Vertically Integrated Human P450 and Oxygen Sensing Film for the

ARTICLE pubs.acs.org/ac

Vertically Integrated Human P450 and Oxygen Sensing Film for the Assays of P450 Metabolic Activities Gang Chang,†,§ Kenichi Morigaki,*,†,‡ Yoshiro Tatsu,† Takashi Hikawa,‡ Tatsushi Goto,‡ and Hiromasa Imaishi*,‡ †

National Institute of Advanced Industrial, Science and Technology (AIST), Midorigaoka, Ikeda 563-8577, 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 ‡

bS Supporting Information ABSTRACT: An assaying method of cytochrome P450 (P450 or CYP) monooxygenase activities for toxicological evaluation of drugs and environmental pollutants was developed by immobilizing P450 on an oxygen sensoring layer. Membrane fractions from E. coli expressing human P450 were entrapped in agarose or silica-based gels and immobilized on 96-well microarrays having an oxygen sensing film at the bottom. 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). P450 activity toward the substrates was monitored through the fluorescence intensity enhancement due to the oxygen consumption by the metabolic reactions. For the metabolism of chlortoluron, a selective herbicide used to control grass weeds, CYP1A1 immobilized in agarose gel showed a higher activity and stability compared with those in silica gels and free suspensions. The luminescence changing rate evaluated by the dynamic transient method (DTM) could be correlated with the substrate concentration. We also compared the metabolic responses of human P450s (CYP1A1,CYP2C8, CYP2E1, CYP3A4) toward various substances. The use of immobilized P450 on an oxygen sensing layer provides a versatile assaying platform owing to the following features. First, the oxygen sensor can detect metabolic reactions of any P450 species, in contrast with assays using fluorogenic substrates. Second, vertical integration of the oxygen sensor and immobilized P450 enhanced the sensitivity because of the effective depletion of oxygen in the vicinity of the oxygen sensing layer. Third, immobilization enables repeated use of P450 by replacing the substrate solutions using a flow cell. Furthermore, the activity of immobilized P450 was retained at least for 3 weeks at 4 °C, suggesting its long-term stability, which is an additional attractive feature.

C

ytochrome P450 monooxygenases consisting of cytochrome P450 (P450 or CYP) species and a generic NADPH-cytochrome P450 oxidoreductase (P450 reductase) play an important role in the biosynthesis of a variety of secondary metabolites, as well as in the metabolism of xenobiotics in organisms. Estimates from the genome projects have implied that there are 57 P450 genes in a human.1,2 Human P450s existing in the liver play an important role for the metabolic reactions including hydroxylation, O-dealkylation, and N-oxidation. The substrates include drugs and toxic compounds as well as metabolic products such as bilirubin (a breakdown product of hemoglobin).3,4 Therefore, metabolic activities of human P450s are important in the screening of drug candidates and the assessment of toxicity of chemicals in food and the environment.5
8 Assay techniques using fluorogenic substrates and microwell plates have been widely used due to their advantages over electrochemical biosensors,9
12 such as high throughput, small size, and no requirement for the reference electrode. However, there are some technological challenges. r 2011 American Chemical Society

One issue is detecting the activities of diverse P450 species using fluorogenic substrates. Each fluorogenic substrate can be used only for a limited number of P450 species that can metabolize it. No generic method appears to identify the activities of different P450 species and drug metabolism in vitro studies. Furthermore, fluorogenic substrates can only be used for competition assays, which cannot distinguish between metabolic activity and inhibition. Another important issue is to immobilize P450s in an appropriate matrix to provide a suitable environment for retaining their enzymatic activities.13
17 Immobilization is advantageous over the use of suspensions of membrane fractions because of the ease in storage and structural control and the potential of miniaturization. This should save both the amount of P450s used and the time for assaying. Received: November 21, 2010 Accepted: March 9, 2011 Published: March 24, 2011 2956

