- Email: [email protected]
Expression of Cytochrome P450 2A6 in Escherichia coli: Purification, Spectral and Catalytic Characterization, and Preparation of Polyclonal Antibodies
Archives of Biochemistry and Biophysics Vol. 370, No. 2, October 15, pp. 190 –200, 1999 Article ID abbi.1999.1388, available online at http://www.idealibrary.com on
Expression of Cytochrome P450 2A6 in Escherichia coli: Purification, Spectral and Catalytic Characterization, and Preparation of Polyclonal Antibodies 1 Pavel Soucˇek 2 Biotransformations Group, Center of Occupational Diseases, National Institute of Public Health, Sˇroba´rova 48, Praha 10, 100 42, Czech Republic
Received March 23, 1999, and in revised form July 6, 1999
Cytochrome P450 (CYP) 2A6 is the principal human enzyme catalyzing coumarin 7-hydroxylation and is known to be involved in the metabolism of halothane, nicotine, and metabolic activation of butadiene and nitrosamines. In this paper expression of CYP2A6 in Escherichia coli is reported. In order to achieve expression, the N-terminus of protein was modified by PCR mutagenesis. The N-terminal variant with only a single amino acid change showed expression of 210 nmol of CYP2A6/liter of culture. Recombinant CYP2A6 protein was purified to electrophoretic homogeneity and further characterized. Absolute spectra were typical for CYP proteins and indicated low spin characteristics of isolated protein. Due to a hydrophobic segment the N-terminal amino acid sequence of recombinant CYP2A6 was blocked. The N-terminal formylmethionine block was removed by mild acid treatment. Purified CYP2A6 had good catalytic activity toward marker substrate coumarin in a reconstituted system (K m 5 1.48 6 0.37 mM, V max 5 3.36 6 0.18 nmol product/min/nmol CYP). Its activity in the reconstituted system was stimulated by the presence of cytochrome b 5 and glutathione. CYP2A6 was shown to metabolize chlorzoxazone in the reconstituted system with activity of 0.32 nmol of product/min/nmol of CYP, and thus caution should be taken when interpretation of CYP2E1 in vivo phenotyping data is performed. Rabbit polyclonal antibodies were produced against recombinant CYP2A6 and proved to be very useful for immunoblotting and immunoinhibition studies. Availability of this expression system and specific antibodies should facilitate characterization of the role of
1 Supported by the Internal Grant Agency of Czech Ministry of Health, Grant IGA 3505-3 (1996-98), and the Grant Agency of Czech Republic, Grant 303/96/0373 (1996). 2 Fax: 1420-2-6731 1236. E-mail: [email protected].
190
CYP2A6 in the metabolism of chemicals and in the study biological relevance of genetic polymorphisms of this enzyme. © 1999 Academic Press Key Words: CYP2A6; heterologous expression; protein purification; catalytic activity; coumarin; antibodies.
Cytochrome P450 (CYP) 3 enzymes (EC 1.14.14.1) are the principal metabolizing enzymes involved in detoxication and/or activation of compounds of both natural and synthetic origin (2). Some CYP-catalyzed reactions result in the formation of cytotoxic or genotoxic intermediates (3). Heterologous expression systems including mammalian cells, yeasts, insect cells, and bacteria (for review, see 4) have been developed in order to increase the understanding of the role of individual CYP enzymes in the metabolism of xenobiotics and its consequences. Heterologous expression systems also serve as a tool to investigate the functional relevance of genetic polymorphisms of metabolizing enzymes. Moreover, expression of mutated CYPs may be the key approach in crystallization and in the elucidation of the secondary structure of these membrane proteins (5). Of the systems mentioned above, bacterial systems possess many advantages in terms of low cost, ease of use, and ammenability to scale up (6). Another advantage of highscale production of ultrapure protein in bacteria is the possibility to raise specific and possibly inhibitory antibodies which could serve as an important aid in studies of the metabolism of CYP substrates in subcellular fractions. N-terminally modified forms of many human 3 Abbreviation used: CYP, cytochrome P450 (for reference to the nomenclature, see 1).
0003-9861/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
EXPRESSION OF CYTOCHROME P450 2A6 IN BACTERIA
CYPs have been expressed in Escherichia coli (for review, see 3 and 4) and proved to be useful for catalytic activity assays. The CYP2A6 gene was cloned and sequenced in 1990 (7, 8). CYP2A6 protein was first purified from human liver and polyclonal antibodies has been produced against this protein (9). However, this preparation was contaminated by at least one CYP form and further purification was hindered by a limited supply of human liver. It was shown that relative content of CYP2A6 between individual human livers differs up to 260-fold and its relative content accounts for about 4% of total CYP content (2, 10, 11). Genetic polymorphism reflecting in three allelic variants (recently designated 2A6*1, 2A6*2, and 2A6*3; see 12) was identified (13). CYP2A6 was expressed in mammalian cells and yeasts (4, 14) and was shown to be responsible for the oxidation of coumarin (marker substrate), halothane, and nicotine and for the metabolic activation of 1,3-butadiene and nitrosamines (15–20). Recently, CYP2A6 was also expressed in E. coli with an attached bacterial signal peptide (21). Although the level of expression was high, this protein was not purified to homogeneity and the activity of the membrane fraction was quite low (21). No spectral or catalytic studies using purified recombinant CYP2A6 as well as the production of inhibitory polyclonal antibodies have been published so far. Here, the expression of a large amount of slightly modified (second amino acid Ala instead of Leu) CYP2A6, purification to electrophoretic homogeneity, and thorough spectral and catalytic study are reported. Moreover, inhibitory polyclonal antibodies against expressed CYP2A6 have been successfully produced. Purified CYP2A6 was used to elucidate its participation in the metabolism of the purported CYP2E1 probe substrate, chlorzoxazone. MATERIALS AND METHODS Chemicals. The expression vector pCW/NF14, HPLC standard for assay of oxidation of tolbutamide, and purified CYP standards for immunoblotting were generous gifts of Dr. F. Peter Guengerich (Vanderbilt University, Nashville, TN), and the CYP2A6 cDNA was provided by Dr. Frank J. Gonzalez (National Institute of Health, Bethesda, MD) cloned into the plasmid pUC/2A6. For bacterial transformation, DH5a E. coli competent cells (Life Technologies Ltd., Paisley, UK) were used. All chemicals used in expression, purification, and activity assays were products of Sigma (St. Louis, MO) except bufuralol and 19-hydroxy bufuralol (Gentest Corp., Woburn, MA). Restriction enzymes were products of BoehringerMannheim GmbH (Mannheim, Germany), and reagents for PCR were purchased from Perkin Elmer Corp. (Norwalk, CT). Purified human cytochrome b 5 was a generous gift of Dr. Hiroshi Yamazaki (Kanazawa University, Kanazawa, Japan). Human liver donors. Human liver samples were obtained from the Transplantation Center (IKEM, Czech Republic) from donors who died accidentally as a result of brain injury. Liver samples were obtained at most 30 min after death and were stored in liquid nitrogen until microsomes were isolated. Some of the donors were
191
submitted to therapy including antibiotics, mannitol, or hormones for 24 – 48 h before death. Microsomes were prepared according to van der Hoeven and Coon (22) and stored in aliquots of 200 ml at 280°C. Construction of expression plasmids. CYP2A6 was cloned between EcoRI (59-end) and XbaI (39-end) sites of pUC18 plasmid. N-terminal modifications of CYP2A6 clone were performed by PCR mutagenesis using primers introducing the NdeI site (CATATG) and the desired N-terminal change at the 59-terminus and the SalI site (GTCGAC) at the 39-terminus. The resulting DNA fragment was cleaved by NdeI and SalI restriction enzymes and resolved by horizontal electrophoresis in 1% agarose gel. An agarose band bearing a DNA fragment of a length corresponding to that of the CYP2A6 clone was extracted by GenecleanII (Bio 101, Vista, CA) and used for ligations into pCW torso, which was prepared by cleavage of expression plasmid pCW/NF14 (23) by NdeI and SalI and further processed in the same way as described above for the CYP2A6 fragment (see Fig. 1). Ligation of CYP2A6 fragment and pCW torso was achieved by the use of the DNA ligation kit according to producer’s protocol (Boehringer-Mannheim). Expression of plasmids in E. coli. Expression of N-terminal variants of pCW/2A6 was assessed following the basic procedure described previously (23). E. coli strain DH5a was transformed with each plasmid by heat shock and selected on Luria–Bertani agar plates containing ampicillin. Single colonies were then grown overnight in Luria–Bertani medium supplemented with ampicillin (100 mg/ml) at 37°C. An overnight culture (1 ml) was used to seed 100 ml of expression medium (23) where trace elements were added (24). Protein expression was induced by 1 mM isopropyl b-D-thiogalactoside for 40 h at 32°C with shaking at 200 RPM in a refrigerated orbital incubator (Sanyo-Gallenkamp PLC, Leicester, UK). Cells were harvested and subcellular fractions were prepared as described (23). Fe 21-CO vs Fe 21 difference spectra were used to detect hemoprotein (25) in whole cells. Large-scale cultures were routinely done by inoculating 0.5 liters of expression medium with 5 ml of overnight culture. Supplementation of expression medium with 0.5 mM d-aminolevulinic acid slightly enhanced expression levels. Subsequent preparation of the membrane fraction was carried out as described elsewhere (23, 26, 27). Purification of recombinant CYP2A6 from E. coli. The recombinant CYP2A6 purification schedule was chosen according to approaches from the literature (23, 26, 27). Basically, the procedure used was very similar to that previously reported for CYP2E1 purification (27). In contrast to some other CYP purification procedures (27), no ligand was used to protect CYP from degradation (formation of P420) during purification. The CYP2A6 membrane fraction was solubilized using high concentrations of detergents sodium cholate and Triton N-101 (both 0.625%). Supernatant of solubilized membranes (105,000g) was then passed through a weak anion-exchanger DEAE-Sephacel. After Triton removal by Amberlite XAD-2, this fraction was applied to weak cation exchanger CM-Sepharose. CYP2A6 formed a brown layer at the top of the column. The potassium phosphate gradient (10 –200 mM) was used for removal of contaminants and elution of CYP2A6. Elution of CYP2A6 started at about 80 mM phosphate. Detergent was removed by application of diluted CYP2A6 preparation to a small hydroxylapatite column and extensive washing with 10 mM potassium phosphate buffer (pH 7.7). CYP2A6 protein was eluted from hydroxylapatite by a high concentration of phosphate and 0.1% sodium cholate. Excess phosphate and cholate was removed by dialysis against 50 mM potassium phosphate (pH 7.4) containing 20% glycerol and 1 mM EDTA. Purity of CYP2A6 preparation during purification was monitored by electrophoresis. The CYP2A6 concentration in the top hydroxylapatiteeluate fraction was 16 mM with a CYP specific content of 12.35 nmol CYP/mg protein.
