The cytochrome P450 (P450) enzymes, a family of heme-containing monooxygenases, play a central role in the metabolism of a variety of drugs, fatty acids, carcinogens, pesticides, and steroids. The P450 enzymes exhibit partially overlapping but distinct substrate specificities (Porter and Coon, 1991; Nelson et al., 1996). It is of great interest to elucidate the structure and function of these enzymes because of their potential uses in industrial and drug discovery processes. The P450 2B enzymes have been extensively studied using homology modeling based on the bacterial P450 and P450 2C5 crystal structures, site-directed mutagenesis, as well as susceptibility to inhibition by mechanism-based inactivators (Szklarz et al., 1995; He et al., 1996; Lewis and Lake, 1997; Dai et al., 1998; Domanski and Halpert, 2001; Kent et al., 2001; Spatzenegger et al., 2001; Wang and Halpert, 2002).

In recent studies using site-directed mutagenesis to elucidate the Tyr residue responsible for the inactivation of P450 2B1 after exposure to peroxynitrite, we demonstrated that Tyr-203, located in the F helix, plays an important role in determining the stereoselectivity for testosterone hydroxylation (Lin et al., 2003). Halpert and coworkers have demonstrated that two other F helix residues, Phe-206 and Leu-209, determine the substrate specificity as well as the regio- and stereoselectivity of P450 2B1 (He et al., 1994; Szklarz et al., 1995). Similar studies with Phe-209 in P450 2A5 have suggested that this F helix residue plays a critical role in determining substrate and product specificity and that the region around residue 209 constitutes the heme-substrate pocket in mammalian P450s (Lindberg and Negishi, 1989; Juvonen et al., 1991). With P450s 3A4, 11A1, and 27A1, residues in the F helix have been found to be critical for controlling the regioselectivity of substrate oxidation (Pikuleva et al., 2001; Xue et al., 2001). All of these residues are located in the putative substrate recognition site 2 for the P450 2 family as defined by Gotoh (1992). The F helix and the F-G loop are thought to comprise part of the substrate pocket or the substrate access channel in bacterial and mammalian P450s (Graham-Lorence et al., 1995; Hasemann et al., 1995; Dai et al., 1998; Williams et al., 2000). Recently, the crystal structure of the P450 BM3-substrate complex has suggested that the “lid domain” of the substrate access channel, consisting of the F and G helices and the loop between them, is involved in a clam shell-like movement to trap substrate and exhibits a rocking motion with the I helix as a fulcrum. The movements in the substrate-docking region seem to position the substrate in the active site for the catalysis (Haines et al., 2001). In the course of a previous study involving conversion of Tyr-203 to Ala, we inadvertently generated a double mutant in which Tyr-203 was converted to Ala and Thr-205 to Arg (Lin et al., 2003). This double mutant displayed a transient reduced CO spectrum together with the generation of a peak at 422 nm and was devoid of catalytic activity. The single mutant involving conversion of Tyr-203 to Ala exhibited higher levels of catalytic activity. We were interested in determining whether the lack of activity of the double mutant was a result of an Arg substitution at position 205. These observations also prompted us to study the F helix residue Thr-205 in greater detail to examine the structural and functional role of the F helix in P450 2B1. Thus, in addition to mutating Thr-205 to Arg, the side chain of Thr was modified by conversion of the 1) -CH₃ to -H for the Ser-mutant; 2) -OH to -H for the Ala-mutant; and 3) -OH to -CH₃ for the Val-mutant to evaluate the functional and structural roles of Thr-205 in P450 2B1. The mutated P450s and the wild-type P450 2B1 (WT) were expressed in Escherichia coli and purified. The reduced CO complexes; substrate-induced spectral changes; and catalytic activities toward 7-ethoxy-4-(trifluoromethyl)coumarin (EFC), benzphetamine, testosterone, and androstenedione were characterized. Two potent mechanism-based inactivators of P450 2B1, 2-ethynylnaphthalene (2EN) and 17α-ethynylestradiol (17EE), were used to assess the susceptibility of these mutated P450s to inactivation (Roberts et al., 1993; Kent et al., 2002).