dx.doi.org/10.1021/ac103059k | Anal. Chem. 2011, 83, 2956–2963

Analytical Chemistry Detecting P450 activities with an oxygen sensing film should provide a pathway to solve the problem of fluorogenic substrate, because all metabolic reactions catalyzed by P450 involve oxygen consumption given as the following reaction equation. R
H þ O2 þ NADPH þ Hþ f R
OH þ H2 O þ NADPþ The activities of P450 for different substrates can be evaluated through monitoring oxygen consumption in the metabolic reaction. Oxygen sensing films, based on Ru and Pt complexes, whose luminescence is quenched by oxygen, can be used for monitoring oxygen consumption, following the Sterm
Volmer equation and evaluating the dissolved oxygen concentration. The use of optical biosensors for monitoring the oxygen consumption by enzymes and microorganisms has been studied extensively.18
25 There have been some studies that detected P450 activities by monitoring oxygen consumption in multiwell plates using suspensions of membrane fractions.26 However, existing approaches using the oxygen sensing layer have suffered from low sensitivity, partially due to the fact that they worked with suspensions of membrane fractions. For the immobilization of P450s, various methods have been developed such as sol
gel immobilization,13,17 lipid membrane containment,27 and emulsion entrapment.28,29 Sol
gel methods have been most widely studied for entrapping enzymes for its facile procedures and accurate control of the microstructures. However, conventional sol
gel methods using silica-based matrixes such as tetramethyl orthosilicate (TMOS) or tetraethyl orthosilicate (TEOS) were of limited use for immobilizing P450s because of the negative effects of alcohol generated from their hydrolysis reactions.17 Kato et al. used aqueous silica (Ludox solution) to prepare hydrogels for encapsulating P450s and combined it with the commercial microarray plate for monitoring the metabolic reactions of P450s toward drugs.13 The activity of P450s could be preserved for 3 weeks at 4 °C. On the other hand, agarose gel provides an attractive alternative to silica-based matrixes for immobilizing P450s, because it is a biomaterial widely used in electrophoresis, enzymatic biosensors, and cellbased biosensors.30
32 In agarose gel, linear polysaccharide chains are combined by hydrogen bonds, forming a pentagonal porous structure that offers a mild environment for enzymes to keep their activities and allows diffusion of the substrates to the enzymes. The diffusion rate of the chemical reagent in agarose gel is reported to be as fast as that in aqueous solution.31 The size of micropores in agarose gel can be facilely tuned by controlling the concentration of the precursor, which makes it an ideal candidate for immobilizing P450-containing membrane fractions. In the present study, a versatile format of P450 array was developed for rapid and reliable evaluation of its metabolic activities toward diverse substrates without fluorogenic substrates. Human P450 and P450 reductase (expressed in E. coli inner membranes) were immobilized on organically modified silica film (ORMOSIL) oxygen sensors to assay their metabolic activities. As the matrix to immobilize P450-containing membrane fractions, agarose gel, and two types of silica-based gels were compared. The oxygen sensing film based on luminescence quenching of tris(4,7-diphenyl-1,10-phenanthroline) ruthenium (Ru(dpp)32þ) can monitor the oxygen consumption in the metabolic reactions catalyzed by P450. Chlortoluron, a selective herbicide used to control grass weeds, and several other substances found in food were used as model substrates.

ARTICLE

The relationship between luminescence changing rate and concentration of substrate was discussed by applying the dynamic transient method (DTM). The long-term stability of P450 in agarose gel was also investigated by keeping the microarray for 3 weeks at 4 °C.

’ MATERIALS AND METHODS Materials. Tetraethyl orthosilicate (TEOS), triethoxy (octyl) silane (Octyl-triEOS), Ludox HS-40 colloidal silica, agarose (Type VII), and sodium silicate solution were purchased from SigmaAldrich. Tris(4,7-diphenyl-1,10-phenanthroline) ruthenium dichloride (Ru(dpp)3Cl2), isopropyl-1-thio-β-D-galactopyranoside, δ-amino levulinic acid, ethanol, methanol, and concentrated hydrochloric acid were obtained from Wako Pure Chemical Industries (Osaka, Japan). Potassium dihydrogenphosphate, coumarin, β-nicotinamide adenine dinucleotide phosphate tetrasodium salt (NADPH), and dipotassium hydrogenphosphate were purchased from Nacalai Tesque (Kyoto, Japan). Chlortoluron was obtained from Riedel-de Haen. Glucose-6phosphate (G6P), safrole, estragole, methoxypsoralen (5-MOP), and 8-methoxyproralen (8-MOP) were purchased from Tokyo Chemical Industry (Tokyo, Japan). Glucose-6-phosphate dehydrogenase (G6PD) was purchased from Toyobo (Osaka, Japan). All chemicals and solvents were reagent grade and used without further purification. Ninety-six microwell plates with a round-bottom were purchased from Nunc. Milli-Q water with the resistivity more than 18 MΩ 3 cm was used to prepare aqueous solutions. Instrumentation. All the luminescence measurements were performed on microplate reader Fluoroskan Ascent CF (Thermo Scientific) controlled by the software Ascent version 2.4 with the excitation and emission wavelength of 400 and 620 nm, respectively. The measurements were conducted from the top of wells (top mode) due to the transparency of agarose gel. Preparation of Membrane Fraction of Human P450 and P450 Reductase from E.coli. Vectors for expressing four species of P450s, CYP1A1, CYP2C8, CYP2E1, and CYP3A4, in E. coli were constructed. The cDNA fragments encoding human CYP1A1, CYP2C8, CYP2E1, and CYP3A4 were obtained by human liver-cDNA libraries using PCR. In order to modify N-terminal regions of cDNA fragments of CYP1A1 and CYP2E1 and add digestion site by NdeI and SalI, primers were prepared as follows: for CYP1A1, h1A1-F, 50 -GGAATTCCATATGGCTTTTCCAATTTCAATGTCAGCAACG-30 , and h1A1-R, 50 -AAGTCGACCTAAGAGCGCAGCTGCATTTGGAAGTGCT-30 ; for CYP2E1, h2E1-F, 50 -GGAATTCCATATGGCAAGACAAGTACATAGTAGTTGGATCTGCCCCCCGGGCCTTTCC-30 , and h2E1-R, 50 -AAGTCGACTCATGAGCGGGGAATGACACA-30 . The cDNA fragment of CYP2C8 with deletion of N-terminus was amplified by the following primers, h2C8-intF, 50 -AAGGAAGCTCCCTCCCGGGCCCACTCCTC-30 , and h2C8-R, 50 -AAGTCGACTCAGACAGGGATGAAGCAGATC30 . The full length cDNA fragment of CYP3A4 was isolated from liver cDNA library by 2-step PCR: hCYP3A4-F, 50 -AACCATGGCTCTCATCCCAGACTTGGCTATGGAA-30 , and hCYP3A4-R-Sal, 50 -TTGTCGACTCAGGCTCCACTTACGGTGC-30 , were used in first step and hCYP3A4-F-NdeOmpA primer, 50 -TTCATATGAAAAAGACAGCTATCGCGATTGCAGTGGCACTGGCTGGTTTCGCTACCGTAGCGCAGGCCATGGCTCTCATCCCAGACTT-30 , and the hCYP3A4-R-Sal primer were used in second step. This cDNA amplification resulted in addition of the bacterial signal peptide of OmpA. All amplified fragments were cloned into pT7 and sequenced. The plasmid 2957