ˇ EK PAVEL SOUC
192
TABLE I
PCR Primers Used to Introduce N-Terminal Modifications and Amino Acid Sequences of Constructs Amino acid sequence b Nucleotide sequence a
Construct
1 3
12
18
21
30
Forward primers (forming a NdeI site and modifications at 59-end) Native A B C D E F
59-agctgacatcatagtcttacatatgctggcctcaggg-39 59-ttaggacatatggctgcttcaggaatgttattagtggccttgctggtc-39 59-ttaggacatatggctctgttattagcagtttttctggtctgcctgact-39 59-catatggctctgttattagcagtttttctgatgtctgtttggcag-39 59-catatggctgttggcaacaacgtaaaagcaaagggaagctg-39 59-catatggtattaatgagtgtttggcaacaaaggaagagcaag-39 59-gacgtgatggtcttgcatatggctcgtcaagttcattcttcttggaatgggaagctgcctccg-39
MLASGMLLVALLVCLTVMVLMSVWQQRKSKGKL MAASGMLLVALLVCLTVMVLMSVWQQRKSKGKL M___ALLLAVFLVCLTVMVLMSVWQQRKSKGKL M___ALLLAVFL________MSVWQQRKSKGKL M_A___________________VWQQRKSKGKL _________________MVLMSVWQQRKSKGKL M_ARQVHSSWN___________________GKL
Universal reverse primer (forming a SalI site at 39-end) 59-catatgattcagatacgtcgactcagcggggcaggaa-39 a b
Start codon is underlined. Amino acid modifications are in bold.
Purification of other proteins. Rat NADPH-P450 reductase was purified from liver microsomes of phenobarbital-treated rats by the method of Yasukochi and Masters (28) with modifications (29). Assays of catalytic activity. The standard CYP2A6 reconstituted system contained 0.5 ml of the following mixture: 25 pmol of CYP2A6, 75 pmol of rat NADPH-P450 reductase, 30 mM L-a-dilauroyl-sn-3-phosphocholine, 50 mM potassium phosphate buffer (pH 7.4), substrate (20 mM coumarin unless otherwise noted), and NADPH-generating system (10 mM MgCl 2, 10 mM glucose 6-phosphate, 1 mM NADP, 0.5 U/mL glucose-6-phosphate dehydrogenase). Coumarin 7-hydroxylation was assayed using the HPLC method with fluorescence detection as described (30). In experiments with human liver microsomes, 100 pmol of total CYP was used in a total volume of 0.5 ml and the following CYP marker substrates were included: 20 mM coumarin (CYP2A6), 3 mM 7-ethoxyresorufin (CYP1A1/2), 2.5 mM tolbutamide (CYP2C9), 0.4 mM bufuralol (CYP2D6), 0.5 mM chlorzoxazone (CYP2E1), and 0.1 mM testosterone (CYP3A4). Incubations have been carried out in a shaking water bath at 37°C for 15 min except for the 7-ethoxyresorufin, tolbutamide, and bufuralol assay, where 30 min incubation was used. Potassium phosphate buffer (50 mM, pH 7.4) and the above-mentioned NADPH-generating system were used. In experiments with anti-CYP2A6 antibodies, microsomes were preincubated with 1–5 mg of IgG per nanomole of CYP for 15 min. After preincubation, the buffer, substrate, and NADPH-generating system were added. Metabolites were detected as described elsewhere (30 –35). Production of antibodies. Polyclonal anti-recombinant CYP2A6 antibodies were raised in five male chinchilla rabbits at 5– 6 months of age (VUFB, Konarovice, Czech Republic). Rabbit sera (1/200 dilution) were taken prior to immunization and subsequently used for preliminary screening by immunoblotting with human liver microsomes. Only rabbits with sera showing no reaction on the blot were selected. Rabbits were immunized on days 1, 14, and 42 with five 0.1-ml subcutaneous shots in their shaved backs. Each immunization was done with 0.5 nmol of purified CYP, diluted to 0.25 ml with 0.9% saline and further diluted to 0.5 ml with Freund’s complete (first immunization) or incomplete (second and third immunization) adjuvants. On day 56, rabbits were bled and IgG were prepared by affinity chromatography using protein A agarose gel according to the producer’s protocol (Pierce, Rockford, IL), followed by extensive dialysis against 100 vol of 50 mM potassium phosphate buffer (pH 7.4). Specificity of the raised antibodies was assessed using immunoblot-
ting (0.1 mg of IgG per blot used) with human liver microsomes and purified CYP standards (CYP 1A1, 1A2, 2A6, 2C9, 2D6, 2E1, 3A4, and 3A5). The inhibitory potency of anti 2A6 IgG was assayed by following the marker activities of various CYP enzymes in human liver microsomes (see assays of catalytic activity) in comparison with IgG taken prior to immunization. Other procedures and assays. Protein concentrations were estimated using the bicinchoninic acid method according to general methods provided by the supplier (Pierce). CYP spectra were recorded with a Specord M400 instrument (CarlZeiss, Jena, Germany) using the general methods of Omura and Sato (25) for quantitation of CYP content. NADPH cytochrome c reduction was determined as described elsewhere (28) using De 550 5 21 mM 21 cm 21 and an assumed specific activity of 3200 nmol of reduced cytochrome c min 21 nmol reductase 21 (28). SDS electrophoresis was done in 10% (w/v) polyacrylamide gels by the method of Laemmli (36) using a MiniProtean apparatus (BioRad, Hercules, CA). Protein staining was done with ammoniacal silver (37) and immunoblotting was performed as described (38) using the conditions for development of blots according to Soucˇek et al. (39). N-terminal amino acid sequencing was done in the Midwest Analytical (St. Louis, MO) facility using a Model 477A Protein Sequencer (Applied Biosystems, Foster City, CA) and methodology based on Edman’s degradation. SDS electrophoresis, transfer of protein to Immobilon-P membrane (Millipore Corp., Bedford, MA), and staining are described elsewhere (39). Blots were treated for 24 h by 0.6 N HCl at 25°C to remove the N-terminal formylmethionine block as described by Hirano et al. (40). Yields at each cycle were estimated by comparison with external standards. The Mfold and LoopViewer programs for MacIntosh computers were used to predict the energy of formation of stable secondary structure at the N-terminus. According to the results of computation, primers with favorable N-terminal sequences (usually with higher AT content) were designed for PCR.
RESULTS
Expression of N-terminal variants. Various N-terminal modifications of the CYP2A6 clone were designed (Table I) according to previous successful ap-
FIG. 1. Construction of pCW/2A6 expressing plasmid. Unique restriction sites are displayed. 193
194
ˇ EK PAVEL SOUC
FIG. 2. Spectrophotometric evaluation of CYP2A6 in whole E. coli cells. A, pCW; B, pCW/2A6native; C, pCW/2A6A. 1 ml of cells after 40 h expression was taken and centrifuged at 12,000g for 5 min. The pellet was resuspended in 3 ml of 100 mM potassium phosphate (pH 7.4) containing 10 mM dextrose and 6 mM magnesium acetate. Then baseline (. . .) and Fe 21.CO vs Fe 21 difference spectra (—) were recorded as described (25).
EXPRESSION OF CYTOCHROME P450 2A6 IN BACTERIA TABLE II
Purification of CYP2A6 from E. coli Membrane Fraction Purification step
Protein a (mg)
CYP (nmol)
Specific content (nmol CYP/mg protein)
Yield (%)
Cells Membranes c DEAE-Sephacel CM-Sepharose HA-Ultrogel
2b 415 87 8.9 3.9
210 178 141 88 48
2b 0.43 1.62 9.93 12.35
100 85 67 42 23
a All values correspond to material recovered from 1 liter of culture. b Protein content in the whole cells was not measured. c Values displayed for solubilized membranes (supernatant after 105,000g spin).
proaches (23, 27, 41). In all sequences the AT content was enhanced in order to reduce the potential for unfavorable mRNA secondary structure formation. Changes in codons were chosen in agreement with the list of E. coli preferred codons (42) where it was possible. Construct A was designed to substitute Leu at the second position by Ala as this N-terminal amino acid change was reported to facilitate expression in E. coli (43). The N-terminal sequence MALLAVFL previously reported to improve expression of various CYP proteins (23, 41, 44, 45) was introduced to constructs B and C. Constructs D and E contained truncated versions of the CYP2A6 native sequence. The sequence modification for construct F was derived from the N-terminal sequence of CYP2E1 successfully expressed in E. coli before (27). pCW/2A6 constructs bearing N-terminal modifications were then prepared (Fig. 1) and the expression level was monitored. As expected, the construct with the native sequence of CYP2A6 did not show detectable hemoprotein production as well as the control carrying pCW only (Figs. 2A and 2B). On the other hand, the construct designated pCW/2A6A, bearing only a single amino acid substitution, appeared to be the most effective in hemoprotein production as 210 nmol of CYP2A6 per liter of culture was detected (Fig. 2C, Table II). This variant of CYP2A6 was selected for further purification and characterization. Expression from the CYP2A6A variant was optimized with respect to temperature (32°C) and time (40 h) of cultivation. Stimulation (about 30% increase) of CYP2A6 production with the addition of 0.5 mM d-aminolevulinic acid was observed. Purification of CYP2A6. The three-step column chromatography employing DEAE-Sephacel, CMSepharose, and hydroxylapatite was used. About 80% of the initial CYP2A6 load was recovered in the DEAE void fraction and a 40% loss of hemoprotein in CMeluate was observed when compared to the DEAE void fraction (Table II). The CYP2A6 concentration in the
195
top hydroxylapatite-eluate fraction was 16 mM with a CYP-specific content of 12.35 nmol CYP/mg protein (Table II). The CYP2A6 preparation was pure as judged by electrophoresis (Fig. 3). Spectral properties of CYP2A6. Absolute spectra of purified recombinant CYP2A6 were recorded (Fig. 4) and the wavelength maxima and extinction coefficients were calculated (Table III). Oxidized CYP2A6 was in a low-spin state as indicated by the peak at 418 nm (Fig. 4A). Addition of the substrate coumarin shifted the spin state of oxidized CYP2A6 from low to high spin, indicating the binding of substrate to the CYP active site (see second derivative spectra, Fig. 4B). Transition of the spin state in the presence of coumarin proceeded in a concentration-dependent manner. In contrast to CYP1A2 (41), CYP2A6 was readily reduced by sodium dithionite. Cytochrome P420 was absent in the CYP2A6 preparation as judged from Fe 21.CO spectra (Fig. 4A). N-terminal sequence analysis of recombinant CYP2A6. N-terminal amino acid analysis showed that the N-terminus of purified CYP2A6 was blocked most probably by N-formylmethionine. This block was removed by treatment of the blot strip containing 20 pmol of purified CYP2A6 with 0.6 N HCl for 24 h at 25°C. The expected N-terminal sequence was then obtained. The sequence was clearly readable, although yields appear to be fairly low (Table IV). Catalytic activities of recombinant CYP2A6. Coumarin 7-hydroxylation was examined as a marker activity of human CYP2A6. Initial experiments were de-
FIG. 3. SDS–polyacrylamide electrophoresis of CYP2A6 fractions obtained during purification. An amount of sample corresponding to 5 pmol of CYP was applied per lane. The direction of migration was from top to bottom. Lanes included: 1, membrane fraction; 2, solubilized membranes (105,000g supernatant fraction), 3, DEAE-Sephacel void fraction; 4, CM-Sepharose gradient eluate fraction; 5, best fraction eluted from hydroxylapatite.