Our results indicate that the 16β-hydroxylation activity for testosterone and androstenedione, a specific marker for P450 2B (Waxman et al., 1983; Wood et al., 1983), was dramatically suppressed in the Val- and Ala-mutants. The susceptibility to inactivation by 2EN and 17EE was also markedly altered in the Val-mutant. These results conclusively demonstrate that the amino acid at position 205 in P450 2B1 contributes to both substrate and product selectivity.

Materials and Methods

Materials. Benzphetamine, NADPH, testosterone, 16α-hydroxytestosterone, 16β-hydroxytestosterone, androstenedione, 16α-hydroxyandrostenedione, l-α-dilauroyl-phosphatidylcholine (DLPC) and 17EE were from Sigma-Aldrich (St. Louis, MO). EFC was from Molecular Probes (Eugene, OR) and 7-hydroxy-(trifluoromethyl)coumarin was from Enzyme System Products (Dublin, CA). Rabbit polyclonal antibody to P450 2B1 was prepared as described previously (Shen et al., 1991). 2EN was synthesized as described previously (Roberts et al., 1993).

Construction of Vectors Used for Protein Expression. Plasmid pCW2B1 was used as a template to construct four mutants at position 205 (Hanna et al., 1998). Mutations were carried out using an in vitro QuickChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). The primer 5′-GAGCTGTTTCTACCGGTCCTTTT CCCTCCTAAG-3′ was used for the Thr to Ser conversion. The primer 5′-GAGCTGTTTCTACCGGGCCTTTTCCCTCCTAAG-3′ was used for the Thr to Ala conversion. The primer 5′-GAGCTGTTTCTACCGGGTCTTTTCCCTCCTAAG-3′ was used for the Thr to Val conversion. The primer 5′-GAGCTGTTTCTACCGGAGGTTTTCCCTCCTAAG-3′ was used for the Thr to Arg conversion. The mutations were confirmed by DNA sequencing carried out at University of Michigan Biomedical Core Facility (Ann Arbor, MI).

Purification of Enzymes. WT P450 2B1 as well as the Arg-, Ser-, Val-, and Ala-mutants were expressed in Escherichia coli MV1304 and NADPH-cytochrome P450 reductase (reductase) was expressed in E. coli Topp3. All the enzymes were purified according to methods described previously (Hanna et al., 1998).

Spectral Analysis. Total P450 concentrations were determined from reduced CO difference spectra (Omura and Sato, 1964). Equal amounts of WT and mutant P450s were used for all the reactions. The substrate-induced spectral changes were performed by addition of 300 μM benzphetamine or 10 μM n-octylamine to 250 pmol of P450 in 50 mM Tris buffer (pH 7.4) containing 20% glycerol and 150 mM KCl (Schenkman et al., 1981). Difference spectra were recorded between 350 and 500 nm using a 3000 spectrophotometer (Milton Roy Company, Rochester, NY).

Determination of Enzymatic Activities. The catalytic activity of each P450 with all the substrates was assessed using the reconstituted system containing 25 pmol of P450, 50 pmol of reductase, and 10 μg of DLPC with preincubation at 22°C for 30 min.

The EFC O-deethylation activities of the P450s were measured as described previously (Buters et al., 1993). Assays were performed at 30°C in 1 ml of assay buffer containing 100 mM potassium phosphate buffer (pH 7.7), 0.2 mM NADPH, and 0.1 mM EFC as substrate. Reactions were quenched after 7 min by the addition of 0.3 ml of acetonitrile. The product, 7-hydroxy-(trifluoromethyl)coumarin, was detected by fluorescence (excitation 410 nm, emission 510 nm) using a SLM Aminco spectrofluorometer.