dx.doi.org/10.1021/ac103059k |Anal. Chem. 2011, 83, 2956–2963

Analytical Chemistry

ARTICLE

Figure 1. Schematic illustration of the vertically integrated microarrays containing P450 in agarose gel and oxygen sensing film.

containing the cDNAs of CYP1A1, CYP2E1, and CYP3A4 were digested with NdeI and SalI, and the fragments obtained were subcloned into the SalI and NdeI sites of the pCW vector, which contains a cDNA of human NADPH-cytochrome P450 reductase.33 The cDNA fragment of CYP2C8 with N-terminal deletion was inserted between SmaI and SalI sites to previously modified pCW vector, resulting in addition of 11 amino acid residues, MAKKTSSKGKL.34 E. coli JM109 cells were transformed with a pCW1A1, pCW2C8, pCW2E1, and pCW3A4 vectors. The transformed cells were cultured in 5 mL of Luria broth (LB) at 37 °C for 1 day, and then, 3 mL of this culture was added to 300 mL of LB and incubated at 37 °C to an OD600 of 0.2 to 0.3. All liquid cultures were shook in a shaking incubator at 150 rpm. After the addition of isopropyl-1-thio-β-D-galactopyranoside (1 mM), and δ-amino levulinic acid (0.5 mM), the cultures were grown at 25 °C for 24 h. The cells were centrifuged and suspended in 4 mL of 100 mM potassium phosphate buffer (pH 7.5) containing 20% glycerol. E.coli cells were homogenized by sonication. The homogenate was centrifuged at 1000g for 15 min to obtain a supernatant fraction, and then, this supernatant was centrifuged at 100 000g for 60 min. Pellets (1 mL) were then resuspended in 10 mL of 100 mM potassium phosphate buffer (pH 7.5) containing 20% glycerol and then stored at
80 °C until use. Preparation of Oxygen Sensing Film on Microplate. The fluorophore doped solution was prepared as described in the ref 35 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 1 h under ambient condition, and then, 1.725 mL of ethanol was added into the solution for dilution of the solution to improve the quality of the oxygen sensing film. The solution was further stirred for 1 h. For preparing Ru(dpp)3Cl2 doped solution, 100 μL of 2 mM Ru(dpp)3Cl2 in ethanol was mixed with 300 μL of the above-mentioned solution. This solution was capped and stirred for 30 min, and 10 μL 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.24,36 The microplate was stored in the dark under ambient condition for gelling and aging for 6 days. For increasing hydrophilicity of the microarrays and attachment with hydrogel, poly (vinyl acetate) (PVAC) was used to modify the surface of the microarrays. Encapsulation of P450-Containing Membrane Fractions in the Agarose Gel, TEOS Gel, and Ludox Gel. Agarose was dissolved in deionized water to form 1.3% (w/w) solution at 60 °C. This solution was cooled down to ca. 38 °C. P450 suspensions (100 μL, 1.43 nmol/mL) were mixed with 300 μL of 1.3% agarose solution, and then, 60 μL of P450 doped agarose solution was pipetted onto the surface of the oxygen sensing layer in each well of the microplate. The thickness of P450 doped agarose gel in the microwell was ca. 2 mm. The microarrays were