196
ˇ EK PAVEL SOUC
FIG. 4. Spectral properties of CYP2A6. (A) Absolute UV-visible spectra of purified recombinant CYP2A6 : Fe 31 (), Fe 21(ª), Fe 21.CO (. . .). (B) Second derivative spectra of purified recombinant CYP2A6 in the absence (. . .) and in the presence of 2 mM (ª) or 20 mM () coumarin. All spectra were recorded with 4.3 mM (A) or 2.4 mM (B) CYP2A6 in 100 mM potassium phosphate (7.4) containing 20% glycerol (v/v) and 1 mM EDTA. Reduction was accomplished by addition of a few milligrams of solid sodium dithionite. The inset shows expanded a and b Soret regions.
signed in order to optimize the composition of the CYP2A6 reconstituted system. In the reconstituted system containing 25 pmol of purified recombinant CYP2A6 and components described under Materials and Methods, the amount of rat NADPH-P450 reductase varied from 6.25 to 250 pmol (CYP/rat NADPHP450 reductase ratio, 4/1 to 1/10). Maximum activity was found when a 1/5 ratio of CYP/rat NADPH-P450 reductase was present (Fig. 5). The activity at 1/3 ratio (4.81 nmol product/min/nmol CYP) was found to be reasonably high, and due to the limited supply of rat NADPH-P450 reductase this ratio was used for subsequent studies. It was published that addition of some chemicals enhances the catalytic activity of CYP in the
reconstituted system (23). Therefore, it was of interest whether effects of these components are CYP-isoenzyme specific or whether it is a general observation. The effect of cytochrome b 5 in a CYP2A6-catalyzed reaction was analyzed as well. From the data presented in Table V it is evident that cytochrome b 5 had a great stimulatory effect on CYP2A6 catalysis. The highest activity was observed at a 1/1 ratio of CYP2A6/ cytochrome b 5 (more than a twofold increase). Addition of 1–5 mM glutathione enhanced CYP2A6 activity up to 50%. The presence of the detergent sodium cholate, variations in the buffer system, or concentrations of MgCl 2 up to 30 mM did not significantly influence the activity of CYP2A6. On the contrary, a higher concen-
EXPRESSION OF CYTOCHROME P450 2A6 IN BACTERIA
197
TABLE III
Spectral Properties of Purified Recombinant Human CYP2A6
Form of CYP2A6 a
Wavelength maximum (nm)
Extinction coefficient (e, mM 21 cm 21)
418 534 568 388 410 544 447 555
131 11.4 10.8 98 113 15.3 127 14.3
Fe 31 Fe 31-coumarin Fe 21 Fe 21.CO Fe 21.CO vs Fe 21 (difference)
91.0 b
448
a
Spectra were recorded in 100 mM potassium phosphate buffer (pH 7.4) containing 1.0 mM EDTA, 20% glycerol (v/v), and 4.3 mM purified CYP2A6. For substrate spectra 20 mM coumarin and 2.4 mM CYP2A6 were used. b Assuming Omura and Sato’s calculations (25).
tration of MgCl 2 strongly inhibited 7-hydroxylation of coumarin (Table V). Analysis of enzymatic kinetics of coumarin 7-hydroxylation in the CYP2A6 reconstituted system revealed typical Michaelis–Menten kinetics (Fig. 6; K m 5 1.48 6 0.37 mM, V max 5 3.36 6 0.18 nmol product/min/nmol CYP). The CYP2A6 reconstituted system was used to investigate the role of CYP2A6 in oxidation of chlorzoxazone, the purported probe of CYP2E1 in vivo. In the standard system described under Materials and Methods, it was found that CYP2A6 (0.05 mM) metabolized 0.5 mM chlorozoxazone to 6-hydroxychlorzoxazone at turnover rate 0.32 nmol product/min/nmol CYP. Under
TABLE IV
N-Terminal Amino Acid Sequence Analysis of Recombinant CYP2A6 Purified from E. coli Amino acid Cycle/position
Expected
Found
Recovered a (pmol)
1 2 3 4 5 6 7 8 9 10
M A A S G M L L V A
M A A S G M L L V A
2.1 2.1 2.1 0.9 1.6 1.5 1.8 2.3 1.5 1.1
a
A nominal amount of 20 pmol of purified CYP2A6 was used for SDS electrophoresis and blotting transfer.
FIG. 5. Coumarin 7-hydroxylation activity of purified recombinant CYP2A6 in the reconstituted system— dependence on rat NADPHP450 reductase amount. The CYP2A6 reconstituted system contained 0.5 ml of the following mixture: 25 pmol of CYP2A6, 6.25, 12.5, 25, 50, 75, 150, or 250 pmol of rat NADPH-P450 reductase, 30 mM L-a-dilauroyl-sn-3-phosphocholine, 50 mM potassium phosphate buffer (pH 7.4), 20 mM coumarin, and the NADPH-generating system (10 mM MgCl 2, 10 mM glucose 6-phosphate, 1 mM NADP, 0.5 U/ml glucose-6-phosphate dehydrogenase). Incubation proceeded for 15 min and metabolite was determined as described under Materials and Methods. All assays were performed in duplicates with variations less than 5%.
the same conditions, chlorzoxazone was metabolized by human liver microsomes (0.2 mM total CYP) at a turnover rate of 0.94 nmol product/min/nmol CYP. Production of polyclonal antibodies against recombinant CYP2A6. Polyclonal antibodies against purified recombinant CYP2A6 have been raised in rabbits. These antibodies proved to be very specific and useful for immunoblotting as they recognize CYP2A6 well (Fig. 7A). Very weak crossreactivity with purified recombinant CYP2C9 and 2E1 (but not with 1A1, 1A2, 2D6, 3A4, and 3A5) was also observed. Crossreactive CYPs migrate with lower mobility than CYP2A6 and therefore may be easily distinguished from the CYP2A6-corresponding band found in human liver microsomes where a minor crossreactive band of higher M r was also detected (Fig. 7B). Anti-CYP2A6 antibodies very specifically and efficiently inhibited CYP2A6 catalytic activity toward coumarin in human liver microsomes (Fig. 8). Significant inhibition was achieved by 1 mg of anti-CYP2A6 IgG/nmol of total CYP. On the other hand, up to 5 mg of anti-2A6 IgG/nmol of total CYP did not significantly inhibit other human CYP (1A, 2C9, 2D6, 2E1, and 3A4) marker activities. Inhibition of coumarin 7-hydroxylation in human liver microsomes by rabbit IgG taken prior to immunization also showed negligible effect (Fig. 8).
ˇ EK PAVEL SOUC
198 TABLE V
Coumarin 7-Hydroxylation in Human Liver Microsomes and the CYP2A6 Reconstituted System—Influence of Cytochrome b 5, Buffers, Salts, Glutathione, and Detergent
Enzyme preparation Human liver microsomes (n 5 10) a Purified recombinant CYP2A6: No addition b 1b 5 - b 5/CYP ratio 1/1 1b 5 - b 5/CYP ratio 2/1 1b 5 - b 5/CYP ratio 4/1 1150 mM KCl/50 mM Tris (pH 7.4) 150 mM Hepes (pH 7.4) 1Glutathione, 1 mM 1Glutathione, 3 mM 1Glutathione, 5 mM 1MgCl 2, 30 mM 1MgCl 2, 60 mM 1Sodium cholate, 50 mg 1Sodium cholate, 100 mg 1Sodium cholate, 200 mg
7-Hydroxylation of coumarin (nmol product/ min/nmol CYP) 1.79 6 0.69 4.06 10.15 8.18 6.92 4.20 4.36 4.27 5.51 6.23 4.08 2.64 3.05 4.60 4.07
Note. For 7-hydroxycoumarin determination see Materials and Methods. All assays were performed in duplicates with less than 5% variation. a Human liver microsomes were incubated 15 min at 37°C with 20 mM coumarin and the NADPH-generating system (10 mM MgCl 2, 10 mM glucose 6-phosphate, 1 mM NADP, 0.5 U/ml glucose-6-phosphate dehydrogenase) in a 0.5-ml total volume of 50 mM potassium phosphate buffer (pH 7.4). b The reconstituted system contained 25 pmol CYP2A6, 75 pmol rat NADPH-P450 reductase, 30 mM L-a-dilauroyl-sn-3-phosphocholine, 50 mM potassium phosphate (pH 7.4), 20 mM coumarin, and the NADPH-generating system (see above) in total volume 0.5 ml. Incubation proceeded for 15 min at 37°C.
Purification of the CYP2A6 protein proceeded quite straitforwardly, contrary to reported difficulties with CYP2D6 and 3A4 proteins (23, 46), and it seems that there is no need to facilitate purification by tagging the protein with oligo-His sequences or other modifications of protein. Spectral properties of purified CYP2A6 protein were not examined so far. Absolute spectra of CYP2A6 presented here appear to give typical results previously seen with other CYP proteins and showed high similarity particularly with CYP2D6 expressed in E. coli (46). Contrary to the majority of other E. coli-expressed CYPs (23, 26, 27, 41, 45, 47), CYP2A6 was isolated in the low-spin state, similarly to CYP2D6 (46). Substrate coumarin shifted the low-spin state of oxidized CYP2A6 to high spin, indicating the binding to the active site of CYP2A6. N-terminal amino acid sequence analysis revealed a N-formylmethionine blockade. It seems that this block is related to the high hydrophobicity of the CYP2A6 N-terminus as it was observed in several E. coli-expressed CYP proteins, all bearing highly hydrophobic MALLAVFL sequence modification (48). Using Kyte and Doolittle’s hydropathy plot (49) it was found that the CYP2A6 N-terminus has also a very high degree of hydrophobicity (results not shown). N-Formylmethionine was removed by mild acid treatment and the amino acid sequence analysis confirmed that the ex-
DISCUSSION
Although CYP2A6 has been considered to be a minor CYP form present in human liver, the amount of information about this particular CYP isoform in the literature is increasing in the last few years. The gene encoding this CYP enzyme has been shown to be present in at least three allelic variants (12, 13). Despite the rising interest in CYP2A6, the isolation of this protein from a heterologous system and the production of antibodies against it have not been reported. Moreover, the E. coli expressing system could also contribute to the study of activity and affinity of CYP2A6 allelic variants. Among the six variants of CYP2A6 examined here, the N-terminal variant with only a single amino acid change (Ala instead of Leu at the second position) showed the best expression efficiency. As in the CYP2D6 protein (46), supplementation of expression medium with d-aminolevulinic acid proved to enhance expression levels of CYP2A6.
FIG. 6. Enzyme kinetics of coumarin 7-hydroxylation activity of purified recombinant CYP2A6 in the reconstituted system. The CYP2A6 reconstituted system contained 0.5 ml of the following mixture: 25 pmol of CYP2A6, 75 pmol of rat NADPH-P450 reductase, 30 mM L-a-dilauroyl-sn-3-phosphocholine, 50 mM potassium phosphate buffer (pH 7.4), 0.2, 0.5, 1, 2, 5, 10, 20, or 50 mM coumarin, and the NADPH-generating system (10 mM MgCl 2, 10 mM glucose 6-phosphate, 1 mM NADP, 0.5 U/ml glucose-6-phosphate dehydrogenase). Incubation proceeded for 10 min and metabolite was determined as described under Materials and Methods. All assays were performed in duplicates with variations less than 5%. K m 5 1.48 6 0.37 mM, V max 5 3.36 6 0.18 nmol product/min/nmol CYP.