Benzphetamine N-demethylation assays were conducted at 37°C for 10 min in 0.5 ml of assay buffer containing 1 mM benzphetamine as substrate. The amount of formaldehyde generated was determined fluorometrically (excitation 410 nm, emission 510 nm) as described previously (de Andrade et al., 1996). An SLM Aminco spectrofluorometer was used to detect the fluorescent product.

The metabolism of testosterone and androstenedione was determined as described previously (Waxman et al., 1983; Wood et al., 1983). The reaction mixtures were incubated at 37°C for 20 min in 1 ml of assay buffer with 0.2 mM testosterone or 0.2 mM androstenedione as substrate. The reactions were terminated by addition of 2 ml of ethyl acetate, the metabolites were extracted from the organic phase, and dried under N₂. The dried products were dissolved in 65% methanol and resolved using a Microsorb-MV C₁₈ reversed phase column (5 μm, 4.6 × 150 mm; Varian, Walnut Creek, CA). Testosterone metabolites were separated isocratically using 65% methanol at a flow rate of 0.85 ml/min. Androstenedione metabolites were separated isocratically using 58% methanol for 16 min followed by a linear gradient to 70% methanol for 12 min. The eluates were monitored by UV detection at 254 nm. The major metabolites for testosterone are 16α-hydroxytestosterone, 16β-hydroxytestosterone, and androstenedione. The major metabolites for androstenedione are 16α-hydroxyandrostenedione and 16β-hydroxyandrostenedione.

Western Blotting Analysis. The P450s (2 pmol) were subjected to SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. The membranes were then incubated with rabbit polyclonal anti-P450 2B1 antibody, probed with goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (Bio-Rad, Hercules, CA), and the immunoreactive bands were detected using SuperSignal West Pico chemiluminescent substrate (Pierce Chemical, Rockford, IL). The blot was exposed to autoradiographic film and photographed.

CO Difference Spectrum of P450 and Reductase Complex. P450 (250 pmol) and reductase (250 pmol) were reconstituted for 30 min and diluted with 1 ml of 100 mM potassium phosphate buffer (pH 7.7), containing 0.5 mM NADPH and 300 μM benzphetamine. The mixture was bubbled with CO for 2 min and the enzymatically reduced CO difference spectrum was scanned until a steady state was attained. A trace of sodium dithionite was added and an additional scan was performed.

Mechanism-Based Inactivation. The primary reaction mixtures contained 150 pmol of P450, 300 pmol of reductase, 40 μg of DLPC, 100 units of catalase, and 10 μM 2EN or 50 μM 17EE in 200 μl of 100 mM potassium phosphate buffer (pH 7.7). After incubating the primary mixture in the absence (100% activity) or the presence of 1 mM NADPH at 30°C for 15 min, a 20-μl aliquot was removed and added to 1 ml of a secondary reaction mixture for the determination of EFC deethylation activity as described above.

Modeling of the P450 2B1 Structure and Conformers of Testosterone and Benzphetamine. Homology modeling of the three-dimensional structure of P450 2B1 was performed using P450 2C5 as a template as described previously (Williams et al., 2000; Lin et al., 2003). Coordinates for testosterone and benzphetamine were constructed using CS Chem3D Pro software (Cambridge Software Corp., Cambridge, MA). Stable conformers of testosterone and benzphetamine were obtained by minimizing internal energies on the basis of calculations using the MOPAC PM3 potential function.

Data Analysis. Results are given as the mean ± S.D., and the statistical evaluations are based on the unpaired, two-tailed Student's t test.