kept in a refrigerator at 4 °C until use. A schematic illustration of the P450 encapsulated agarose gel in the oxygen sensing microarrays is shown in Figure 1. TEOS solution was prepared by mixing 0.5 mL of TEOS, 0.25 mL of deionized water, and 12.5 μL of 0.1 M HCl and stirring for 3 h to form a homogeneous solution. The solution was diluted four times with deionized water. Diluted TEOS solution (300 μL) was mixed with 100 μL of a P450-containing membrane fraction suspension, and 60 μL of the mixed solution was pipetted onto the surface of the oxygen sensing layer in each well of the microplate. The microplate was also kept in the refrigerator at 4 °C. Ludox solution was prepared as described in the literature.13,37 Briefly, 0.5 mL of 8.5 M Ludox colloidal silica was mixed with 0.5 mL of 0.16 M sodium silicate solution while stirring. HCl (4.0 M) was added to neutralize the pH value to around 7, and then, 100 μL of P450-containing membrane fraction suspensions was mixed with 300 μL of the above Ludox silica solution. P450 doped solution (60 μL) was dropped to each well of the microplate. The microplate was stored in the refrigerator at 4 °C for 2 days for aging before use. P450 Assays in Microarrays. For the measurement of P450 activities toward the substrates, substrate solutions were prepared as follows. In the case of chlortoluron, 25 μL of methanol solutions with different concentrations of chlortoluron (0.8, 4, 8, 20, 40 mM) were added to 1975 μL of 0.1 mM KPB solution containing a NADPH regenerating system, 0.1 mM NADPH, 3 mM MgCl2, 3 mM G6P, and 0.4 U/mL G6PD. The final concentrations of the chlortoluron were 0.01, 0.05, 0.1, 0.25, and 0.5 mM, respectively. A 250 μL portion of the standard solution was added to each well of the microplates. A transparent polymer tape was used to seal the plate and prevent the oxygen penetration into the well. After the addition of the substrate solution into the microarrays, the microplate was quickly placed onto the microplate reader for the fluorescence measurement. Fluorescence intensity was recorded every 5 min for 3 h. The same procedure was applied also for other substrates. HPLC Analysis. Metabolic activities of CYP1A1 were analyzed by an HPLC system (HITACHI High-Technologies Co., Tokyo, Japan) equipped with a C18 column, ODS-TSK gel (150  4.6 mm; Tosoh Co., Tokyo, Japan). The mobility phase was changed with linear gradient from 37.5% to 100% of methanol for 15 min, and metabolites were monitored at 324 nm. The metabolic products were quantified by integrating their peaks.

’ RESULTS AND DISCUSSION Oxygen Sensor Responses toward the Metabolic Activities of CYP1A1 Encapsulated in Agarose, Ludox Silica, and TEOS Silica. We immobilized CYP1A1 on the oxygen sensing

layer using three types of gel matrixes, agarose, Ludox silica, and TEOS silica. Figure 2 shows the changes in luminescence of the 2958

dx.doi.org/10.1021/ac103059k |Anal. Chem. 2011, 83, 2956–2963

Analytical Chemistry

ARTICLE

Figure 2. Fluorescence responses of oxygen sensors caused by the metabolic reaction of CYP1A1 encapsulated in different gels. The substrate was chlortoluron (0.5 mM). (A) Agarose gel, (B) the changing rate of fluorescence derived from (A), (C) Ludox silica gel, and (D) TEOS silica gel.

oxygen sensing layer after the addition of a substrate solution (0.5 mM chlortoluron) to each well. As control experiments, NADPH solution without chlortoluron was also added to other wells. In the case of CYP1A1 in agarose gels, a distinct fluorescence intensity increase was observed in the presence of chlortoluron (Figure 2A red). A small increase of fluorescence intensity was observed also without chlortoluron (black), which is attributed to the side reaction from NADPH oxidation by CYP1A1.26 The larger fluorescence increase in the presence of the substrate suggests that CYP1A1 encapsulated in agarose gel retains the enzymatic activity as in aqueous solution.26 Interestingly, the kinetic courses of the fluorescence intensity change fit well to the following sigmoid equation (fitting coefficient: 0.99). It ¼ I1 þ ðI0
I1 Þ=ð1 þ eðt
tmax Þdt Þ It is the fluorescence intensity at the time t. I0 and I1 represents the fluorescence intensities at the beginning and after reaching the saturation, respectively. tmax represents the time point where the maximum rate of fluorescence intensity increase was observed (the time point of inflection). dt is the time range for the most significant fluorescence changes, which is used solely for the fitting purpose. We used this equation to fit the data points and derive the rate of fluorescence changes. The rate of fluorescence changes should reflect the reaction rate and has been widely used to estimate the value of biosensor, significantly reducing the

