system were filled with enzyme-coated microbeads bearing CYP2D6 and UGT2B7, respectively. A substrate solution containing 200 μM DEX was delivered microfluidically and concentrations of DOR, MOM, HOM, and dextrorphan-O-glucuronide (DORGlu) products were quantified by online LC/MS (Figure 4C− F, respectively). Similar to that observed for BFC metabolism, formation of products followed a characteristic profile, and reversal of the substrate flow from phase II → phase I compartment (gray curves) led to no formation of phase II product, as expected. Despite the fact that metabolite formation of both enzymes was increased by the presence of bovine serum albumin (BSA) in suspension assays (Figures S5, S6), the addition of BSA to the substrate solution led to no significant effect in the microfluidic setting (data not shown). The successful generation of phase II product DORGlu in our compartmentalized microreactor (Figure 4F) is remarkable, because no phase II product was obtained when the coupled phase I/phase II transformation of DEX was carried out in a one-pot incubation with a mixture of CYP2D6- and UGT2B7Supersomes in homogeneous suspension (Figure S7, Supporting Information). We speculate that concentrations of phase I intermediate are too low to enable efficient phase II conversions in this one-pot assay. Evidence for the formation of CYP-UGT protein−protein association30 suggests that dynamic assembly of metabolic multienzyme complexes in the cell membrane and the concomitant increase in local concentrations of intermediate products could improve the low reaction rates of isolated enzymes. Therefore, our results suggest that fluidic coupling by laminar flow enables a more efficient transport of substrates than random diffusion. In general, compartmentalized reactors offer the principal advantages that (i) reactions occur in separated spatial confinements to avoid cross-talk and formation of sideproducts, (ii) individual reaction compartments can be coupled by simple logic control of fluidics, and (iii) additional modules can conveniently be embedded to enable processsing, such as separation of intermediates, addition of substrates and cofactors, or online detection and quantification of reaction compounds. The compartmentalized microchip presented here can be used for the fluidic coupling of phase I and II enzymatic transformations and online detection of metabolites. Because it can be readily equipped with a variety of recombinant microsomal enzyme preparations, it provides a highly modular platform to emulate and investigate processes of the liver metabolism. The microchip offers the perspective to fluidically

■

ASSOCIATED CONTENT

S Supporting Information *

LC-MS2 data analysis, characterization of enzymes, Figures S1− 7, and Tables S1−4. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel./Fax: + 49 (0)721-608-2-5546. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Bundesministerium für Bildung und Forschung (BMBF) in the course of the program “Ersatzmethoden zum Tierversuch” (project no. 0315358).