EXPRESSION OF CYTOCHROME P450 2A6 IN BACTERIA
pressed CYP2A6A variant had the desired N-terminal sequence. Recombinant CYP2A6 expressed and purified in this study was shown to be catalytically active toward the widely used marker substrate coumarin with a K m (1.48 mM) comparable to values found in human liver microsomes (7, 11). The presence of an equal amount (to CYP) of cytochrome b 5 greatly stimulated 7-hydroxylation of coumarin similarly as it was noted for E. coli-expressed CYP’s 2C9, 2E1, 3A4, and 3A5 (23, 26, 27, 45). CYP2A6 activity was also enhanced by the addition of glutathione (up to 5 mM). Stimulation of catalytic activity by glutatione was reported for recombinant CYP3A4 (23). On the other hand, neither the change in buffer composition nor the change in the concentration of divalent metal ions or cholate stimulates CYP2A6 activity, in contrast to data on CYP3A4 but in agreement with data on CYP2C9 (50). Moreover, it was shown that CYP2A6 in the reconstituted system metabolizes chlorzoxazone at a more
199
FIG. 8. Immunoinhibition of CYP marker activities in human liver microsomes by anti-CYP2A6 IgG. Incubations were carried out as described under Materials and Methods. Control (uninhibited) CYP activities in nmol product/min/nmol CYP were: 0.12, 7-ethoxyresorufin O-deethylation (*); 2.20, coumarin 7-hydroxylation (E); 0.39, tolbutamide methylhydroxylation (‚); 0.33, bufuralol 19-hydroxylation ( ); 0.94, chlorzoxazone 6-hydroxylation (h); 2.33, testosterone 6bhydroxylation (3). Rabbit IgG isolated from the sera taken prior to immunization ({) were used in amounts of 1, 3, or 5 mg IgG/nmol CYP. All assays were performed in duplicates with variations less than 5%.
than notable rate and thus it should be taken into consideration when in vivo phenotyping of human CYP2E1 is performed. From data presented here it seems that it would be very interesting to follow the activity of the reconstituted system with added cytochrome b 5, glutathione, and sodium cholate together or to compare enzyme kinetics data with the experiment done in the presence of cytochrome b 5, where a change of kinetic parameters may be expected similarly as it was observed for recombinant CYP2E1 (27). The catalytic activity of CYP2A6 toward chlorzoxazone in the reconstituted system may be much higher in the presence of cytochrome b 5 as well. These experiments are in progress now. Polyclonal antibodies raised against recombinant CYP2A6 showed very good specificity in immunoblotting and immunoinhibition and may serve as an important tool in future studies of the role of CYP2A6 in the metabolism of chemicals and in procarcinogen activation. FIG. 7. Immunoblotting of CYP proteins and human liver microsomes with anti-CYP2A6 IgG. (A) 1 pmol of purified recombinant CYP was used per lane: 1, CYP 1A1; 2, CYP1A2; 3, CYP2A6; 4, CYP2C9; 5, CYP2D6; 6, CYP2E1; 7, CYP3A4; 8, CYP3A5. (B) Lane 1, 1 pmol of purified CYP2A6; lanes 2–9, 20 mg of protein from eight individual samples of human liver microsomes was used per lane. Blots were developed using 0.1 mg of anti-2A6 IgG as described under Materials and Methods.
ACKNOWLEDGMENTS The author expresses his sincere thanks to Dr. Guengerich for his generous gifts (CYP’s, pCW/NF14, and methylhydroxy tolbutamide) and essential expertize, Dr. Gonzalez for pUC2A6, Dr. Yamazaki for human cytochrome b 5, and Drs. Pavel Anzenbacher, Hermann Esselmann, Elizabeth Gillam, and David McCourt for helpful discussions during this project.
200
ˇ EK PAVEL SOUC
REFERENCES 1. Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J., Feyereisen, R., Waxman, D. J., Waterman, M. R., Gotoh, O., Coon, M. J., Estabrook, R. W., Gunsalus, I. C., and Nebert, D. W. (1996) Pharmacogenetics 6, 1– 42. 2. Guengerich, F. P. (1995) in Cytochrome P450: Structure, Mechanism, and Biochemistry (Ortiz de Montellano, P. R., Ed.), pp. 473–535, Plenum, New York. 3. Sawada, M., and Kamataki, T. (1998) Mutat. Res. 411, 19 – 43. 4. Clarke, S. E. (1998) Xenobiotica 28, 1167–1202. 5. Dong, M. S., Yamazaki, H., Guo, Z., and Guengerich, F. P. (1996) Arch. Biochem. Biophys. 327, 11–19. 6. Waterman, M. R., Jenkins, C. M., and Pikuleva, I. (1995) Toxicol. Lett. 82/83, 807– 813. 7. Yamano, S., Tatsuno, J., and Gonzalez, F. J. (1990) Biochemistry 29, 1322–1329. 8. Miles, J. S., McLaren, A. W., Forrester, L. M., Glancey, M. J., Lang, M. A., and Wolf, C. R. (1990) Biochem. J. 267, 365–371. 9. Yun, C. H., Shimada, T., and Guengerich, F. P. (1991) Mol. Pharmacol. 40, 679 – 685. 10. Shimada, T., Yamazaki, H., Mimura, M., Inui, Y., and Guengerich, F. P. (1994) J. Pharm. Exp. Ther. 270, 414 – 423. 11. Shimada, T., Yamazaki, H., and Guengerich, F. P. (1996) Xenobiotica 26, 395– 403. 12. Oscarson, M., Gullsten, H., Rautio, A., Bernal, M. L., Sinues, B., Dahl, J. H., Stengard, J. H., Pelkonen, O., Raunion, H., and Ingelman-Sundberg, M. (1998) FEBS Lett. 438, 201–205. 13. Fernandez-Salguero, P., Hoffman, S. M. G., Cholerton, S., Mohrenweiser, H., Raunio, H., Raution, A., Pelkonen, O., Huang, J., Evans, W. E., Idle, J. R., and Gonzalez, F. J. (1995) Am. J. Hum. Genet. 57, 651– 660. 14. Imaoka, S., Yamada, T., Hiroi, T., Hayashi, K., Sakaki, T., Yabusaki, Y., and Funae, Y. (1996) Biochem. Pharmacol. 51, 1041–1050. 15. Spracklin, D. K., Thummel, K. E., and Kharasch, E. D. (1996) Drug Metab. Dispos. 24, 976 –983. 16. Nakajima, M., Yamamoto, T., Nunoya, K-I., Yokoi, T., Nagashima, K., Inoue, K., Funae, Y., Shimada, N., Kamataki, T., and Kuroiwa, Y. (1996) Drug Metab. Dispos. 24, 1212–1217. 17. Aoyama, T., Shigeru, Y., Guzelian, P. S., Gelboin, H. V., and Gonzalez, F. J. (1990) Proc. Natl. Acad. Sci. USA 87, 4790 – 4793. 18. Duescher, R. J., and Elfarra, A. A. (1994) Arch. Biochem. Biophys. 311, 342–349. 19. Crespi, C. L., Penman, B. W., Leakey, J. A., Arlotto, M. P., Stark, A., Parkinson, A., Turner, T., Steimel, D. T., Rudo, K., Davies, R. L., and Langenbach, R. (1990) Carcinogenesis 11, 1293–1300. 20. Yamazaki, H., Inui, Y., Yun, C-H., Guengerich, F. P., and Shimada, T. (1992) Carcinogenesis 13, 1789 –1794. 21. Pritchard, M. P., Ossetian, R., Li, D. N., Henderson, C. J., Burchell, B., Wolf, C. R., and Friedberg, T. (1997) Arch. Biochem. Biophys. 345, 342–354. 22. van der Hoeven, T. A., and Coon, M. J. (1974) J. Biol. Chem. 249, 6302– 6310. 23. Gillam, E. M. J., Baba, T., Kim, B-R., Ohmori, S., and Guengerich, F. P. (1993) Arch. Biochem. Biophys. 305, 123–131. 24. Bauer, S., and Shiloach, J. (1974) Biotechnol. Bioeng. 16, 933– 941.
25. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239, 2370 –2378. 26. Sandhu, P., Baba, T., and Guengerich, F. P. (1993) Arch. Biochem. Biophys. 306, 443– 450. 27. Gillam, E. M. J., Guo, Z., and Guengerich, F. P. (1994) Arch. Biochem. Biophys. 312, 59 – 66. 28. Yasukochi, Y., and Masters, B. S. S. (1976) J. Biol. Chem. 251, 5337–5344. 29. Guengerich, F. P. (1994) in Principles and methods of toxicology (Hayes, A. W., Ed.), pp. 1259 –1313, Raven, New York. 30. Soucˇek, P. (1999) A novel sensitive HPLC method for assay of coumarin 7-hydroxylation. J. Chromatogr., in press. 31. Lubet, R. A., Mayer, R. T., Cameron, J. W., Nims, R. W., Burke, M. D., Wolf, T., and Guengerich, F. P. (1985) Arch. Biochem. Biophys. 283, 43– 48. 32. Knodell, R. G., Hall, S. D., Wilkinson, G. R., and Guengerich, F. P. (1987) J. Pharm. Exp. Ther. 241, 1112–1119. 33. Yamazaki, H., Guo, Z., Persmark, M., Mimura, M., Inoue, K., Guengerich, F. P., and Shimada, T. (1994) Mol. Pharmacol. 16, 568 –577. 34. Peter, R., Bocker, R., Beaune, P. H., Iwasaki, M., Guengerich, F. P., and Yang, C. S. (1990) Chem. Res. Toxicol. 3, 566 –573. 35. Brian, W. R., Sari, M. A., Iwasaki, M., Shimada, T., Kaminsky, L. S., and Guengerich, F. P. (1990) Biochemistry 29, 11280 – 11292. 36. Laemmli, U. K. (1970) Nature 227, 680 – 685. 37. Wray, W., Boulikas, T., Wray, V. P., and Hancock, R. (1981) Anal. Biochem. 118, 197–203. 38. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350 – 4356. 39. Soucˇek, P., Martin, M. V., Ueng, Y. F., and Guengerich, F. P. (1995) Biochemistry 34, 16013–16021. 40. Hirano, H., Komatsu, S., Kajiwara, H., Takagi, Y., and Tsunasawa, S. (1993) Electrophoresis 14, 839 – 846. 41. Sandhu, P., Guo, Z., Baba, T., Martin, M. V., Tukey, R. H., and Guengerich, F. P. (1994) Arch. Biochem. Biophys. 309, 168 –177. 42. (1987) in Current Protocols in Molecular Biology (Ausubel, F. M., et al., Eds.), Greene, PA; Wiley, Brooklyn, NY. 43. Looman, A. C., Bodlaender, J., Comstock, J. L., Eaton, D., Jhurani, P., deBoer, H. A., and Knippenberg, P. H. (1987) EMBO J. 6, 2489 –2492. 44. Barnes, H. J., Arlotto, M. P., and Waterman, M. R. (1991) Proc. Natl. Acad. Sci. USA 88, 5597–5601. 45. Gillam, E. M. J., Guo, Z., Ueng, Y-F., Yamazaki, H., Cock, I., Reilly, P. E. B., Hooper, W. D., and Guengerich, F. P. (1995) Arch. Biochem. Biophys. 317, 374 –384. 46. Gillam, E. M. J., Guo, Z., Martin, M. V., Jenkins, C. M., and Guengerich, F. P. (1995) Arch. Biochem. Biophys. 319, 540 –550. 47. Guo, Z., Gillam, E. M. J., Ohmori, S., Tukey, R. H., and Guengerich, F. P. (1994) Arch. Biochem. Biophys. 312, 436 – 446. 48. Dong, M-S., Bell, C. L., Guo, Z., Phillips, D. R., Blair, I. A., and Guengerich, F. P. (1996) Biochemistry 35, 10031–10040. 49. Kyte, J., and Doolittle, R. F. (1982) J. Mol. Biol. 157, 105–132. 50. Yamazaki, H., Gillam, E. M. J., Dong, M-S., Johnson, W., Guengerich, F. P., and Shimada, T. (1997) Arch. Biochem. Biophys. 342, 3329 –3337.