measurement time.25,38
40 Therefore, we also employed the luminescence changing rate for the evaluation of kinetic behaviors (Figure 2B; the analytical technique is generally called the dynamic transient method (DTM)).25,38,40 Although DTM is an empirical method, it can fit the complex kinetic responses of optical biosensors including substrate penetration into the gel, oxygen consumption by the enzymatic reaction, and oxygen diffusion/depletion in the gel. The fluorescence changing rate increased in the first hour due to the consumption of oxygen resulting from substrate diffusion into the gel and P450 catalysis. Subsequently, the rate decreased due to the exhaustion of substrate or oxygen in the well. Similar behaviors have been observed in microbial biochemical oxygen demand biosensors (BOD).38,41
44 We used the maximum changing rate for evaluating the P450 activities. To investigate the effects of the matrixes on the P450 activities, Ludox gel and TEOS gel were also used for immobilizing CYP1A1 on the oxygen sensing layer. Figure 2C,D shows the fluorescence increase of oxygen sensing layer due to the metabolic activities of P450 encapsulated Ludox gel and TEOS gel in the presence and absence of the substrate (0.5 mM chlortoluron), respectively. Higher background oxygen consumption from NADPH was observed in Ludox gel, compared with the results of CYP1A1 encapsulated in agarose gel (Figure 2C black dots-line). The fluorescence showed only a limited increase in the presence of chlortoluron. The small mesoporous structures in Ludox silica may have restricted the 2959

dx.doi.org/10.1021/ac103059k |Anal. Chem. 2011, 83, 2956–2963

Analytical Chemistry

ARTICLE

Figure 3. Comparison of the oxygen sensor responses toward metabolic reactions of CYP1A1 dispersed in suspension and encapsulated in agarose gel.

diffusion of substrate molecules. The mean pore sizes of agarose gel and Ludox silica gel were reported to be 400 and 10 nm, respectively.30,32,45 CYP1A1 in TEOS silica gel showed lower background oxygen consumption in the absence of chlortoluron (Figure 2D black dots-line). However, the fluorescence increase was not significantly enhanced in the presence of the substrate, which should be due to the denaturation effect of alcohol produced during the hydrolysis of TEOS.17 Immobilization of P450 has two distinctive advantages compared with the assays using suspensions of membrane fractions. First, vertical integration of immobilized P450 and oxygen sensor can enhance the sensitivity of oxygen sensor toward the oxygen consumption by P450. Figure 3 shows the comparison of the responses toward chlortoluron metabolism by P450 dispersed in suspensions and immobilized in agarose gel, respectively. Compared with P450 in suspensions, P450 in agarose gel showed much higher responses (10 times), indicating that it is more sensitive toward P450 metabolism. The different responses should stem from differences in the oxygen consumption and permeation in the gel and in the solution. In the case of immobilized P450, oxygen consumption occurs in the vicinity of the oxygen sensor and the gel matrixes should impede the permeation of oxygen molecules, causing oxygen depression in the gel. In the case of P450 dispersed in suspension, oxygen consumption occurs uniformly in the well and oxygen molecules can diffuse freely, restricting the formation of the concentration gradient. The differences in permeability apply also to substrate molecules. Therefore, a quick fluorescence change was observed in dispersed P450, while initial responses for immobilized P450 were rather slow. The second advantage is the fact that immobilization of P450 in agarose gel enables one to assay the P450 activities repeatedly using a flow cell. We observed that P450 enzymes were active in the second and third assays using the same wells. Furthermore, immobilized P450 was found to be stable for a prolonged period. Immobilized P450 were kept at 4 °C for 21 days, and their activity was evaluated. CYP1A1 in agarose gel showed a similar catalytic behavior even after 3 weeks (results not shown). This result confirms that P450 activities can be preserved well by

Figure 4. (A) Fluorescence responses of CYP1A1-oxygen sensors toward different concentrations of chlortoluron. (B) Correlation between the maximum luminescence rate and the concentration of chlortoluron. The red line is the linear fitting result.

