Microarray of Human P450 with an Integrated Oxygen Sensing Film for

May 8, 2012 - National Institute of Advanced Industrial, Science and Technology (AIST), ... the activities of diverse P450 isoforms in a high-throughp...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/ac

Microarray of Human P450 with an Integrated Oxygen Sensing Film for High-Throughput Detection of Metabolic Activities Gang Chang,†,⊥ Yoshinao Mori,†,‡ Saori Mori,† Takashi Irie,† Hidenori Nagai,† Tatsushi Goto,§ Yoshiro Tatsu,† Hiromasa Imaishi,*,‡,§ and Kenichi Morigaki*,†,‡,§ †

National Institute of Advanced Industrial, Science and Technology (AIST), Midorigaoka, Ikeda 563-8577, Japan Graduate School of Agricultural Science, Kobe University, Rokkodaicho 1-1, Nada, Kobe 657-8501 Japan § Research Center for Environmental Genomics, Kobe University, Rokkodaicho 1-1, Nada, Kobe 657-8501 Japan ⊥ Ministry-of-Education Key Laboratory for the Green Preparation and Application of Functional Materials, Faculty of Materials Sciences and Engineering, Hubei University, No.11 Xueyuan Road, Wuchang, Wuhan 430062, China ‡

S Supporting Information *

ABSTRACT: A microarray chip containing human P450 isoforms was constructed for the parallel assay of their metabolic activities. The chip had microwells that contained vertically integrated P450 and oxygen sensing layers. The oxygen sensing film was made of an organically modified silica film (ORMOSIL) doped with tris(4,7-diphenyl-1,10-phenanthroline) ruthenium dichloride (Ru(dpp)3Cl2). Human P450s (23 types) expressed in E. coli and purified as membrane fractions were immobilized in agarose matrixes on the oxygen sensing layer. The activities of P450s were determined by evaluating the fluorescence intensity enhancement of the oxygen sensor due to the oxygen consumption by the metabolic reaction. By normalizing the responses with the amounts of oxygen sensor and P450 enzymes in microwells, we could obtain fluorescence enhancement patterns that were characteristic to the combination of P450 isoforms and substrate material. The patterns obtained from two psoralen derivatives resembled each other, whereas a structurally different substrate (capsaicin) resulted in a distinct pattern. These results suggest the potential of the microarray to analyze the activities of diverse P450 isoforms in a high-throughput fashion. Furthermore, mechanism-based inactivation (MBI) of P450 could be detected by successively incubating a chip with different substrate solutions and measuring the residual activities.

T

he activities of cytochrome P450 (P450) enzyme super family have been extensively studied, because they are crucial in the metabolism of drugs and dietary materials in the human body.1,2 The potential applications range from identification of toxicity in drug candidates in an early stage of the drug development to food and environmental safety inspections.3−5 Despite the recent instrumental developments (including HPLC-MS), the gap between the number of potential candidates for metabolic inspections and the number of samples that can be handled is widening.3 To amend the bottleneck, high-throughput assaying platforms using P450 microarrays have been developed.5−10 However, most of the assays are based on the evaluation of P450 activities using fluorogenic substrates (e.g., 7-ethoxycoumarin (7-EC)), which are applicable only to a limited number of P450 isoforms.6,8,9 Detecting P450 activities with an oxygen sensing film should solve the problem of fluorogenic substrate, because all metabolic reactions catalyzed by P450 involve oxygen consumption given as the following reaction equation.

Ru and Pt complexes, whose luminescence is quenched by oxygen, can be used for monitoring oxygen consumption, following the Stern−Volmer equation and evaluating the dissolved oxygen concentration.11−13 The use of optical biosensors for monitoring the oxygen consumption by enzymes and microorganisms has been studied extensively.14−21 There have been some studies that detected P450 activities by monitoring oxygen consumption in multiwell plates using suspensions of membrane fractions.22 However, existing approaches using an oxygen sensing layer have suffered from low sensitivity. Recently, we developed a methodology to immobilize P450s on the surface of the oxygen sensor and applied the integrated structure to the evaluation of the metabolism by several P450 isoforms.10 We demonstrated that vertical integration of immobilized P450 and oxygen sensor could improve the sensitivity. Here, we report on the development of a P450 microarray based on the vertically integrated P450 and oxygen sensing