Expression of Cytochrome P450 2A6 in Escherichia coli: Purification, Spectral and Catalytic Characterization, and Preparation of Polyclonal Antibodies 1 Pavel Soucˇek 2 Biotransformations Group, Center of Occupational Diseases, National Institute of Public Health, Sˇroba´rova 48, Praha 10, 100 42, Czech Republic
Received March 23, 1999, and in revised form July 6, 1999
Cytochrome P450 (CYP) 2A6 is the principal human enzyme catalyzing coumarin 7-hydroxylation and is known to be involved in the metabolism of halothane, nicotine, and metabolic activation of butadiene and nitrosamines. In this paper expression of CYP2A6 in Escherichia coli is reported. In order to achieve expression, the N-terminus of protein was modified by PCR mutagenesis. The N-terminal variant with only a single amino acid change showed expression of 210 nmol of CYP2A6/liter of culture. Recombinant CYP2A6 protein was purified to electrophoretic homogeneity and further characterized. Absolute spectra were typical for CYP proteins and indicated low spin characteristics of isolated protein. Due to a hydrophobic segment the N-terminal amino acid sequence of recombinant CYP2A6 was blocked. The N-terminal formylmethionine block was removed by mild acid treatment. Purified CYP2A6 had good catalytic activity toward marker substrate coumarin in a reconstituted system (K m 5 1.48 6 0.37 mM, V max 5 3.36 6 0.18 nmol product/min/nmol CYP). Its activity in the reconstituted system was stimulated by the presence of cytochrome b 5 and glutathione. CYP2A6 was shown to metabolize chlorzoxazone in the reconstituted system with activity of 0.32 nmol of product/min/nmol of CYP, and thus caution should be taken when interpretation of CYP2E1 in vivo phenotyping data is performed. Rabbit polyclonal antibodies were produced against recombinant CYP2A6 and proved to be very useful for immunoblotting and immunoinhibition studies. Availability of this expression system and specific antibodies should facilitate characterization of the role of
1 Supported by the Internal Grant Agency of Czech Ministry of Health, Grant IGA 3505-3 (1996-98), and the Grant Agency of Czech Republic, Grant 303/96/0373 (1996). 2 Fax: 1420-2-6731 1236. E-mail: [email protected].
190
CYP2A6 in the metabolism of chemicals and in the study biological relevance of genetic polymorphisms of this enzyme. © 1999 Academic Press Key Words: CYP2A6; heterologous expression; protein purification; catalytic activity; coumarin; antibodies.
Cytochrome P450 (CYP) 3 enzymes (EC 1.14.14.1) are the principal metabolizing enzymes involved in detoxication and/or activation of compounds of both natural and synthetic origin (2). Some CYP-catalyzed reactions result in the formation of cytotoxic or genotoxic intermediates (3). Heterologous expression systems including mammalian cells, yeasts, insect cells, and bacteria (for review, see 4) have been developed in order to increase the understanding of the role of individual CYP enzymes in the metabolism of xenobiotics and its consequences. Heterologous expression systems also serve as a tool to investigate the functional relevance of genetic polymorphisms of metabolizing enzymes. Moreover, expression of mutated CYPs may be the key approach in crystallization and in the elucidation of the secondary structure of these membrane proteins (5). Of the systems mentioned above, bacterial systems possess many advantages in terms of low cost, ease of use, and ammenability to scale up (6). Another advantage of highscale production of ultrapure protein in bacteria is the possibility to raise specific and possibly inhibitory antibodies which could serve as an important aid in studies of the metabolism of CYP substrates in subcellular fractions. N-terminally modified forms of many human 3 Abbreviation used: CYP, cytochrome P450 (for reference to the nomenclature, see 1).
0003-9861/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
EXPRESSION OF CYTOCHROME P450 2A6 IN BACTERIA
CYPs have been expressed in Escherichia coli (for review, see 3 and 4) and proved to be useful for catalytic activity assays. The CYP2A6 gene was cloned and sequenced in 1990 (7, 8). CYP2A6 protein was first purified from human liver and polyclonal antibodies has been produced against this protein (9). However, this preparation was contaminated by at least one CYP form and further purification was hindered by a limited supply of human liver. It was shown that relative content of CYP2A6 between individual human livers differs up to 260-fold and its relative content accounts for about 4% of total CYP content (2, 10, 11). Genetic polymorphism reflecting in three allelic variants (recently designated 2A6*1, 2A6*2, and 2A6*3; see 12) was identified (13). CYP2A6 was expressed in mammalian cells and yeasts (4, 14) and was shown to be responsible for the oxidation of coumarin (marker substrate), halothane, and nicotine and for the metabolic activation of 1,3-butadiene and nitrosamines (15–20). Recently, CYP2A6 was also expressed in E. coli with an attached bacterial signal peptide (21). Although the level of expression was high, this protein was not purified to homogeneity and the activity of the membrane fraction was quite low (21). No spectral or catalytic studies using purified recombinant CYP2A6 as well as the production of inhibitory polyclonal antibodies have been published so far. Here, the expression of a large amount of slightly modified (second amino acid Ala instead of Leu) CYP2A6, purification to electrophoretic homogeneity, and thorough spectral and catalytic study are reported. Moreover, inhibitory polyclonal antibodies against expressed CYP2A6 have been successfully produced. Purified CYP2A6 was used to elucidate its participation in the metabolism of the purported CYP2E1 probe substrate, chlorzoxazone. MATERIALS AND METHODS Chemicals. The expression vector pCW/NF14, HPLC standard for assay of oxidation of tolbutamide, and purified CYP standards for immunoblotting were generous gifts of Dr. F. Peter Guengerich (Vanderbilt University, Nashville, TN), and the CYP2A6 cDNA was provided by Dr. Frank J. Gonzalez (National Institute of Health, Bethesda, MD) cloned into the plasmid pUC/2A6. For bacterial transformation, DH5a E. coli competent cells (Life Technologies Ltd., Paisley, UK) were used. All chemicals used in expression, purification, and activity assays were products of Sigma (St. Louis, MO) except bufuralol and 19-hydroxy bufuralol (Gentest Corp., Woburn, MA). Restriction enzymes were products of BoehringerMannheim GmbH (Mannheim, Germany), and reagents for PCR were purchased from Perkin Elmer Corp. (Norwalk, CT). Purified human cytochrome b 5 was a generous gift of Dr. Hiroshi Yamazaki (Kanazawa University, Kanazawa, Japan). Human liver donors. Human liver samples were obtained from the Transplantation Center (IKEM, Czech Republic) from donors who died accidentally as a result of brain injury. Liver samples were obtained at most 30 min after death and were stored in liquid nitrogen until microsomes were isolated. Some of the donors were
191
submitted to therapy including antibiotics, mannitol, or hormones for 24 – 48 h before death. Microsomes were prepared according to van der Hoeven and Coon (22) and stored in aliquots of 200 ml at 280°C. Construction of expression plasmids. CYP2A6 was cloned between EcoRI (59-end) and XbaI (39-end) sites of pUC18 plasmid. N-terminal modifications of CYP2A6 clone were performed by PCR mutagenesis using primers introducing the NdeI site (CATATG) and the desired N-terminal change at the 59-terminus and the SalI site (GTCGAC) at the 39-terminus. The resulting DNA fragment was cleaved by NdeI and SalI restriction enzymes and resolved by horizontal electrophoresis in 1% agarose gel. An agarose band bearing a DNA fragment of a length corresponding to that of the CYP2A6 clone was extracted by GenecleanII (Bio 101, Vista, CA) and used for ligations into pCW torso, which was prepared by cleavage of expression plasmid pCW/NF14 (23) by NdeI and SalI and further processed in the same way as described above for the CYP2A6 fragment (see Fig. 1). Ligation of CYP2A6 fragment and pCW torso was achieved by the use of the DNA ligation kit according to producer’s protocol (Boehringer-Mannheim). Expression of plasmids in E. coli. Expression of N-terminal variants of pCW/2A6 was assessed following the basic procedure described previously (23). E. coli strain DH5a was transformed with each plasmid by heat shock and selected on Luria–Bertani agar plates containing ampicillin. Single colonies were then grown overnight in Luria–Bertani medium supplemented with ampicillin (100 mg/ml) at 37°C. An overnight culture (1 ml) was used to seed 100 ml of expression medium (23) where trace elements were added (24). Protein expression was induced by 1 mM isopropyl b-D-thiogalactoside for 40 h at 32°C with shaking at 200 RPM in a refrigerated orbital incubator (Sanyo-Gallenkamp PLC, Leicester, UK). Cells were harvested and subcellular fractions were prepared as described (23). Fe 21-CO vs Fe 21 difference spectra were used to detect hemoprotein (25) in whole cells. Large-scale cultures were routinely done by inoculating 0.5 liters of expression medium with 5 ml of overnight culture. Supplementation of expression medium with 0.5 mM d-aminolevulinic acid slightly enhanced expression levels. Subsequent preparation of the membrane fraction was carried out as described elsewhere (23, 26, 27). Purification of recombinant CYP2A6 from E. coli. The recombinant CYP2A6 purification schedule was chosen according to approaches from the literature (23, 26, 27). Basically, the procedure used was very similar to that previously reported for CYP2E1 purification (27). In contrast to some other CYP purification procedures (27), no ligand was used to protect CYP from degradation (formation of P420) during purification. The CYP2A6 membrane fraction was solubilized using high concentrations of detergents sodium cholate and Triton N-101 (both 0.625%). Supernatant of solubilized membranes (105,000g) was then passed through a weak anion-exchanger DEAE-Sephacel. After Triton removal by Amberlite XAD-2, this fraction was applied to weak cation exchanger CM-Sepharose. CYP2A6 formed a brown layer at the top of the column. The potassium phosphate gradient (10 –200 mM) was used for removal of contaminants and elution of CYP2A6. Elution of CYP2A6 started at about 80 mM phosphate. Detergent was removed by application of diluted CYP2A6 preparation to a small hydroxylapatite column and extensive washing with 10 mM potassium phosphate buffer (pH 7.7). CYP2A6 protein was eluted from hydroxylapatite by a high concentration of phosphate and 0.1% sodium cholate. Excess phosphate and cholate was removed by dialysis against 50 mM potassium phosphate (pH 7.4) containing 20% glycerol and 1 mM EDTA. Purity of CYP2A6 preparation during purification was monitored by electrophoresis. The CYP2A6 concentration in the top hydroxylapatiteeluate fraction was 16 mM with a CYP specific content of 12.35 nmol CYP/mg protein.
ˇ EK PAVEL SOUC
192
TABLE I
PCR Primers Used to Introduce N-Terminal Modifications and Amino Acid Sequences of Constructs Amino acid sequence b Nucleotide sequence a
Construct
1 3
12
18
21
30
Forward primers (forming a NdeI site and modifications at 59-end) Native A B C D E F
59-agctgacatcatagtcttacatatgctggcctcaggg-39 59-ttaggacatatggctgcttcaggaatgttattagtggccttgctggtc-39 59-ttaggacatatggctctgttattagcagtttttctggtctgcctgact-39 59-catatggctctgttattagcagtttttctgatgtctgtttggcag-39 59-catatggctgttggcaacaacgtaaaagcaaagggaagctg-39 59-catatggtattaatgagtgtttggcaacaaaggaagagcaag-39 59-gacgtgatggtcttgcatatggctcgtcaagttcattcttcttggaatgggaagctgcctccg-39
MLASGMLLVALLVCLTVMVLMSVWQQRKSKGKL MAASGMLLVALLVCLTVMVLMSVWQQRKSKGKL M___ALLLAVFLVCLTVMVLMSVWQQRKSKGKL M___ALLLAVFL________MSVWQQRKSKGKL M_A___________________VWQQRKSKGKL _________________MVLMSVWQQRKSKGKL M_ARQVHSSWN___________________GKL
Universal reverse primer (forming a SalI site at 39-end) 59-catatgattcagatacgtcgactcagcggggcaggaa-39 a b
Start codon is underlined. Amino acid modifications are in bold.