encapsulating membrane fractions in agarose gel, suggesting its utility for long-term applications. Responses of P450-Oxygen Sensors toward Different Substrate Concentrations and Types. The results described above have shown that agarose was the most suitable matrix for encapsulating P450. We, therefore, employed the agarose-based matrix to construct integrated sensors (we call this configuration P450-oxygen sensor hereafter). We evaluated their responses toward different substrate types and concentrations. Figure 4A shows the fluorescence changes for different chlortoluron concentrations. CYP1A1-oxygen sensor exhibited a larger luminescence increase for higher substrate concentrations. The fluorescence increase was differentiated with time, again fitting well with sigmoidal curves (with the high coefficient of 0.99). Figure 4B summarizes the maximum luminescence changing rate as a function of the chlortoluron concentration. The error bar represents the standard deviation. The substrate concentration showed a high linear correlation with the maxim rate (R = 0.997) in the concentration range studied (i.e., 0.01
0.5 mM), although the response without chlortoluron (0 mM) slightly deviated from the linearity (Figure 4B). The fluorescence intensity 2960

dx.doi.org/10.1021/ac103059k |Anal. Chem. 2011, 83, 2956–2963

Analytical Chemistry

ARTICLE

Figure 5. Fluorescence responses of P450-oxygen sensors with four P450s (CYP1A1, CYP2C8, CYP2E1, CYP3A4) toward various substances.

changes are caused as a total effect of complex procedures, including substrate permeation, oxygen consumption by the P450 catalysis, and oxygen permeation. The linear dependency of responses toward the substrate concentrations suggest that the rate of one of these processes, presumably oxygen consumption by the P450 catalysis considering its relatively low rate, is limited by the substrate concentration. One important advantage of the assay technique monitoring oxygen consumption is the fact that it directly detects the metabolic reactions of P450 (unlike conventional assays using fluorogenic substrates, which can only evaluate the competition). Furthermore, it can be applied to monitor the activities of diverse P450 species. In order to demonstrate this potential, P450-oxygen sensors with four human P450s (CYP1A1, CYP2C8, CYP2E1, and CYP3A4) were applied to evaluate their activities toward seven natural organic compounds found in food (estragole, safrole, capsaicin, 5-methoxypsoralen (5-MOP), 8-methoxypsoralen (8-MOP), and coumarin) and a pesticide (chlortoluron). These natural organic compounds widely exist in herbal medicines and food additives and are suspected to undergo metabolic activation, which might result in reactive intermediates that covalently bind with DNA and protein and finally cause the genotoxicity and cytotoxcity.46 Thus, it is important to evaluate the metabolic activities of various human P450s toward these chemicals. Figure 5 shows the fluorescence changes, and Figure 6 shows the maximum changing rates of fluorescence intensity derived using the DTM method. The values of maximum changing rate are compiled as normalized values to the

background responses (no substrate, designated as NADPH). The fluorescence changing rates depended critically on the combination of substrates and P450 species. For example, in the case of CYP1A1, chlortoluron and capsaicin had significantly higher responses compared with the background. Especially, capsaicin showed the highest fluorescence increase, indicating a high activity of CYP1A1, which is consistent with previous reports.47 On the other hand, the values for safrole, 5-MOP, and 8-MOP were similar or even lower compared with the background. Lower values found in the metabolism of 5-MOP and 8-MOP may reflect their inhibiting properties to P450 catalysis.48 In order to ascertain that the fluorescence enhancement of oxygen sensor reflect the metabolic activities of P450, we measured the metabolic reactions of CYP1A1 toward capsaicin, estragole, and safrole using HPLC. Although the amounts of products could not be determined because of different extinction coefficients, significantly larger number and quantity of metabolites were observed for capsaicin and estragole compared with safrole, in agreement with the oxygen sensor responses. Furthermore, the effects of coexisting enzymes in E. coli inner membranes were evaluated by measuring the response of the membrane fraction without P450 (see the supplementary data). E. coli membranes without P450 did not have any enhanced responses in the presence of capsaicin, confirming that the fluorescence enhancement after the substrate addition was due to the metabolic reaction of P450. As discussed above, monitoring oxygen consumption is advantageous over the use of fluorogenic substrates in that it can 2961

dx.doi.org/10.1021/ac103059k |Anal. Chem. 2011, 83, 2956–2963

Analytical Chemistry

ARTICLE

Figure 6. Maximum changing rates of fluorescence intensity in Figure 5 are compiled for each P450. (1) NADPH (background responses without substrates); (2) chlortoluron; (3) capsaicin; (4) safrole; (5) estragole; (6) coumarin; (7) 5-MOP; (8) 8-MOP.

detect the metabolic activities. The relative fluorescence intensity changes should provide a useful insight into the activities of various P450 species toward diverse chemical substances. One technological limitation at the moment is the background oxygen consumption that is caused by either the uncoupling reactions of P450 or enzymatic activities caused by intrinsic substances (e.g., fatty acids) present in the membrane fragments. The responses for different P450 species and substrates may be affected by different levels of background oxygen consumption. Therefore, we should state at the moment that one should be careful in the use of the quantitative values obtained. However, P450-oxygen sensor arrays should provide useful information on the metabolic activities of various types of P450s toward diverse chemicals, as supported by the correlation with the HPLC results. 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 (Supporting Information).