R‐H + O2 + NADPH + H+

Received: February 6, 2012 Accepted: May 8, 2012 Published: May 8, 2012

+

→ R‐OH + H 2O + NADP

© 2012 American Chemical Society

5292

dx.doi.org/10.1021/ac300355w | Anal. Chem. 2012, 84, 5292−5297

Analytical Chemistry

Article

of 0.2 to 0.3. After the addition of isopropyl-1-thio-β-Dgalactopyranoside (1 mM) and delta-amino levulinic acid (0.5 mM), the cultures were grown at 25 °C for 24 h. The cells were centrifuged and suspended in 25 mL of 100 mM potassium phosphate buffer (pH 7.25) containing 20% glycerol. Preparation of E.coli membrane fractions were carried out according to the method of Goto et al.25 The presence of P450 isoforms was confirmed by reduced CO difference spectra, according to the previously reported procedure.26 Furthermore, we tested the activity of P450 isoforms by applying various substrates and detecting the conversion by HPLC (Supporting Information Table S1). Fabrication of Microwells in Cyclo-Olefin-Polymer Plates. P450 arrays were generated on the surface of a thermoplastic biocompatible polymer (cyclo-olefin polymer (COP)) with a thickness of 2 mm (ZEONOR 1060R, Zeon, Tokyo, Japan). Microwells were fabricated by cutting work with a numerical control machine that was automatically operated using CAD programs (Micro MC-2, PMT Corp., Fukuoka, Japan). The diameter and depth of the microwells were 2 and 1.5 mm, respectively. The bottom of microwells had a round shape (Figure 1).

layer in microwells. The microwells (width 2 mm, depth 1.5 mm) were generated on a cyclo-olefin-polymer (COP) plate by the numerical-control-machining. The oxygen sensing film was made of an organically modified silica film (ORMOSIL) doped with tris(4,7-diphenyl-1,10-phenanthroline) ruthenium dichloride (Ru(dpp)3Cl2). Twenty-three types of human P450s and a generic P450 reductase were expressed in the E. coli inner membrane, purified as membrane fractions, and immobilized in agarose matrixes on the surface of the oxygen sensing layer in microwells. The activities of P450s were determined by evaluating the fluorescence intensity enhancement of the oxygen sensor due to the oxygen consumption by the metabolic reaction. Combinations of the P450 isoforms and substrates resulted in unique patterns of fluorescence enhancement, which should convey information on the degree of metabolic activities by P450 isoforms toward the substrates tested. One advantage of the present method is that it can assess the metabolic activities of diverse P450 isoforms, unlimited by the specificity of P450 isoforms toward probe materials (fluorogenic substrates). The microarray generated on a COP plate also significantly reduced the volume of samples required for each well (from 250 μL to less than 4 μL). Furthermore, mechanism-based inactivation (MBI) of P450 could be detected by successively incubating a chip with different substrate solutions and measuring the residual activities, extending the utility of the P450 microarray.

■

MATERIALS AND METHODS Materials. Tetraethyl orthosilicate (TEOS), triethoxy (octyl) silane (Octyl-triEOS), and agarose (Type VII) were purchased from Sigma-Aldrich. Tris(4,7-diphenyl-1,10-phenanthroline) ruthenium dichloride (Ru(dpp)3Cl2), isopropyl-1thio-β-D-galactopyranoside, delta-amino levulinic acid, ethanol, methanol, and concentrated hydrochloric acid were obtained from Wako Pure Chemical Industries (Osaka, Japan). Coumarin, testosterone, β-nicotinamide adenine dinucleotide phosphate tetrasodium salt (NADPH), potassium dihydrogenphosphate, and dipotassium hydrogenphosphate were purchased from Nacalai Tesque (Kyoto, Japan). Glucose-6phosphate (G6P), capsaicin, 5-methoxypsoralen (5-MOP), and 8-methoxypsoralen (8-MOP) were purchased from Tokyo Chemical Industry (Tokyo, Japan). Glucose-6-phosphate dehydrogenase (G6PD) was purchased from Toyobo (Osaka, Japan). Bergamottin was purchased from Indofine Chemical Company (Hillsborough, NJ). All chemicals and solvents were reagent grade and used without further purification. Milli-Q water with the resistivity of more than 18 MΩ cm was used to prepare aqueous solutions. Preparation of Membrane Fraction of Human P450 and P450 Reductase from E.coli. Vectors for expression of human P450 genes in E. coli were constructed. We first subjected the genes encoding each human P450 by RT-PCR using a human cDNA library. The amplified fragments of each human cDNA were modified by PCR to create restriction enzyme sites and modified N-terminal amino acid residues.23 The amplified fragments were cloned into the SalI/SpeI and NdeI sites of the pCW vector, which contains a cDNA of human NADPH-cytochrome P450 reductase and was constructed as described by Iwata et al.24 E. coli JM109 cells were transformed with a pCW vector containing a P450 insert. The transformed cells were preincubated in Luria broth (LB) at 37 °C for 1 day, and then, 3 mL of this culture was added to 500 mL of terrific broth (TB) and incubated at 37 °C to an OD600