Results

Spectral Analysis. The concentrations of the purified WT and mutated P450s were determined from the dithionite-reduced CO difference spectra. All of the P450s exhibited maximal absorptions at 450 nm, and the reduced CO complexes were stable for up to 10 min with the exception of the Arg-mutant. The Arg-mutant displayed a reduced CO complex with a peak initially at 450 nm but that was rapidly converted to a denatured form with a peak at 422 nm within 0.5 min after adding dithionite (Fig. 1, spectrum a). The spectra of the reduced CO complexes recorded 3 and 6 min later were markedly changed from that at 0.5 min (Fig. 1, spectrum b and spectrum c, respectively).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Dithionite-reduced CO difference spectra for the Arg-mutant. The purified P450 was bubbled with CO, a trace amount of dithionite was added to the sample cuvette, and the spectrum was recorded from 400 to 600 nm after 0.5 min (a), 3 min (b), and 6 min (c). Experiments were performed at least four times.

The addition of benzphetamine to WT, as well as the Ser-, Ala-, and Val-mutants of P450 2B1 resulted in classical type I spectral changes and the addition of n-octylamine to these P450s resulted in classical type II spectral changes (data not shown). These findings suggest that there are no major conformational changes within the active sites of the Ser-, Val-, and Ala-mutants. However, the addition of benzphetamine or n-octylamine to the Arg-mutant induced no spectral change, suggesting an alteration in the active site of the Arg-mutant that either prevents binding of these compounds or eliminates the spectral change normally elicited by their binding.

Western Blotting Analysis. Equal amounts of WT and mutant P450s were subjected to Western immunoblotting analysis and probed with the anti-P450 2B1 antibody (data not shown). The Ser-, Val-, and Ala-mutants expressed immunoreactive P450 2B1 at levels comparable with the WT apoprotein. However, the Arg-mutant displayed a much stronger signal for P450 2B1 apoprotein, indicating either that the Arg-mutant does not incorporate heme into the apoprotein well as the other P450s or that this mutant cannot maintain a stable heme environment.

Catalytic Activities. As shown in Fig. 2, the overall catalytic activity of the mutants was generally lower than that of the WT enzyme toward the four substrates tested except that the Val-mutant retained essentially full activity toward benzphetamine and the Ser-mutant exhibited slightly increased activity toward EFC. The catalytic activity of the Arg-mutant was less than 4% compared with the activity of WT enzyme (data not shown). In general, the changes in the catalytic activities due to any given mutation were similar with EFC, testosterone, and androstenedione as substrates. The Ser-mutant displayed the least change in its activities. The Ala-mutant, lacking the γ-hydroxyl on the side chain, was approximately 30 to 50% as active compared with the WT enzyme. The Val-mutant, in which the γ-hydroxyl was replaced with a methyl group, exhibited even less catalytic activity than the Ala-mutant. These results suggest the requirement for a γ-hydroxyl group in determining maximal catalytic activity toward EFC, testosterone, and androstenedione. The K_m values determined for EFC deethylation activity are 15 μM for WT and 18 μM for the Val-mutant. The Eadie-Hofstee plots are linear for both WT and Val-mutant enzymes (data not shown). In contrast, benzphetamine demethylation activity was fully retained in the Val-mutant, but was approximately one-half as active in the Ser- and Ala-mutants compared with the WT. These results suggest that the methyl group, not the hydroxyl group, on the γ-position is required for benzphetamine to exhibit maximal activity.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Comparison of enzymatic activities of WT and mutant P450s. EFC, benzphetamine, testosterone, and androstenedione were used as substrates to determine the enzymatic activities of WT, Ser-, Ala-, Val-, and Arg-mutants with the same concentrations of P450 and the conditions, as described under Materials and Methods. The specific activity of wild-type P450 2B1 in the reconstituted system was 1.7 nmol of HFC/min/nmol of P450 for EFC deethylation, 74 nmol of formaldehyde/min/nmol of P450 for benzphetamine demethylation, 12 nmol of total products/min/nmol of P450 for testosterone hydroxylation, and 24 nmol of total products/min/nmol of P450 for androstenedione hydroxylation. The experiments were performed at least four times in duplicate.