’ CONCLUSIONS A new format of P450 microarray was developed by combining an oxygen-sensing film with immobilized P450s in agarose gel. The two layers were vertically integrated so that the oxygen consumption by P450s in the agarose gel layer could be effectively detected by the oxygen-sensing layer beneath. In this configuration, a heightened response of the oxygen sensor

was detected compared with P450s in membrane fractions suspensions (10 times higher response). Agarose was found to be a superior gel matrix compared with silica-based gels due to higher P450 enzymatic activities and lower background oxygen consumption. CYP1A1-oxygen sensor showed varied responses depending on the type and concentration of the chemical substances in the aqueous solution. For chlortoluron, a linear concentration dependency of the response was found in a certain concentration range, suggesting that the responses can be correlated with the oxygen consumption by the P450 metabolism. An important advantage of P450 assays based on oxygen sensing is the fact that it detects the enzymatic reactions, whereas fluorogenic substrates can only be used for competitive assays. We demonstrated this advantage by evaluating the activities of human P450s toward several chemicals found in food. Although the absolute values of P450 activities are not obtained at present due to the background oxygen consumption, there is a clear correlation between the metabolic activities and the oxygen sensor responses. The use of optical detection also enables one to assay a large number of P450 species in a parallel fashion. The activity of CYP1A1 was retained at least for 3 weeks at 4 °C. Owing to these advantages, vertically integrated P450 and oxygen sensors seems to be a feasible alternative to the assays based on fluorogenic substrates or HPLC, which are currently widely used for monitoring the activities of human P450s. Combining with microfabricated chips and microfluidics, the P450-oxygen sensors should provide a new tool for high 2962

dx.doi.org/10.1021/ac103059k |Anal. Chem. 2011, 83, 2956–2963

Analytical Chemistry throughput P450 metabolism assays needed in drug development and environmental monitoring.

’ ASSOCIATED CONTENT

bS

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

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected] (K.M.), himaish@kobe-u. ac.jp (H.I.).

’ ACKNOWLEDGMENT This work was financially supported by Program for Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN). We thank Ms. Saori Mori, Mr. Kazuyuki Mizutani, and Ms. Emi Kanemura (AIST) for their assistance in the experiments. We thank Dr. Masanori Horie (Health Technology Research Center, AIST) for allowing us to use the Fluoroskan Ascent CF multiwell plate reader. Discussions with Dr. Kunio Isshiki and Dr. Akira Arisawa (Mercian Corp.) were also 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) Korzekwa, K. R.; Jones, J. P. Pharmacogenetics 1993, 3, 1–18. (4) Anzenbacher, P.; Anzenbacherova, E. Cell. Mol. Life Sci. 2001, 58, 737–747. (5) Bistolas, N.; Wollenberger, U.; Jung, C.; Scheller, F. W. Biosens. Bioelectron. 2005, 20, 2408–2423. (6) Ansede, J. H.; Thakker, D. R. J. Pharm. Sci. 2004, 93, 239–255. (7) Walsky, R. L.; Boldt, S. E. Curr. Drug Metab. 2008, 9, 928–939. (8) Tsotsou, G. E.; Cass, A. E. G.; Gilardi, G. Biosens. Bioelectron. 2002, 17, 119–131. (9) Lvov, Y. M.; Lu, Z. Q.; Schenkman, J. B.; Zu, X. L.; Rusling, J. F. J. Am. Chem. Soc. 1998, 120, 4073–4080. (10) Sultana, N.; Schenkman, J. B.; Rusling, J. F. Electroanalysis 2007, 19, 2499–2506. (11) Paternolli, C.; Antonini, M.; Ghisellini, P.; Nicolini, C. Langmuir 2004, 20, 11706–11712. (12) Sukumaran, S. M.; Potsaid, B.; Lee, M.-Y.; Clark, D. S.; Dordick, J. S. J. Biomol. Screening 2009, 14, 668–678. (13) Sakai-Kato, K.; Kato, M.; Homma, H.; Toyo’oka, T.; Utsunomiya-Tate, N. Anal. Chem. 2005, 77, 7080–7083. (14) Lee, M. Y.; Park, C. B.; Dordick, J. S.; Clark, D. S. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 983–987. (15) 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. (16) Dave, B. C.; Dunn, B.; Valentine, J. S.; Zink, J. I. Anal. Chem. 1994, 66, A1120–A1127. (17) Gill, I. Chem. Mater. 2001, 13, 3404–3421. (18) Preininger, C.; Klimant, I.; Wolfbeis, O. S. Anal. Chem. 1994, 66, 1841–1846. (19) Shi, Y. N.; Seliskar, C. J. Chem. Mater. 1997, 9, 821–829. (20) Borisov, S. M.; Wolfbeis, O. S. Chem. Rev. 2008, 108, 423–461. (21) Li, X. M.; Ruan, F. C.; Ng, W. Y.; Wong, K. Y. Sens. Actuators, B: Chem. 1994, 21, 143–149.