Figure 1. (A) Schematic illustration of P450 microarray: Oxygen sensing layer and P450 immobilized in agarose gel were integrated in each microwell. (B) Photograph of a P450 microarray plate compared with a Japanese coin (100 yen). (C) P450 assay procedure using the microarray. The substrate solution was first put on the surface of the chip and introduced into microwells with the aid of gentle centrifugation. Then, the chips were sealed with a sealing sheet to isolate the wells. It should be noted that the enzymatic reaction starts as the substrate solution is introduced into microwells, before the sealing of the chip and measurement by the fluorescence microscopy.

Fabrication of the Oxygen Sensing Layer. The fluorophore doped solution was prepared as described in ref 11 with small modifications as follows. TEOS (0.29 mL) was mixed with 0.612 mL of octyl-triEOS, 0.625 mL of ethanol, and 0.2 mL of 0.1 M HCl by stirring for 30 min under ambient condition. Then, 1.725 mL of ethanol was added into the solution for dilution to improve the quality of the oxygen sensing film. The solution was further stirred for 1 h. For 5293

dx.doi.org/10.1021/ac300355w | Anal. Chem. 2012, 84, 5292−5297

Analytical Chemistry

Article

preparing Ru(dpp)3Cl2 doped solution, 100 μL of 2 mM Ru(dpp)3Cl2 in ethanol was mixed with 300 μL of the abovementioned solution. This solution was vortexed for 1 min, and 2 μL of the fluorophore doped solution was pipetted into each well of the microplate. The thickness of the oxygen sensing layer was less than 10 μm based on the amount of the solution/ dye mixture deposited at the bottom.13,20 The microplate was stored in the dark under ambient condition for gelling and aging for 6 days. Immobilization of P450 in Agarose Gel in Microwells. Agarose was dissolved in 0.1 M potassium phosphate buffer (KPB) to form 0.8% (w/w) solution at 60 °C. This solution was cooled down to ca. 37 °C. Twenty microliters P450 suspensions were mixed with 20 μL of 0.8% agarose solution (the final concentration was 0.4%), and then, 2 μL of P450 doped agarose solution was pipetted onto the surface of the oxygen sensing layer in each microwell. The chip was centrifuged for 15 min at 2000 rpm (4 °C) to settle P450/ agarose at the bottom of the well . The microarrays were kept at 4 °C for 3−6 days for gelation. P450 Assays in Microarrays. The measurements of P450 activities were conducted by introducing a substrate solution into microwells and sealing the wells with a sealing sheet (Figure 1C). For the substrates, we used capsaicin, 8-MOP, and 5-MOP. The reaction solution contained 0.1 mM KPB, 0.1 mM NADPH, 3 mM MgCl2, 3 mM G6P, and 0.4 U/mL G6PD. (These concentrations were set from preliminary experiments to ensure that the supply of NADPH is not rate-limiting.) The substrates were dissolved in dimethylsulfoxide (DMSO) and added to the reaction solution so that the final concentration was 0.2 mM (the final DMSO concentration was 2% (v/v)). The reaction was initiated by adding 530 μL of the reaction solution to the surface of the chip. A gentle centrifugation was applied for introducing the substrate solution into each microwell and expelling air bubbles. Then, the wells were sealed with a sealing sheet (PET) to prevent the diffusion of oxygen from the atmosphere. After sealing the wells, the chip was quickly placed on the fluorescence microscope (BX51WI, Olympus) for the automated fluorescence intensity measurement. The microarray was positioned on the microscopy stage, and fluorescence intensity of the oxygen sensor in each microwell was measured every 4 min for 20 min. (We started the measurements 20 min after the introduction of reaction solution onto the microarray.) Fluorescence microscopy images were obtained with a CCD camera (DP-50, Olympus) and processed by MetaMorph program (Molecular Devices, Sunnyvale, CA).