Regio- and Stereoselectivity of Androgen Metabolism.Fig. 3 illustrates representative chromatographic metabolite profiles for the three major products generated from P450 2B1-catalyzed testosterone metabolism. The androstenedione shown in the elution profile from the reaction mixture that contained no P450 is due to an impurity in the testosterone stock. It can be seen that the Arg-mutant is essentially catalytically inactive for the metabolism of testosterone. The Ser-mutant displayed a chromatographic profile that was very similar to that of the WT enzyme. The generation of 16β-hydroxytestosterone is negligible in the Val-mutant and suppressed to a very low level in the Ala-mutant. Overall, it seems that the 16β-hydroxylation activity was suppressed to a greater extent than 16α-hydroxylation activity.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

High-performance liquid chromatography chromatograms of the testosterone metabolites generated by the P450s. The separations of the metabolites were performed using a C₁₈ reversed phase column eluted isocratically with 65% methanol for the WT, Ser-, Val-, Ala-, and Arg-mutants, as described under Materials and Methods. A separation for a reaction mixture incubated without P450 was also included as a background control. Chromatograms were monitored at 254 nm for the metabolites.

The stereoselectivity was further studied by using androstenedione as a substrate for the WT, Ser-Val-, and Alamutants (Fig. 4). Interestingly, the generation of 16α-hydroxyl product occurred to a similar extent with all four proteins, whereas the generation of 16β-hydroxyl product varied dramatically depending on the identity of the mutated amino acid. The Ser-mutant still retained almost all the activity for 16β-hydroxylation, the Ala-mutant retained approximately one-half the activity, and the Val-mutant only retained ∼20% of the activity compared with the WT. The molar ratios for 16α-hydroxytestosterone/16β-hydroxytestosterone/androstenedione are 1:0.7:0.7 for WT, 1:0.8:0.8 for the Ser-mutant, 1:0.2:0.7 for the Ala-mutant, and 1:0.01:0.37 for the Val-mutant. The molar ratios for 16α-hydroxyandrostenedione/16β-hydroxyandrostenedione are 1:9 for WT, 1:8 for the Ser-mutant, 1:5 for the Ala-mutant, and 1:2 for the Val-mutant. The data clearly show that 16β-hydroxylation activity is severely impaired upon the replacement of Thr-205 by Val or Ala. Once again, the Arg-mutant exhibited no catalytic activity.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

High-performance liquid chromatography chromatograms of the androstenedione metabolites formed by the P450s. Analyses were performed using a C₁₈ reversed phase column, as described under Materials and Methods. Chromatograms were monitored at 254 nm for the metabolites.

Based on the results in Figs. 3 and 4, we can conclude that 1) the substitution of Ser for Thr resulted in the least modification of catalytic activity and the substitution of Ala for Thr caused moderate decreases in enzymatic activity, suggesting that the hydroxyl group in the γ-position of Thr is critical for the maintenance of maximal activity and for the regio- and stereoselectivity of androgen oxidation; 2) the Val-mutant, in which the γ-hydroxyl was replaced with a methyl group, resulted in the most pronounced attenuation of the 16β-hydroxylase activity, suggesting that the hydrophobic environment alters the binding orientation at the 16β-face of androgen in the active site; 3) when Thr was replaced by Val, the molar ratios of 16α-hydroxytestosterone/16β-hydroxytestosterone/androstenedione changed from 1:0.7:0.7 for WT to 1:0.01:0.37 for the Val-mutant, indicating that both stereo- and regioselectivity were altered. In short, Thr-205 plays an important role in determining the product specificity for androgen metabolism by P450 2B1.