■

REFERENCES

(1) Ruiz-Garcia, A.; Bermejo, M.; Moss, A.; Casabo, V. G. J. Pharm. Sci. 2008, 97, 654−690. (2) Strolin Benedetti, M.; Whomsley, R.; Poggesi, I.; Cawello, W.; Mathy, F.-X.; Delporte, M.-L.; Papeleu, P.; Watelet, J.-B. Drug Metab. Rev. 2009, 41, 344−390. (3) Leenaars, M.; Hooijmans, C. R.; van Veggel, N.; ter Riet, G.; Leeflang, M.; Hooft, L.; van der Wilt, G. J.; Tillema, A.; RitskesHoitinga, M. Lab. Anim. 2012, 46, 24−31. (4) Hendriksen, C. F. ILAR J. 2002, 43 (Suppl), S43−S48. (5) Flecknell, P. ALTEX 2002, 19, 73−78. (6) van Midwoud, P. M.; Merema, M. T.; Verpoorte, E.; Groothuis, G. M. Lab Chip 2010, 10, 2778−2786. (7) van Midwoud, P. M.; Janssen, J.; Merema, M. T.; de Graaf, I. A.; Groothuis, G. M.; Verpoorte, E. Anal. Chem. 2011, 83, 84−91. (8) van Midwoud, P. M.; Merema, M. T.; Verweij, N.; Groothuis, G. M. M.; Verpoorte, E. Biotechnol. Bioeng. 2011, 108, 1404−1412. (9) Mahler, G. J.; Esch, M. B.; Glahn, R. P.; Shuler, M. L. Biotechnol. Bioeng. 2009, 104, 193−205. (10) Sung, J. H.; Kam, C.; Shuler, M. L. Lab Chip 2010, 10, 446− 455. (11) Toh, Y. C.; Lim, T. C.; Tai, D.; Xiao, G.; van Noort, D.; Yu, H. Lab Chip 2009, 9, 2026−2035. (12) Viravaidya, K.; Sin, A.; Shuler, M. L. Biotechnol. Prog. 2004, 20, 316−323. (13) Ma, B.; Zhang, G.; Qin, J.; Lin, B. Lab Chip 2009, 9, 232−238. (14) Mao, S.; Gao, D.; Liu, W.; Wei, H.; Lin, J.-M. Lab Chip 2012, 12, 219−226. (15) Kim, H. S.; Wainer, I. W. Anal. Chem. 2006, 78, 7071−7077. (16) Sakai-Kato, K.; Kato, M.; Toyo’oka, T. J. Chromatogr. A 2004, 1051, 261−266. (17) Tanvir, S.; Adenier, H.; Pulvin, S. Enzyme Microb. Technol. 2009, 45, 112−117. (18) van Liempd, S. M.; Kool, J.; Reinen, J.; Schenk, T.; Meerman, J. H.; Irth, H.; Vermeulen, N. P. J. Chromatogr. A 2005, 1075, 205−212.

3073



Article

(19) Nicoli, R.; Bartolini, M.; Rudaz, S.; Andrisano, V.; Veuthey, J. L. J. Chromatogr. A 2008, 1206, 2−10. (20) Haumont, M.; Magdalou, J.; Ziegler, J. C.; Bidault, R.; Siest, J. P.; Siest, G. Appl. Microbiol. Biotechnol. 1991, 35, 440−446. (21) Sukumaran, S. M.; Potsaid, B.; Lee, M. Y.; Clark, D. S.; Dordick, J. S. J Biomol. Screen. 2009, 14, 668−678. (22) Sakai-Kato, K.; Kato, M.; Homma, H.; Toyo’oka, T.; Utsunomiya-Tate, N. Anal. Chem. 2005, 77, 7080−7083. (23) Renwick, A. B.; Surry, D.; Price, R. J.; Lake, B. G.; Evans, D. C. Xenobiotica 2000, 30, 955−969. (24) Baranczewski, P.; Kallin, A.; Andersson, A.; Hagigi, S.; Aberg, M.; Postlind, H.; Mankowitz, L. Assay Drug. Dev. Technol. 2004, 2, 345−352. (25) Brown, C.; Fezoui, M.; Selig, W. M.; Schwartz, C. E.; Ellis, J. L. Br. J. Pharmacol. 2004, 141, 233−240. (26) Drugs of Abuse, 2011 ed.; U.S. Department of Justice, Drug Enforcement Administration; U.S. Government Printing Office: Washington, DC, 2012 (27) Bryner, J. K.; Wang, U. K.; Hui, J. W.; Bedodo, M.; MacDougall, C.; Anderson, I. B. Arch. Pediatr. Adolesc. Med. 2006, 160, 1217−1222. (28) Lutz, U.; Volkel, W.; Lutz, R. W.; Lutz, W. K. J. Chromatogr. B 2004, 813, 217−225. (29) Rowland, A.; Knights, K. M.; Mackenzie, P. I.; Miners, J. O. Drug Metab. Dispos. 2008, 36, 1056−1062. (30) Takeda, S.; Ishii, Y.; Iwanaga, M.; Nurrochmad, A.; Ito, Y.; Mackenzie, P. I.; Nagata, K.; Yamazoe, Y.; Oguri, K.; Yamada, H. Mol. Pharmacol. 2009, 75, 956−964.

3074


Phase II

Recommend Documents