The responses of P450 isoforms were measured using three model substrates found in food (capsaicin, 5-MOP, and 8MOP: Figure 2). Figure 3 shows the fluorescence micrographs

Figure 2. Chemical structures of the substrates tested by the P450 microarray: (A) capsaicin, (B) 5-MOP, and (C) 8-MOP.

Figure 3. Fluorescence micrographs of microwells (5 × 5) containing 23 different P450 isoforms and two controls were compared before and after the incubation (A) with 0.2 mM capsaicin and (B) without capsaicin. In some wells, significant enhancement of the fluorescence intensities was observed due to the oxygen consumption.

■

of each well during the measurement for 20 min. It should be noted that the measurements were started 20 min after the initial introduction of the substrate solution into the wells (the time was necessary for the introduction, sealing, and positioning of the wells under the microscope). Figure 3A are the responses in the presence of 0.2 mM capsaicin, whereas Figure 3B is the response without substrates (except for the substrate, all other ingredients, including NADPH, were present). The fluorescence intensity of oxygen sensor in the wells increased because of the decrease of oxygen concentration, which should be correlated with the enzymatic activities. By comparing Figure 3A,B, it can be clearly seen that considerable enhancement of fluorescence intensities was observed in some microwells in the presence of capsaicin. We normalized the fluorescence intensities with the initial values to eliminate

RESULTS AND DISCUSSION Responses of P450 Microwell Array toward Substrates. A schematic illustration of the P450 microarray is shown in Figure 1A. Each microwell contains two layers, one being agarose gel with immobilized P450 and the other being the oxygen sensor. The actual size of a P450 array is shown together with a Japanese coin (Figure 1B). This configuration can simultaneously measure the activities of P450 up to 25 isoforms. In the present study, we incorporated 23 human P450s in 23 wells. The 24th and 25th wells contained control membrane fractions without P450 (pCW) and CYP1A1 (the same P450 isoforms as the first well to check the possible changes of enzymatic activities during the microarray fabrication process), respectively. 5294

dx.doi.org/10.1021/ac300355w | Anal. Chem. 2012, 84, 5292−5297

Analytical Chemistry

Article

primarily from the lack of normalization. One should obtain more consistent results, if one applies the normalization of data. We obtained relative activities of 23 P450 isoforms for the three model substrates (5-MOP, 8-MOP, and capsaicin) (Figure 5). It is interesting to note that 5-MOP and 8-MOP,

the data variation due to a slightly different amount of oxygen sensing layer present in each well (Supporting Information Figure S1). The fluorescence intensity in each microwell increased with time, depending on the P450 isoforms. The increases of fluorescence intensities were observed also without the substrate, and it was regarded to be due to the background oxygen consumptions by either P450 (uncoupling) or other intrinsic proteins in E. coli inner membranes. Figure 4A,B shows

Figure 5. Relative activities of 25 microwells (23 P450 isoforms, pCW, and CYP1A1 (duplicate)) toward different substrates: (A) capsaicin, (B) 5-MOP, and (C) 8-MOP. Figure 4. Changes of the fluorescence intensities in microwells during the incubation (20 min): (A) with 0.2 mM capsaicin and (B) without capsaicin (background). The “relative activity” was calculated by dividing the fluorescence increase with capsaicin by the background (C).