P450 and Reductase Complex. Because the Arg-mutant exhibited essentially no catalytic activity with any of substrates tested, the ability of the Arg-mutant to be reduced by electrons transferred from NADPH by the reductase was investigated. The CO difference spectra of the P450s reconstituted with reductase and then incubated with NADPH were compared before and after adding dithionite. The amount of ferrous-carbonyl complex reduced enzymatically versus that further reduced by dithionite was very similar for WT as well as the Ser-, Ala-, and Val-mutants (a representative P450 spectrum for the WT protein is shown in Fig. 5A). In contrast, no detectable spectrum at 450 nm was seen when NADPH was added to the Arg-mutant in the presence of reductase, whereas the P450 spectrum could be detected after adding a trace amount of dithionite (Fig. 5B). It seems that the transfer of the first electron by the reductase from NADPH to the heme iron in this mutant is severely impaired.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

CO difference spectra of the P450 and reductase complex for the WT (A) and Arg-mutant (B). Repetitive difference spectra were recorded from samples containing 250 pmol of P450, 500 pmol of reductase, and 0.5 mM NADPH after bubbling CO gas until a steady state was attained. A trace of sodium dithionite was then added to the samples and additional scans were performed. The wavelength for maximal absorbance is at 450 nm.

Mechanism-Based Inactivation. EFC deethylation was used as a marker activity to determine whether the residue at position 205 was involved in the inactivation of P450 2B1 by 2EN or 17EE (Fig. 6). The extent of inactivation by 17EE in the Ser- and Ala-mutants was essentially the same as that observed in the WT. However, substitution of a Val residue at position 205 completely abolished the inactivation by 17EE. The ability of 2EN to inactivate the Ala- and Val-mutants, was significantly decreased compared with WT. 2EN inactivated the Ser-mutant as efficiently as the WT. When the concentrations of 2EN and 17EE were increased 10-fold, there was no significant change in the activity remaining in Val-mutant. These results indicate that a single amino acid substitution can alter the susceptibility of a P450 to inactivators and confirms the importance of the Thr-205 in P450 2B1 as a determinant of the substrate specificity.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 6.

Inactivation of the EFC deethylation activity of the WT, Ser-, Ala-, and Val-mutant P450s by 2EN and 17EE. Each P450 isoform was incubated with 17EE or 2EN in the absence (100%) or presence of 1 mM NADPH, and then the EFC activity was measured. The experiments were performed three times in triplicate. *, p < 0.001, compared with the WT enzyme.

Discussion

Thr-205 was mutated to Arg in P450 2B1 to test our hypothesis that the disruption of the reduced CO spectrum in the double mutant (Tyr-203 to Ala and Thr-205 to Arg) was due to the Arg residue at position 205. The Arg-mutant did not exhibit any spectral change with benzphetamine and had no catalytic activity with four different substrates. The Arg-mutant could not form a CO complex in the presence of NADPH and reductase, but a transient CO complex was observed after the addition of dithionite. The specific content of P450 was 11 nmol/mg for WT and 2 nmol/mg for the Arg-mutant. Western immunoblotting analysis revealed that the majority of apoprotein cannot form holoenzyme compared with WT. Our results with the Thr-205 to Arg conversion in P450 2B1 are similar to those reported for the Thr-301 to His conversion in P450 2C (Imai and Nakamura, 1988). Presumably the bulky, positively charged Arg residue at position 205 impacts negatively on the interaction of the heme group with the apoprotein. These results suggest that the residue at position 205 may play an important role in substrate binding and in maintaining a required structural environment for the heme. Therefore, the functional role of Thr-205 was further examined by replacing Thr-205 with other amino acids.

Halpert and coworkers have substituted Ser for Thr-205 in P450 2B1 and have demonstrated that this mutation did not change the enzymatic activities for the metabolism of androstenedione and progesterone (Szklarz et al., 1995). Either Thr or Ser at position 205 is well conserved in the P450 2B family (Lewis and Lake, 1997). The catalytic activities for the metabolism of EFC, testosterone, and androstenedione after mutation were Ser-mutant > Ala-mutant > Val-mutant. In contrast, benzphetamine metabolism was fully retained in the Val-mutant. From these experimental data, it seems that the specific catalytic activity may be related to the structures and flexibility of the substrates as described previously (Furuya et al., 1989). Benzphetamine is a cationic, hydrophobic, and flexible molecule, whereas testosterone, androstenedione, and EFC are rigid molecules with fused ring systems.