Purification of other proteins. Rat NADPH-P450 reductase was purified from liver microsomes of phenobarbital-treated rats by the method of Yasukochi and Masters (28) with modifications (29). Assays of catalytic activity. The standard CYP2A6 reconstituted system contained 0.5 ml of the following mixture: 25 pmol of CYP2A6, 75 pmol of rat NADPH-P450 reductase, 30 mM L-a-dilauroyl-sn-3-phosphocholine, 50 mM potassium phosphate buffer (pH 7.4), substrate (20 mM coumarin unless otherwise noted), and NADPH-generating system (10 mM MgCl 2, 10 mM glucose 6-phosphate, 1 mM NADP, 0.5 U/mL glucose-6-phosphate dehydrogenase). Coumarin 7-hydroxylation was assayed using the HPLC method with fluorescence detection as described (30). In experiments with human liver microsomes, 100 pmol of total CYP was used in a total volume of 0.5 ml and the following CYP marker substrates were included: 20 mM coumarin (CYP2A6), 3 mM 7-ethoxyresorufin (CYP1A1/2), 2.5 mM tolbutamide (CYP2C9), 0.4 mM bufuralol (CYP2D6), 0.5 mM chlorzoxazone (CYP2E1), and 0.1 mM testosterone (CYP3A4). Incubations have been carried out in a shaking water bath at 37°C for 15 min except for the 7-ethoxyresorufin, tolbutamide, and bufuralol assay, where 30 min incubation was used. Potassium phosphate buffer (50 mM, pH 7.4) and the above-mentioned NADPH-generating system were used. In experiments with anti-CYP2A6 antibodies, microsomes were preincubated with 1–5 mg of IgG per nanomole of CYP for 15 min. After preincubation, the buffer, substrate, and NADPH-generating system were added. Metabolites were detected as described elsewhere (30 –35). Production of antibodies. Polyclonal anti-recombinant CYP2A6 antibodies were raised in five male chinchilla rabbits at 5– 6 months of age (VUFB, Konarovice, Czech Republic). Rabbit sera (1/200 dilution) were taken prior to immunization and subsequently used for preliminary screening by immunoblotting with human liver microsomes. Only rabbits with sera showing no reaction on the blot were selected. Rabbits were immunized on days 1, 14, and 42 with five 0.1-ml subcutaneous shots in their shaved backs. Each immunization was done with 0.5 nmol of purified CYP, diluted to 0.25 ml with 0.9% saline and further diluted to 0.5 ml with Freund’s complete (first immunization) or incomplete (second and third immunization) adjuvants. On day 56, rabbits were bled and IgG were prepared by affinity chromatography using protein A agarose gel according to the producer’s protocol (Pierce, Rockford, IL), followed by extensive dialysis against 100 vol of 50 mM potassium phosphate buffer (pH 7.4). Specificity of the raised antibodies was assessed using immunoblot-
ting (0.1 mg of IgG per blot used) with human liver microsomes and purified CYP standards (CYP 1A1, 1A2, 2A6, 2C9, 2D6, 2E1, 3A4, and 3A5). The inhibitory potency of anti 2A6 IgG was assayed by following the marker activities of various CYP enzymes in human liver microsomes (see assays of catalytic activity) in comparison with IgG taken prior to immunization. Other procedures and assays. Protein concentrations were estimated using the bicinchoninic acid method according to general methods provided by the supplier (Pierce). CYP spectra were recorded with a Specord M400 instrument (CarlZeiss, Jena, Germany) using the general methods of Omura and Sato (25) for quantitation of CYP content. NADPH cytochrome c reduction was determined as described elsewhere (28) using De 550 5 21 mM 21 cm 21 and an assumed specific activity of 3200 nmol of reduced cytochrome c min 21 nmol reductase 21 (28). SDS electrophoresis was done in 10% (w/v) polyacrylamide gels by the method of Laemmli (36) using a MiniProtean apparatus (BioRad, Hercules, CA). Protein staining was done with ammoniacal silver (37) and immunoblotting was performed as described (38) using the conditions for development of blots according to Soucˇek et al. (39). N-terminal amino acid sequencing was done in the Midwest Analytical (St. Louis, MO) facility using a Model 477A Protein Sequencer (Applied Biosystems, Foster City, CA) and methodology based on Edman’s degradation. SDS electrophoresis, transfer of protein to Immobilon-P membrane (Millipore Corp., Bedford, MA), and staining are described elsewhere (39). Blots were treated for 24 h by 0.6 N HCl at 25°C to remove the N-terminal formylmethionine block as described by Hirano et al. (40). Yields at each cycle were estimated by comparison with external standards. The Mfold and LoopViewer programs for MacIntosh computers were used to predict the energy of formation of stable secondary structure at the N-terminus. According to the results of computation, primers with favorable N-terminal sequences (usually with higher AT content) were designed for PCR.
RESULTS
Expression of N-terminal variants. Various N-terminal modifications of the CYP2A6 clone were designed (Table I) according to previous successful ap-
FIG. 1. Construction of pCW/2A6 expressing plasmid. Unique restriction sites are displayed. 193
194
ˇ EK PAVEL SOUC
FIG. 2. Spectrophotometric evaluation of CYP2A6 in whole E. coli cells. A, pCW; B, pCW/2A6native; C, pCW/2A6A. 1 ml of cells after 40 h expression was taken and centrifuged at 12,000g for 5 min. The pellet was resuspended in 3 ml of 100 mM potassium phosphate (pH 7.4) containing 10 mM dextrose and 6 mM magnesium acetate. Then baseline (. . .) and Fe 21.CO vs Fe 21 difference spectra (—) were recorded as described (25).
EXPRESSION OF CYTOCHROME P450 2A6 IN BACTERIA TABLE II
Purification of CYP2A6 from E. coli Membrane Fraction Purification step
Protein a (mg)
CYP (nmol)
Specific content (nmol CYP/mg protein)
Yield (%)
Cells Membranes c DEAE-Sephacel CM-Sepharose HA-Ultrogel
2b 415 87 8.9 3.9
210 178 141 88 48
2b 0.43 1.62 9.93 12.35
100 85 67 42 23
a All values correspond to material recovered from 1 liter of culture. b Protein content in the whole cells was not measured. c Values displayed for solubilized membranes (supernatant after 105,000g spin).
proaches (23, 27, 41). In all sequences the AT content was enhanced in order to reduce the potential for unfavorable mRNA secondary structure formation. Changes in codons were chosen in agreement with the list of E. coli preferred codons (42) where it was possible. Construct A was designed to substitute Leu at the second position by Ala as this N-terminal amino acid change was reported to facilitate expression in E. coli (43). The N-terminal sequence MALLAVFL previously reported to improve expression of various CYP proteins (23, 41, 44, 45) was introduced to constructs B and C. Constructs D and E contained truncated versions of the CYP2A6 native sequence. The sequence modification for construct F was derived from the N-terminal sequence of CYP2E1 successfully expressed in E. coli before (27). pCW/2A6 constructs bearing N-terminal modifications were then prepared (Fig. 1) and the expression level was monitored. As expected, the construct with the native sequence of CYP2A6 did not show detectable hemoprotein production as well as the control carrying pCW only (Figs. 2A and 2B). On the other hand, the construct designated pCW/2A6A, bearing only a single amino acid substitution, appeared to be the most effective in hemoprotein production as 210 nmol of CYP2A6 per liter of culture was detected (Fig. 2C, Table II). This variant of CYP2A6 was selected for further purification and characterization. Expression from the CYP2A6A variant was optimized with respect to temperature (32°C) and time (40 h) of cultivation. Stimulation (about 30% increase) of CYP2A6 production with the addition of 0.5 mM d-aminolevulinic acid was observed. Purification of CYP2A6. The three-step column chromatography employing DEAE-Sephacel, CMSepharose, and hydroxylapatite was used. About 80% of the initial CYP2A6 load was recovered in the DEAE void fraction and a 40% loss of hemoprotein in CMeluate was observed when compared to the DEAE void fraction (Table II). The CYP2A6 concentration in the
195
top hydroxylapatite-eluate fraction was 16 mM with a CYP-specific content of 12.35 nmol CYP/mg protein (Table II). The CYP2A6 preparation was pure as judged by electrophoresis (Fig. 3). Spectral properties of CYP2A6. Absolute spectra of purified recombinant CYP2A6 were recorded (Fig. 4) and the wavelength maxima and extinction coefficients were calculated (Table III). Oxidized CYP2A6 was in a low-spin state as indicated by the peak at 418 nm (Fig. 4A). Addition of the substrate coumarin shifted the spin state of oxidized CYP2A6 from low to high spin, indicating the binding of substrate to the CYP active site (see second derivative spectra, Fig. 4B). Transition of the spin state in the presence of coumarin proceeded in a concentration-dependent manner. In contrast to CYP1A2 (41), CYP2A6 was readily reduced by sodium dithionite. Cytochrome P420 was absent in the CYP2A6 preparation as judged from Fe 21.CO spectra (Fig. 4A). N-terminal sequence analysis of recombinant CYP2A6. N-terminal amino acid analysis showed that the N-terminus of purified CYP2A6 was blocked most probably by N-formylmethionine. This block was removed by treatment of the blot strip containing 20 pmol of purified CYP2A6 with 0.6 N HCl for 24 h at 25°C. The expected N-terminal sequence was then obtained. The sequence was clearly readable, although yields appear to be fairly low (Table IV). Catalytic activities of recombinant CYP2A6. Coumarin 7-hydroxylation was examined as a marker activity of human CYP2A6. Initial experiments were de-
FIG. 3. SDS–polyacrylamide electrophoresis of CYP2A6 fractions obtained during purification. An amount of sample corresponding to 5 pmol of CYP was applied per lane. The direction of migration was from top to bottom. Lanes included: 1, membrane fraction; 2, solubilized membranes (105,000g supernatant fraction), 3, DEAE-Sephacel void fraction; 4, CM-Sepharose gradient eluate fraction; 5, best fraction eluted from hydroxylapatite.