which are structural isomers, showed patterns that resembled each other. The patterns for 5-MOP and 8-MOP showed a higher correlation of fluorescence increase to each other (r = 0.912) compared with capsaicin (5-MOP vs capsaicin: r = 0.809; 8-MOP vs capsaicin: r = 0.799). To the best of our knowledge, this is the first attempt to directly evaluate a large variety of human P450 isoforms (23 types) using the oxygen consumptions in a single microarray. Although it is rather premature for generalization, the different P450 response profiles for 5-MOP, 8-MOP, and capsaicin may indicate the possibility that we could correlate the chemical types or structures of the substrate with the oxygen consumption patterns. One technological challenge for the present P450 microarray is the background oxygen consumption caused by either the uncoupling reactions of P450 or enzymatic activities caused by intrinsic substances (e.g., fatty acids) present in the membrane fragments. Since each P450 isoform has a different degree of uncoupling reactions, the responses for different P450 isoforms and substrates may be affected by different levels of background oxygen consumption. It prevents fully quantitative analyses of the metabolic activities from the oxygen consumption patterns. However, The patterns should at least qualitatively indicate the extent of P450 catalysis by different isoforms toward a specific substrate, as supported by the high correlation with the HPLC results.10 Therefore, P450-oxygen sensor arrays should provide useful information on the metabolic activities of various types of P450s toward diverse chemicals. It should give us the possibility to screen the P450 activities toward new compounds

the increase of fluorescence for each P450 isoform with and without capsaicin, respectively. Since we did not adjust the amount of P450 in the preparation, the background oxygen consumption was used for normalizing the differences in the amount of P450 for each isoforms, assuming that the amount of P450 enzymes were proportional to the degree of background reaction for the same P450 isoforms. The fluorescence increase in the presence of a substrate was divided with the background fluorescence increase, yielding the “relative activity”, which should represent the degree of metabolic reaction by P450. The relative activities obtained from the data in Figure 4A,B are shown in Figure 4C. This normalization could significantly reduce the data fluctuation due to differences of P450 amounts in different samples (Supporting Information Figure S2). Without the normalization, the obtained patterns varied significantly, depending on the amount of P450 enzymes in the sample (Supporting Information Figure S2). For example, the profiles of selected P450 isoforms toward capsaicin obtained in our previous work looked significantly different from the present results.10 However, the results in the previous study were presented without normalization of the data with the background oxygen consumption. Therefore, the discrepancy between these two set of results should have arisen 5295

dx.doi.org/10.1021/ac300355w | Anal. Chem. 2012, 84, 5292−5297

Analytical Chemistry

Article

Figure 6. Measuring MBI of P450 isoforms caused by 8-MOP and bergamottin: P450 microarrays containing four selected P450 isoforms and pCW (control) were incubated with either 8-MOP or bergamottin (2 mM) for 30 min. The microarrays were rinsed with KPB solution, and the residual enzymatic activities were examined. Coumarin and testosterone (0.5 mM) were used as the assaying substrate for the MBI by 8-MOP and bergamottin, respectively. The obtained responses were normalized with the control experiment, in which P450 microarrays were incubated with a KPB solution without substrates. (A) MBI by 8-MOP and (B) MBI by bergamottin.

show that 8-MOP inhibited CYP2A6 selectively, whereas bergamottin inhibited CYP2A6, 3A4, and 3A5. Other P450 isoforms tested were not affected by the incubation with these substances. These results agree with the previous findings that 8-MOP and bergamottin had MBI effects on 2A6 and 3A4, respectively.28,30 Therefore, using the P450 microarray, we should be able to assess the potential risk of MBI by the substances in food or drugs with a high throughput.