A number of structural studies have suggested that the C-terminal portion of the F helix may be important in forming the ceiling of the substrate binding pocket in P450s (Hasemann et al., 1995; Williams et al., 2000; Pikuleva et al., 2001). To facilitate the interpretation of our experimental data, we constructed a P450 2B1 homology model based on the crystal structure of P450 2C5 (Williams et al., 2000; Lin et al., 2003). The structure of the distal surface of P450 2B1 is displayed in Fig. 7A. Thr-205 is located at the C-terminal end of the F helix and its hydroxyl group is exposed to the substrate-heme pocket. The distance between the hydroxyl group of Thr and the heme iron is approximately 15 Å. Figure 7B shows the stable conformers of testosterone with dimensions of 11.73 × 10.07 × 7.89 Å and benzphetamine with dimensions of 11.89 × 11.44 × 8.33 Å. Testosterone has one hydrophilic oxygen atom at C-3 on the A-ring and another at C-17 on the D-ring. In contrast, benzphetamine has two hydrophobic phenyl groups, one on each side of the nitrogen atom. Moreover, the P450 2B1-substrate complex models propose that both C-16 and C-17 of testosterone are nearest to the heme iron and that there are numerous hydrophobic interactions between the two benzene rings in benzphetamine and residues in the 2B1 active site (Dai et al., 1998). Thus, we suggest that the entry of testosterone and androstenedione into the active site from the F-G loop is governed by the orientation of C-16 and C-17 toward the heme iron and the orientation of C-3 toward the Thr-205 side chain through hydrogen bonding or electrostatic properties. On the other hand, the hydrophobic side chain of residue 205 favors the proper delivery of benzphetamine to the substrate-heme pocket.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 7.

Homology model of P450 2B1 showing the distal surface and the location of Thr-205 (A) and the structures of two P450 2B1 substrates (B). The distance between the hydroxyl group of Thr-205 and the heme iron is ∼15 Å. The skeletons of testosterone with dimensions of 11.73 × 10.07 × 7.89 Å and benzphetamine with dimensions of 11.89 × 11.44 × 8.33 Å are embedded in their electrostatic surfaces. The red color represents the electronegative surface (3-keto on the A-ring and 17-hydroxyl on the D-ring of testosterone; nitrogen atom of benzphetamine), whereas the blue color represents the electropositive surface.

When the regio- and stereoselectivity for androgen metabolism were studied in detail, the importance of the hydroxyl side chain of Thr-205 became more obvious. P450s 2B1 and 2B2 hydroxylate the 16α- and 16β-positions of testosterone to about the same extent, whereas they hydroxylate androstenedione primarily at the 16β-position (Waxman et al., 1983; Wood et al., 1983). P450 2B4 and 2B6 also hydroxylate the 16α- and 16β-position of testosterone to about the same extent (H. Lin and P. F. Hollenberg, unpublished data). In the Ser-mutant, where the hydroxyl group on the side chain is retained, the total activity, as well as the ratio of the 16α-OH to the 16β-OH product is very similar to that of WT. The Ala- and Val-mutants, which lack the γ-hydroxyl group, exhibited a marked suppression of the 16β-hydroxylation activity with little alteration of 16α-hydroxylation activity for testosterone and preferentially decreased the 16β-hydroxylation activity for androstenedione (Figs. 3 and 4). Because 16β-hydroxylase activity is a unique marker for P450 2B activity, the hydroxyl group of residue 205 is functionally and structurally important in steering the 16β-face of androgens toward enzyme active site. Perhaps the absence of the hydroxyl group in the Val-mutant of P450 2B1 distorts or twists the four-ring system of androgen and reorients the 16β-face away from the heme.