196
ˇ EK PAVEL SOUC
FIG. 4. Spectral properties of CYP2A6. (A) Absolute UV-visible spectra of purified recombinant CYP2A6 : Fe 31 (), Fe 21(ª), Fe 21.CO (. . .). (B) Second derivative spectra of purified recombinant CYP2A6 in the absence (. . .) and in the presence of 2 mM (ª) or 20 mM () coumarin. All spectra were recorded with 4.3 mM (A) or 2.4 mM (B) CYP2A6 in 100 mM potassium phosphate (7.4) containing 20% glycerol (v/v) and 1 mM EDTA. Reduction was accomplished by addition of a few milligrams of solid sodium dithionite. The inset shows expanded a and b Soret regions.
signed in order to optimize the composition of the CYP2A6 reconstituted system. In the reconstituted system containing 25 pmol of purified recombinant CYP2A6 and components described under Materials and Methods, the amount of rat NADPH-P450 reductase varied from 6.25 to 250 pmol (CYP/rat NADPHP450 reductase ratio, 4/1 to 1/10). Maximum activity was found when a 1/5 ratio of CYP/rat NADPH-P450 reductase was present (Fig. 5). The activity at 1/3 ratio (4.81 nmol product/min/nmol CYP) was found to be reasonably high, and due to the limited supply of rat NADPH-P450 reductase this ratio was used for subsequent studies. It was published that addition of some chemicals enhances the catalytic activity of CYP in the
reconstituted system (23). Therefore, it was of interest whether effects of these components are CYP-isoenzyme specific or whether it is a general observation. The effect of cytochrome b 5 in a CYP2A6-catalyzed reaction was analyzed as well. From the data presented in Table V it is evident that cytochrome b 5 had a great stimulatory effect on CYP2A6 catalysis. The highest activity was observed at a 1/1 ratio of CYP2A6/ cytochrome b 5 (more than a twofold increase). Addition of 1–5 mM glutathione enhanced CYP2A6 activity up to 50%. The presence of the detergent sodium cholate, variations in the buffer system, or concentrations of MgCl 2 up to 30 mM did not significantly influence the activity of CYP2A6. On the contrary, a higher concen-
EXPRESSION OF CYTOCHROME P450 2A6 IN BACTERIA
197
TABLE III
Spectral Properties of Purified Recombinant Human CYP2A6
Form of CYP2A6 a
Wavelength maximum (nm)
Extinction coefficient (e, mM 21 cm 21)
418 534 568 388 410 544 447 555
131 11.4 10.8 98 113 15.3 127 14.3
Fe 31 Fe 31-coumarin Fe 21 Fe 21.CO Fe 21.CO vs Fe 21 (difference)
91.0 b
448
a
Spectra were recorded in 100 mM potassium phosphate buffer (pH 7.4) containing 1.0 mM EDTA, 20% glycerol (v/v), and 4.3 mM purified CYP2A6. For substrate spectra 20 mM coumarin and 2.4 mM CYP2A6 were used. b Assuming Omura and Sato’s calculations (25).
tration of MgCl 2 strongly inhibited 7-hydroxylation of coumarin (Table V). Analysis of enzymatic kinetics of coumarin 7-hydroxylation in the CYP2A6 reconstituted system revealed typical Michaelis–Menten kinetics (Fig. 6; K m 5 1.48 6 0.37 mM, V max 5 3.36 6 0.18 nmol product/min/nmol CYP). The CYP2A6 reconstituted system was used to investigate the role of CYP2A6 in oxidation of chlorzoxazone, the purported probe of CYP2E1 in vivo. In the standard system described under Materials and Methods, it was found that CYP2A6 (0.05 mM) metabolized 0.5 mM chlorozoxazone to 6-hydroxychlorzoxazone at turnover rate 0.32 nmol product/min/nmol CYP. Under
TABLE IV
N-Terminal Amino Acid Sequence Analysis of Recombinant CYP2A6 Purified from E. coli Amino acid Cycle/position
Expected
Found
Recovered a (pmol)
1 2 3 4 5 6 7 8 9 10
M A A S G M L L V A
M A A S G M L L V A
2.1 2.1 2.1 0.9 1.6 1.5 1.8 2.3 1.5 1.1
a
A nominal amount of 20 pmol of purified CYP2A6 was used for SDS electrophoresis and blotting transfer.
FIG. 5. Coumarin 7-hydroxylation activity of purified recombinant CYP2A6 in the reconstituted system— dependence on rat NADPHP450 reductase amount. The CYP2A6 reconstituted system contained 0.5 ml of the following mixture: 25 pmol of CYP2A6, 6.25, 12.5, 25, 50, 75, 150, or 250 pmol of rat NADPH-P450 reductase, 30 mM L-a-dilauroyl-sn-3-phosphocholine, 50 mM potassium phosphate buffer (pH 7.4), 20 mM coumarin, and the NADPH-generating system (10 mM MgCl 2, 10 mM glucose 6-phosphate, 1 mM NADP, 0.5 U/ml glucose-6-phosphate dehydrogenase). Incubation proceeded for 15 min and metabolite was determined as described under Materials and Methods. All assays were performed in duplicates with variations less than 5%.
the same conditions, chlorzoxazone was metabolized by human liver microsomes (0.2 mM total CYP) at a turnover rate of 0.94 nmol product/min/nmol CYP. Production of polyclonal antibodies against recombinant CYP2A6. Polyclonal antibodies against purified recombinant CYP2A6 have been raised in rabbits. These antibodies proved to be very specific and useful for immunoblotting as they recognize CYP2A6 well (Fig. 7A). Very weak crossreactivity with purified recombinant CYP2C9 and 2E1 (but not with 1A1, 1A2, 2D6, 3A4, and 3A5) was also observed. Crossreactive CYPs migrate with lower mobility than CYP2A6 and therefore may be easily distinguished from the CYP2A6-corresponding band found in human liver microsomes where a minor crossreactive band of higher M r was also detected (Fig. 7B). Anti-CYP2A6 antibodies very specifically and efficiently inhibited CYP2A6 catalytic activity toward coumarin in human liver microsomes (Fig. 8). Significant inhibition was achieved by 1 mg of anti-CYP2A6 IgG/nmol of total CYP. On the other hand, up to 5 mg of anti-2A6 IgG/nmol of total CYP did not significantly inhibit other human CYP (1A, 2C9, 2D6, 2E1, and 3A4) marker activities. Inhibition of coumarin 7-hydroxylation in human liver microsomes by rabbit IgG taken prior to immunization also showed negligible effect (Fig. 8).
ˇ EK PAVEL SOUC
198 TABLE V
Coumarin 7-Hydroxylation in Human Liver Microsomes and the CYP2A6 Reconstituted System—Influence of Cytochrome b 5, Buffers, Salts, Glutathione, and Detergent
Enzyme preparation Human liver microsomes (n 5 10) a Purified recombinant CYP2A6: No addition b 1b 5 - b 5/CYP ratio 1/1 1b 5 - b 5/CYP ratio 2/1 1b 5 - b 5/CYP ratio 4/1 1150 mM KCl/50 mM Tris (pH 7.4) 150 mM Hepes (pH 7.4) 1Glutathione, 1 mM 1Glutathione, 3 mM 1Glutathione, 5 mM 1MgCl 2, 30 mM 1MgCl 2, 60 mM 1Sodium cholate, 50 mg 1Sodium cholate, 100 mg 1Sodium cholate, 200 mg
7-Hydroxylation of coumarin (nmol product/ min/nmol CYP) 1.79 6 0.69 4.06 10.15 8.18 6.92 4.20 4.36 4.27 5.51 6.23 4.08 2.64 3.05 4.60 4.07
Note. For 7-hydroxycoumarin determination see Materials and Methods. All assays were performed in duplicates with less than 5% variation. a Human liver microsomes were incubated 15 min at 37°C with 20 mM coumarin and the NADPH-generating system (10 mM MgCl 2, 10 mM glucose 6-phosphate, 1 mM NADP, 0.5 U/ml glucose-6-phosphate dehydrogenase) in a 0.5-ml total volume of 50 mM potassium phosphate buffer (pH 7.4). b The reconstituted system contained 25 pmol CYP2A6, 75 pmol rat NADPH-P450 reductase, 30 mM L-a-dilauroyl-sn-3-phosphocholine, 50 mM potassium phosphate (pH 7.4), 20 mM coumarin, and the NADPH-generating system (see above) in total volume 0.5 ml. Incubation proceeded for 15 min at 37°C.
Purification of the CYP2A6 protein proceeded quite straitforwardly, contrary to reported difficulties with CYP2D6 and 3A4 proteins (23, 46), and it seems that there is no need to facilitate purification by tagging the protein with oligo-His sequences or other modifications of protein. Spectral properties of purified CYP2A6 protein were not examined so far. Absolute spectra of CYP2A6 presented here appear to give typical results previously seen with other CYP proteins and showed high similarity particularly with CYP2D6 expressed in E. coli (46). Contrary to the majority of other E. coli-expressed CYPs (23, 26, 27, 41, 45, 47), CYP2A6 was isolated in the low-spin state, similarly to CYP2D6 (46). Substrate coumarin shifted the low-spin state of oxidized CYP2A6 to high spin, indicating the binding to the active site of CYP2A6. N-terminal amino acid sequence analysis revealed a N-formylmethionine blockade. It seems that this block is related to the high hydrophobicity of the CYP2A6 N-terminus as it was observed in several E. coli-expressed CYP proteins, all bearing highly hydrophobic MALLAVFL sequence modification (48). Using Kyte and Doolittle’s hydropathy plot (49) it was found that the CYP2A6 N-terminus has also a very high degree of hydrophobicity (results not shown). N-Formylmethionine was removed by mild acid treatment and the amino acid sequence analysis confirmed that the ex-
DISCUSSION
Although CYP2A6 has been considered to be a minor CYP form present in human liver, the amount of information about this particular CYP isoform in the literature is increasing in the last few years. The gene encoding this CYP enzyme has been shown to be present in at least three allelic variants (12, 13). Despite the rising interest in CYP2A6, the isolation of this protein from a heterologous system and the production of antibodies against it have not been reported. Moreover, the E. coli expressing system could also contribute to the study of activity and affinity of CYP2A6 allelic variants. Among the six variants of CYP2A6 examined here, the N-terminal variant with only a single amino acid change (Ala instead of Leu at the second position) showed the best expression efficiency. As in the CYP2D6 protein (46), supplementation of expression medium with d-aminolevulinic acid proved to enhance expression levels of CYP2A6.
FIG. 6. Enzyme kinetics of coumarin 7-hydroxylation activity of purified recombinant CYP2A6 in the reconstituted system. The CYP2A6 reconstituted system contained 0.5 ml of the following mixture: 25 pmol of CYP2A6, 75 pmol of rat NADPH-P450 reductase, 30 mM L-a-dilauroyl-sn-3-phosphocholine, 50 mM potassium phosphate buffer (pH 7.4), 0.2, 0.5, 1, 2, 5, 10, 20, or 50 mM coumarin, and the NADPH-generating system (10 mM MgCl 2, 10 mM glucose 6-phosphate, 1 mM NADP, 0.5 U/ml glucose-6-phosphate dehydrogenase). Incubation proceeded for 10 min and metabolite was determined as described under Materials and Methods. All assays were performed in duplicates with variations less than 5%. K m 5 1.48 6 0.37 mM, V max 5 3.36 6 0.18 nmol product/min/nmol CYP.