in a high-throughput fashion, especially if we have to study the activities of a large variety of P450 isoforms, including human single nucleotide polymorphism (SNP). One approach to amend the technical problem associated with the background oxygen consumption is to optimize the NADPH concentration, since the degree of background can be suppressed by lowering the NADPH concentration used in the assay.10 (It should be noted that P450 catalytic rates could be affected, if the NADPH concentration is too low. We found that the NADPH concentration could be reduced down to 0.1 mM in the present regeneration system without significantly lowering the P450 catalysis rates (data not shown).) Another possibility is to evaluate the P450 activities by measuring both oxygen and NADPH consumption in the presence of reactive oxygen species quenchers (superoxide dismutase and catalase), as recently shown by Traylor et al.,27 although simultaneous measurements of oxygen and NADPH consumption in microwells should pose technological challenges. Assaying the Mechanism-Based Inactivation (MBI) of P450. One advantage of immobilizing P450 in microwells is the possibility to exchange solutions on the chip and repeat the assays. An application of this possibility is the measurement of mechanism-based inactivation (MBI) of P450. MBI is a commonly observed route of P450 inactivation, which occurs due to the covalent binding of the reaction product to P450. It is responsible for a wide range of negative effects arising from combined uptake of dietary materials and drugs (a prominent instance is the uptake of grapefruit juice and some types of drugs).28 We assessed MBI of P450 isoforms using P450 arrays as follows. (The general procedures are depicted in Supporting Information, Figure S3.) First, a substrate solution, whose MBI potency is to be measured, is introduced to the chip and incubated for a defined period to allow the enzymatic reactions (and possible MBI) to proceed. Then, the chip is rinsed extensively with a buffer solution to remove the first substrate. Subsequently, the second substrate solution is introduced to assess the residual enzymatic activities. The second substrate is used solely for the assessment of P450 activities, so it should not cause MBI to the P450 isoforms. Figure 6 shows the data obtained for 8-MOP and bergamottin, which are commonly found natural products in food and known to cause MBI to some P450 isoforms.28,29 Coumarin and testosterone were used as the assaying substrate for evaluating the degree of MBI by 8MOP and bergamottin, respectively. The results in Figure 6

■

CONCLUSIONS

■

ASSOCIATED CONTENT

We developed a P450-microarray that can evaluate the metabolic activities of a wide variety of P450 by integrating immobilized P450 and an oxygen sensing layer in microwells. Using the oxygen sensing platform, we could assess the activities of P450 isoforms irrespective of their substrate specificities. We fabricated a parallel array of 23 human P450s expressed in E. coli. and obtained fluorescence enhancement patterns that were specific to the combination of P450 isoforms and substrate materials. Although oxygen consumption is not quantitatively linked with the enzymatic conversion of substrates, its correlation to the enzymatic activities makes the integrated P450 chip a uniquely useful tool to evaluate the activities of diverse P450 isoforms. Furthermore, an immobilized array of P450 could be used to detect MBI of P450 isoforms by chemicals commonly found in food. Miniaturization of the chip by assembling microwells on the COP plate enabled assays with significantly less sample volume compared with the commercial microwell plates (less than 4 μL of the assay solution for each microwell), which also resulted in a shorter assay time (less than 30 min compared with 1 h for commercial microwell plates). Further downsizing of the chip should be possible by applying microfabrication and printing technologies. Such microarrays may find applications in clinical and drug development settings by incorporating diverse genetic variants of human P450, which are closely related with the variable effects of medicinal materials to individuals.

S Supporting Information *

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. 5296

dx.doi.org/10.1021/ac300355w | Anal. Chem. 2012, 84, 5292−5297

Analytical Chemistry

■

Article

(26) Imaishi, H.; Matsuo, S.; Swai, E.; Ohkawa, H. Biosci. Biotechnol. Biochem. 2000, 64, 1696−1701. (27) Traylor, M. J.; Ryan, J. D.; Arnon, E. S.; Dordick, J. S.; Clark, D. S. J. Am. Chem. Soc. 2011, 133, 14476−14479. (28) Koenigs, L. L.; Trager, W. F. Biochemistry 1998, 37, 10047− 10061. (29) Hollenberg, P. F.; Kent, U. M.; Bumpus, N. N. Chem. Res. Toxicol. 2008, 21, 189−205. (30) Yukinaga, H.; Takami, T.; Shioyama, S.-h.; Tozuka, Z.; Masumoto, H.; Okazaki, O.; Sudo, K. Chem. Res. Toxicol. 2007, 20, 1373−1378.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (H.I.); [email protected]. ac.jp (K.M.). Fax: +81-78-803-5940 (H.I.); +81-78-803-5941 (K.M.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS G.C., Y.M., S.M., and T.I. contributed equally to this work. This work was financially supported by Program for Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN). Discussions with Dr. Kunio Isshiki and Dr. Akira Arisawa (Mercian Corp.) were appreciated.