Three amino acid residues have been identified from two distinct allelic variants in P450s 2B1 and 2B2, which contribute to the unique 16β-hydroxylation activity for testosterone and androstenedione. The 2B2 variant having Phe at position 58 instead of Leu and Phe at position 114 instead of Ile did not catalyze the 16β-hydroxylation of testosterone and androstenedione (Aoyama et al., 1989). The 2B1 variant with a substitution of Ala for Gly at position 478 exhibited a 10-fold lower androstenedione 16β-hydroxylation activity (Kedzie et al., 1991). The functional and structural characteristics of P450 2B1 have been studied extensively using site-directed mutagenesis by Domanski and Halpert (2001). Their studies have identified several residues required for 16β-hydroxylation of testosterone and androstenedione. For example, mutations of Phe-115 to Ala, Phe-206 to Leu, Leu-209 to Ala, Ser-294 to Ala, Ala-298 to Val, Thr-302 to Ser, and Val-363 to Ala all diminished the 16β-hydroxylation activity for androgens (He et al., 1994; Szklarz et al., 1995; Domanski et al., 2001). Together, Leu-58, Ile-114, Phe-115, Phe-206, Leu-209, Ser-294, Ala-298, Thr-302, Val-363, Gly-478, as well as the Thr-205 that has been identified in this study, contribute to the unique characteristics of P450 2B1.

Mechanism-based inactivators have proven to be valuable probes to study functionally important residues in P450 2B1. Residues 114, 302, 363, 367, and 478 have been identified as being in the active site by using mechanism-based inactivators such as secobarbital, N-benzyl-1-aminobenzotriazole, chloramphenicol and N-(2-p-nitrophenethyl)-chlorofluoroacetamide (Kedzie et al., 1991; He et al., 1994, 1996; Kent et al., 1997). The importance of Thr-205 was further emphasized by investigating the susceptibility of the mutant P450s to inactivation by two well characterized mechanism-based inactivators of P450 2B1. The substitution of Val for Thr abolished the inactivation by 17EE and markedly decreased the sensitivity to inactivation by 2EN compared with WT. We have previously demonstrated that both 2EN and 17EE modify the apoprotein rather than the heme (Roberts et al., 1993; Kent et al., 2002). The resistance of the mutants to inactivation can be attributed to either an inability of the P450 to generate a reactive intermediate or to decrease covalent binding to the protein. The role of Thr-205 in the inactivation of P450 by these two inactivators is currently being investigated. Our preliminary results show that the Val-mutant metabolizes 2EN to 2-naphthylacetic acid as efficiently as the WT and that the catalytic activity for 17EE metabolism by the Val-mutant is 70% of that by the WT enzyme.

The contributions of several amino acids in the F helix to substrate specificity and the regio- and stereoselectivity of product formation have been identified in P450 2A4/2A5, P450 2B1, P450 27A1, P450 11A1, and P450 3A4 (Lindberg and Negishi 1989; Domanski and Halpert, 2001; Xue et al., 2001; Pikuleva et al., 2001). Studies with several P450 models suggest that the F-G loop serves as a hydrophobic membrane anchor and substrate entrance channel and residues lining the interior of the substrate entrance channel may determine the orientation of substrate as it enters the active site (Graham-Lorence et al., 1995; Hasemann et al., 1995; Dai et al., 1998). Moreover, the crystal structure of a complex between P450 BM3 and N-palmitoylglycine indicates that the movement of the “lid domain” positions the substrate molecule properly in the active site (Haines et al., 2001). Presumably, the oxygen atom at C-3 of testosterone is in contact with the hydroxyl group of Thr-205, and the movement of the F helix may propagate into the active site region and ultimately stabilize the orientation of C-16 and C-17 of testosterone.

In conclusion, using site-directed mutagenesis and mechanism-based inactivators, we have provided evidence that Thr-205 in the F helix plays an important role as a determinant for P450 2B1 specificity. These findings also support the hypothesis that residues lining the interior of a substrate entrance channel may function as substrate recognition sites and control the orientation of a substrate in the process of entering and subsequently binding to the active site.