EXPRESSION OF CYTOCHROME P450 2A6 IN BACTERIA
pressed CYP2A6A variant had the desired N-terminal sequence. Recombinant CYP2A6 expressed and purified in this study was shown to be catalytically active toward the widely used marker substrate coumarin with a K m (1.48 mM) comparable to values found in human liver microsomes (7, 11). The presence of an equal amount (to CYP) of cytochrome b 5 greatly stimulated 7-hydroxylation of coumarin similarly as it was noted for E. coli-expressed CYP’s 2C9, 2E1, 3A4, and 3A5 (23, 26, 27, 45). CYP2A6 activity was also enhanced by the addition of glutathione (up to 5 mM). Stimulation of catalytic activity by glutatione was reported for recombinant CYP3A4 (23). On the other hand, neither the change in buffer composition nor the change in the concentration of divalent metal ions or cholate stimulates CYP2A6 activity, in contrast to data on CYP3A4 but in agreement with data on CYP2C9 (50). Moreover, it was shown that CYP2A6 in the reconstituted system metabolizes chlorzoxazone at a more
199
FIG. 8. Immunoinhibition of CYP marker activities in human liver microsomes by anti-CYP2A6 IgG. Incubations were carried out as described under Materials and Methods. Control (uninhibited) CYP activities in nmol product/min/nmol CYP were: 0.12, 7-ethoxyresorufin O-deethylation (*); 2.20, coumarin 7-hydroxylation (E); 0.39, tolbutamide methylhydroxylation (‚); 0.33, bufuralol 19-hydroxylation ( ); 0.94, chlorzoxazone 6-hydroxylation (h); 2.33, testosterone 6bhydroxylation (3). Rabbit IgG isolated from the sera taken prior to immunization ({) were used in amounts of 1, 3, or 5 mg IgG/nmol CYP. All assays were performed in duplicates with variations less than 5%.
than notable rate and thus it should be taken into consideration when in vivo phenotyping of human CYP2E1 is performed. From data presented here it seems that it would be very interesting to follow the activity of the reconstituted system with added cytochrome b 5, glutathione, and sodium cholate together or to compare enzyme kinetics data with the experiment done in the presence of cytochrome b 5, where a change of kinetic parameters may be expected similarly as it was observed for recombinant CYP2E1 (27). The catalytic activity of CYP2A6 toward chlorzoxazone in the reconstituted system may be much higher in the presence of cytochrome b 5 as well. These experiments are in progress now. Polyclonal antibodies raised against recombinant CYP2A6 showed very good specificity in immunoblotting and immunoinhibition and may serve as an important tool in future studies of the role of CYP2A6 in the metabolism of chemicals and in procarcinogen activation. FIG. 7. Immunoblotting of CYP proteins and human liver microsomes with anti-CYP2A6 IgG. (A) 1 pmol of purified recombinant CYP was used per lane: 1, CYP 1A1; 2, CYP1A2; 3, CYP2A6; 4, CYP2C9; 5, CYP2D6; 6, CYP2E1; 7, CYP3A4; 8, CYP3A5. (B) Lane 1, 1 pmol of purified CYP2A6; lanes 2–9, 20 mg of protein from eight individual samples of human liver microsomes was used per lane. Blots were developed using 0.1 mg of anti-2A6 IgG as described under Materials and Methods.
ACKNOWLEDGMENTS The author expresses his sincere thanks to Dr. Guengerich for his generous gifts (CYP’s, pCW/NF14, and methylhydroxy tolbutamide) and essential expertize, Dr. Gonzalez for pUC2A6, Dr. Yamazaki for human cytochrome b 5, and Drs. Pavel Anzenbacher, Hermann Esselmann, Elizabeth Gillam, and David McCourt for helpful discussions during this project.
200
ˇ EK PAVEL SOUC
REFERENCES 1. Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J., Feyereisen, R., Waxman, D. J., Waterman, M. R., Gotoh, O., Coon, M. J., Estabrook, R. W., Gunsalus, I. C., and Nebert, D. W. (1996) Pharmacogenetics 6, 1– 42. 2. Guengerich, F. P. (1995) in Cytochrome P450: Structure, Mechanism, and Biochemistry (Ortiz de Montellano, P. R., Ed.), pp. 473–535, Plenum, New York. 3. Sawada, M., and Kamataki, T. (1998) Mutat. Res. 411, 19 – 43. 4. Clarke, S. E. (1998) Xenobiotica 28, 1167–1202. 5. Dong, M. S., Yamazaki, H., Guo, Z., and Guengerich, F. P. (1996) Arch. Biochem. Biophys. 327, 11–19. 6. Waterman, M. R., Jenkins, C. M., and Pikuleva, I. (1995) Toxicol. Lett. 82/83, 807– 813. 7. Yamano, S., Tatsuno, J., and Gonzalez, F. J. (1990) Biochemistry 29, 1322–1329. 8. Miles, J. S., McLaren, A. W., Forrester, L. M., Glancey, M. J., Lang, M. A., and Wolf, C. R. (1990) Biochem. J. 267, 365–371. 9. Yun, C. H., Shimada, T., and Guengerich, F. P. (1991) Mol. Pharmacol. 40, 679 – 685. 10. Shimada, T., Yamazaki, H., Mimura, M., Inui, Y., and Guengerich, F. P. (1994) J. Pharm. Exp. Ther. 270, 414 – 423. 11. Shimada, T., Yamazaki, H., and Guengerich, F. P. (1996) Xenobiotica 26, 395– 403. 12. Oscarson, M., Gullsten, H., Rautio, A., Bernal, M. L., Sinues, B., Dahl, J. H., Stengard, J. H., Pelkonen, O., Raunion, H., and Ingelman-Sundberg, M. (1998) FEBS Lett. 438, 201–205. 13. Fernandez-Salguero, P., Hoffman, S. M. G., Cholerton, S., Mohrenweiser, H., Raunio, H., Raution, A., Pelkonen, O., Huang, J., Evans, W. E., Idle, J. R., and Gonzalez, F. J. (1995) Am. J. Hum. Genet. 57, 651– 660. 14. Imaoka, S., Yamada, T., Hiroi, T., Hayashi, K., Sakaki, T., Yabusaki, Y., and Funae, Y. (1996) Biochem. Pharmacol. 51, 1041–1050. 15. Spracklin, D. K., Thummel, K. E., and Kharasch, E. D. (1996) Drug Metab. Dispos. 24, 976 –983. 16. Nakajima, M., Yamamoto, T., Nunoya, K-I., Yokoi, T., Nagashima, K., Inoue, K., Funae, Y., Shimada, N., Kamataki, T., and Kuroiwa, Y. (1996) Drug Metab. Dispos. 24, 1212–1217. 17. Aoyama, T., Shigeru, Y., Guzelian, P. S., Gelboin, H. V., and Gonzalez, F. J. (1990) Proc. Natl. Acad. Sci. USA 87, 4790 – 4793. 18. Duescher, R. J., and Elfarra, A. A. (1994) Arch. Biochem. Biophys. 311, 342–349. 19. Crespi, C. L., Penman, B. W., Leakey, J. A., Arlotto, M. P., Stark, A., Parkinson, A., Turner, T., Steimel, D. T., Rudo, K., Davies, R. L., and Langenbach, R. (1990) Carcinogenesis 11, 1293–1300. 20. Yamazaki, H., Inui, Y., Yun, C-H., Guengerich, F. P., and Shimada, T. (1992) Carcinogenesis 13, 1789 –1794. 21. Pritchard, M. P., Ossetian, R., Li, D. N., Henderson, C. J., Burchell, B., Wolf, C. R., and Friedberg, T. (1997) Arch. Biochem. Biophys. 345, 342–354. 22. van der Hoeven, T. A., and Coon, M. J. (1974) J. Biol. Chem. 249, 6302– 6310. 23. Gillam, E. M. J., Baba, T., Kim, B-R., Ohmori, S., and Guengerich, F. P. (1993) Arch. Biochem. Biophys. 305, 123–131. 24. Bauer, S., and Shiloach, J. (1974) Biotechnol. Bioeng. 16, 933– 941.
25. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239, 2370 –2378. 26. Sandhu, P., Baba, T., and Guengerich, F. P. (1993) Arch. Biochem. Biophys. 306, 443– 450. 27. Gillam, E. M. J., Guo, Z., and Guengerich, F. P. (1994) Arch. Biochem. Biophys. 312, 59 – 66. 28. Yasukochi, Y., and Masters, B. S. S. (1976) J. Biol. Chem. 251, 5337–5344. 29. Guengerich, F. P. (1994) in Principles and methods of toxicology (Hayes, A. W., Ed.), pp. 1259 –1313, Raven, New York. 30. Soucˇek, P. (1999) A novel sensitive HPLC method for assay of coumarin 7-hydroxylation. J. Chromatogr., in press. 31. Lubet, R. A., Mayer, R. T., Cameron, J. W., Nims, R. W., Burke, M. D., Wolf, T., and Guengerich, F. P. (1985) Arch. Biochem. Biophys. 283, 43– 48. 32. Knodell, R. G., Hall, S. D., Wilkinson, G. R., and Guengerich, F. P. (1987) J. Pharm. Exp. Ther. 241, 1112–1119. 33. Yamazaki, H., Guo, Z., Persmark, M., Mimura, M., Inoue, K., Guengerich, F. P., and Shimada, T. (1994) Mol. Pharmacol. 16, 568 –577. 34. Peter, R., Bocker, R., Beaune, P. H., Iwasaki, M., Guengerich, F. P., and Yang, C. S. (1990) Chem. Res. Toxicol. 3, 566 –573. 35. Brian, W. R., Sari, M. A., Iwasaki, M., Shimada, T., Kaminsky, L. S., and Guengerich, F. P. (1990) Biochemistry 29, 11280 – 11292. 36. Laemmli, U. K. (1970) Nature 227, 680 – 685. 37. Wray, W., Boulikas, T., Wray, V. P., and Hancock, R. (1981) Anal. Biochem. 118, 197–203. 38. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350 – 4356. 39. Soucˇek, P., Martin, M. V., Ueng, Y. F., and Guengerich, F. P. (1995) Biochemistry 34, 16013–16021. 40. Hirano, H., Komatsu, S., Kajiwara, H., Takagi, Y., and Tsunasawa, S. (1993) Electrophoresis 14, 839 – 846. 41. Sandhu, P., Guo, Z., Baba, T., Martin, M. V., Tukey, R. H., and Guengerich, F. P. (1994) Arch. Biochem. Biophys. 309, 168 –177. 42. (1987) in Current Protocols in Molecular Biology (Ausubel, F. M., et al., Eds.), Greene, PA; Wiley, Brooklyn, NY. 43. Looman, A. C., Bodlaender, J., Comstock, J. L., Eaton, D., Jhurani, P., deBoer, H. A., and Knippenberg, P. H. (1987) EMBO J. 6, 2489 –2492. 44. Barnes, H. J., Arlotto, M. P., and Waterman, M. R. (1991) Proc. Natl. Acad. Sci. USA 88, 5597–5601. 45. Gillam, E. M. J., Guo, Z., Ueng, Y-F., Yamazaki, H., Cock, I., Reilly, P. E. B., Hooper, W. D., and Guengerich, F. P. (1995) Arch. Biochem. Biophys. 317, 374 –384. 46. Gillam, E. M. J., Guo, Z., Martin, M. V., Jenkins, C. M., and Guengerich, F. P. (1995) Arch. Biochem. Biophys. 319, 540 –550. 47. Guo, Z., Gillam, E. M. J., Ohmori, S., Tukey, R. H., and Guengerich, F. P. (1994) Arch. Biochem. Biophys. 312, 436 – 446. 48. Dong, M-S., Bell, C. L., Guo, Z., Phillips, D. R., Blair, I. A., and Guengerich, F. P. (1996) Biochemistry 35, 10031–10040. 49. Kyte, J., and Doolittle, R. F. (1982) J. Mol. Biol. 157, 105–132. 50. Yamazaki, H., Gillam, E. M. J., Dong, M-S., Johnson, W., Guengerich, F. P., and Shimada, T. (1997) Arch. Biochem. Biophys. 342, 3329 –3337.