■

REFERENCES

(1) Gueguen, Y.; Mouzat, K.; Ferrari, L.; Tissandie, E.; Lobaccaro, J. M. A.; Batt, A. M.; Paquet, F.; Voisin, P.; Aigueperse, J.; Gourmelon, P.; Souidi, M. Ann. Biol. Clin. 2006, 64, 535−548. (2) Zhou, S. F.; Liu, J. P.; Chowbay, B. Drug Metab. Rev. 2009, 41, 89−295. (3) Bistolas, N.; Wollenberger, U.; Jung, C.; Scheller, F. W. Biosens. Bioelectron. 2005, 20, 2408−2423. (4) Walsky, R. L.; Boldt, S. E. Curr. Drug Metab. 2008, 9, 928−939. (5) Tsotsou, G. E.; Cass, A. E. G.; Gilardi, G. Biosens. Bioelectron. 2002, 17, 119−131. (6) Sakai-Kato, K.; Kato, M.; Homma, H.; Toyo’oka, T.; Utsunomiya-Tate, N. Anal. Chem. 2005, 77, 7080−7083. (7) Ansede, J. H.; Thakker, D. R. J. Pharm. Sci. 2004, 93, 239−255. (8) Lee, M. Y.; Kumar, R. A.; Sukumaran, S. M.; Hogg, M. G.; Clark, D. S.; Dordick, J. S. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 59−63. (9) Lee, M. Y.; Park, C. B.; Dordick, J. S.; Clark, D. S. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 983−987. (10) Chang, G.; Morigaki, K.; Tatsu, Y.; Hikawa, T.; Goto, T.; Imaishi, H. Anal. Chem. 2011, 83, 2956−2963. (11) Tang, Y.; Tehan, E. C.; Tao, Z. Y.; Bright, F. V. Anal. Chem. 2003, 75, 2407−2413. (12) Lin, L.; Xiao, L. L.; Huang, S.; Zhao, L.; Cui, J. S.; Wang, X. H.; Chen, X. Biosens. Bioelectron. 2006, 21, 1703−1709. (13) Pang, H. L.; Kwok, N. Y.; Chow, L. M. C.; Yeung, C. H.; Wong, K. Y.; Chen, X.; Wang, X. R. Sens. Actuators, B: Chem. 2007, 123, 120− 126. (14) Preininger, C.; Klimant, I.; Wolfbeis, O. S. Anal. Chem. 1994, 66, 1841−1846. (15) Shi, Y. N.; Seliskar, C. J. Chem. Mater. 1997, 9, 821−829. (16) Borisov, S. M.; Wolfbeis, O. S. Chem. Rev. 2008, 108, 423−461. (17) Li, X. M.; Ruan, F. C.; Ng, W. Y.; Wong, K. Y. Sens. Actuators, B: Chem. 1994, 21, 143−149. (18) John, G. T.; Klimant, I.; Wittmann, C.; Heinzle, E. Biotechnol. Bioeng. 2003, 81, 829−836. (19) Kwok, N. Y.; Dong, S. J.; Lo, W. H.; Wong, K. Y. Sens. Actuators, B: Chem. 2005, 110, 289−298. (20) Pang, H. L.; Kwok, H. Y.; Chan, P. H.; Yeung, C. H.; Lo, W. H.; Wong, K. Y. Environ. Sci. Technol. 2007, 41, 4038−4044. (21) Wang, X. D.; Zhou, T. Y.; Chen, X.; Wong, K. Y.; Wang, X. R. Sens. Actuators, B: Chem. 2008, 129, 866−873. (22) Olry, A.; Schneider-Belhaddad, F.; Heintz, D.; Werck-Reichhart, D. Plant J. 2007, 51, 331−340. (23) Yun, C. H.; Yim, S. K.; Kim, D. H.; Ahn, T. Curr. Drug Metab. 2006, 7, 411−429. (24) Iwata, H.; Fujita, K. I.; Kushida, H.; Suzuki, A.; Konno, Y.; Nakamura, K.; Fujino, A.; Kamataki, T. Biochem. Pharmacol. 1998, 55, 1315−1325. (25) Goto, T.; Moriuchi, H.; Fu, X.; Ikegawa, T.; Matsubara, T.; Chang, G.; Uno, T.; Morigaki, K.; Isshiki, K.; Imaishi, H. Drug Metab. Dispos. 2010, 38, 2110−2116. 5297

dx.doi.org/10.1021/ac300355w | Anal. Chem. 2012, 84, 5292−5297