ARTICLE

(22) John, G. T.; Klimant, I.; Wittmann, C.; Heinzle, E. Biotechnol. Bioeng. 2003, 81, 829–836. (23) Kwok, N. Y.; Dong, S. J.; Lo, W. H.; Wong, K. Y. Sens. Actuators, B: Chem. Sens. Mater. 2005, 110, 289–298. (24) 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. (25) Wang, X. D.; Zhou, T. Y.; Chen, X.; Wong, K. Y.; Wang, X. R. Sens. Actuators, B: Chem. 2008, 129, 866–873. (26) Olry, A.; Schneider-Belhaddad, F.; Heintz, D.; WerckReichhart, D. Plant J. 2007, 51, 331–340. (27) Ueda, Y.; Morigaki, K.; Tatsu, Y.; Yumoto, N.; Imaishi, H. Biochem. Biophys. Res. Commun. 2007, 355, 926–931. (28) Ryan, J. D.; Clark, D. S. Biotechnol. Bioeng. 2008, 99, 1311–1319. (29) Lamb, S. B.; Lamb, D. C.; Kelly, S. L.; Stuckey, D. C. FEBS Lett. 1998, 431, 343–346. (30) Griess, G. A.; Guiseley, K. B.; Serwer, P. Biophys. J. 1993, 65, 138–148. (31) Wolfaardt, G. M.; Lawrence, J. R.; Hendry, M. J.; Robarts, R. D.; Caldwell, D. E. Appl. Environ. Microbiol. 1993, 59, 2388–2396. (32) Pernodet, N.; Maaloum, M.; Tinland, B. Electrophoresis 1997, 18, 55–58. (33) Uno, T.; Nakao, A.; Masuda, S.; Taniguchi, Y.; Kanamaru, K.; Yamagata, H.; Nakamura, M.; Imaishi, H.; Oono, K. J. Ind. Microbiol. Biotechnol. 2006, 33, 1043–1050. (34) 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. (35) Tang, Y.; Tehan, E. C.; Tao, Z. Y.; Bright, F. V. Anal. Chem. 2003, 75, 2407–2413. (36) 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. (37) Nassif, N.; Roux, C.; Coradin, T.; Rager, M. N.; Bouvet, O. M. M.; Livage, J. J. Mater. Chem. 2003, 13, 203–208. (38) Xiao, D.; Choi, M. M. F. Anal. Chem. 2002, 74, 863–870. (39) Jiang, Y. Q.; Xiao, L. L.; Zhao, L.; Chen, X.; Wang, X. R.; Wong, K. Y. Talanta 2006, 70, 97–103. (40) Yang, X. F.; Zhou, Z. D.; Xiao, D.; Choi, M. M. F. Biosens. Bioelectron. 2006, 21, 1613–1620. (41) Yang, Z.; Suzuki, H.; Sasaki, S.; McNiven, S.; Karube, I. Sens. Actuators, B: Chem. 1997, 45, 217–222. (42) Jiang, Y. Q.; Xiao, L. L.; Zhao, L.; Chen, X.; Wang, X. R.; Wong, K. Y. China/Japan/Korea Environmental Analytical Chemistry Symposium; Elsevier Science Bv: Chiba, JAPAN, Sep 01
03, 2005; pp 97
103. (43) Lin, L.; Xiao, L. L.; Huang, S.; Zhao, L.; Cui, J. S.; Wang, X. H.; Chen, X. Biosens. Bioelectron. 2006, 21, 1703–1709. (44) Xin, L. L.; Wang, X. D.; Guo, G. M.; Wang, X. R.; Chen, X. Meas. Sci. Technol. 2007, 18, 2878–2884. (45) Coiffier, A.; Coradin, T.; Roux, C.; Bouvet, O. M. M.; Livage, J. J. Mater. Chem. 2001, 11, 2039–2044. (46) Zhou, S. F.; Xue, C. C.; Yu, X. Q.; Wang, G. Curr. Drug Metab. 2007, 8, 526–553. (47) Reilly, C. A.; Ehlhardt, W. J.; Jackson, D. A.; Kulanthaivel, P.; Mutlib, A. E.; Espina, R. J.; Moody, D. E.; Crouch, D. J.; Yost, G. S. Chem. Res. Toxicol. 2003, 16, 336–349. (48) Tassaneeyakul, W.; Birkett, D. J.; Veronese, M. E.; McManus, M. E.; Tukey, R. H.; Quattrochi, L. C.; Gelboin, H. V.; Miners, J. O. J. Pharm. Exp. Ther. 1993, 265, 401–407.

2963

dx.doi.org/10.1021/ac103059k |Anal. Chem. 2011, 83, 2956–2963