Binding Specificity Determines the Cytochrome ... - ACS Publications

Article pubs.acs.org/JPCB

Cite This: J. Phys. Chem. B XXXX, XXX, XXX−XXX

Binding Specificity Determines the Cytochrome P450 3A4 Mediated Enantioselective Metabolism of Metconazole Shulin Zhuang,† Leili Zhang,‡ Tingjie Zhan,† Liping Lu,†,§ Lu Zhao,† Haifei Wang,† Joseph A. Morrone,‡ Weiping Liu,*,† and Ruhong Zhou*,‡,∥ †

College of Environmental and Resource Sciences, Zhejiang University, Hangzhou 310058, China Computational Biology Center, IBM TJ Watson Research Center, Yorktown Heights, New York 10598, United States § Institute of Quantitative Biology, Department of Physics, Zhejiang University, Hangzhou 310058, China ∥ Department of Chemistry, Columbia University, New York, New York 10027, United States ‡

S Supporting Information *

ABSTRACT: Cytochrome P450 3A4 (CYP3A4) is a promiscuous enzyme, mediating the biotransformations of ∼50% of clinically used drugs, many of which are chiral molecules. Probing the interactions between CYP3A4 and chiral chemicals is thus essential for the elucidation of molecular mechanisms of enantioselective metabolism. We developed a stepwise-restrained-molecular-dynamics (MD) method to model human CYP3A4 in a complex with cis-metconazole (MEZ) isomers and performed conventional MD simulations with a total simulation time of 2.2 μs to probe the molecular interactions. Our current study, which employs a combined experimental and theoretical approach, reports for the first time on the distinct conformational changes of CYP3A4 that are induced by the enantioselective binding of cis-MEZ enantiomers. CYP3A4 preferably metabolizes cis-RS MEZ over the cis-SR isomer, with the resultant enantiomer fraction for cis-MEZ increasing rapidly from 0.5 to 0.82. cis-RS MEZ adopts a more extended structure in the active pocket with its Cl atom exposed to the solvent, whereas cis-SR MEZ sits within the hydrophobic core of the active pocket. Free-energy-perturbation calculations indicate that unfavorable van der Waals interactions between the cis-MEZ isomers and the CYP3A4 binding pocket predominantly contribute to their binding-affinity differences. These results demonstrate that binding specificity determines the cytochrome P450 3A4 mediated enantioselective metabolism of cis-MEZ.

1. INTRODUCTION

Triazole derivatives are widely used as fungicides that can be metabolized by CYP3A4.8−12 Because of their chemical stability, triazole-fungicide residues have been detected in many environmental matrices causing adverse effects.13,14 It was previously found that the metabolism of triazole fungicides may influence their toxicokinetics.15 Although there were some reports on the metabolism of a triazole chemical with the CYP450 enzyme,8,9 absolute stereochemical properties were resolved for very few triazole chemicals.16,17 The shortage of structural information hinders the mechanistic elucidation of the enantioselective metabolism at the atomic level. In this study, using a multitechnique experimental and computational approach, we attempt to conduct an in-depth investigation on how human CYP3A4 mediates the enantioselective metabolism of metconazole (MEZ, Figure S1), which is a relatively new triazole fungicide with a substituted triazole moiety that presumably binds to the heme portion of the fungal cytochrome P450 and has been effectively used to protect many

Cytochrome P450 3A4 (CYP3A4) is the predominant isoform among the cytochrome P450 (CYP) family and is mainly found in the human liver; it is also the most promiscuous CYP due to its flexible active site.1−3 CYP3A4 mediates the biotransformations of about 50% of clinically used drugs and is responsible for some side effects caused by drug interactions.4 The active site of CYP3A4 is connected to its surface via the substrate access channel.5−7 Interactions between CYP3A4 and its ligands mainly utilize the active site and adjacent heme iron. During their entry into the access channel, the ligands dynamically interact with the surrounding residues. Williams et al. reported that ligand binding to human CYP3A4 may potentially trigger conformational movements of the residues in the phenylalanine-rich hydrophobic cluster of the active site, thus facilitating the entry of a substrate suitable for metabolism.5 Within the CYP family, substrate−P450 interactions and access-route preferences were previously reported to influence substrate specificity.7 The characterization of ligand interactions is thus essential for probing the molecular mechanism of substrate metabolism. © XXXX American Chemical Society

Received: November 11, 2017 Revised: December 28, 2017 Published: January 8, 2018 A

DOI: 10.1021/acs.jpcb.7b11170 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B

our works. All simulations were performed with particle-mesh Ewald (PME) treating the electrostatics. The van der Waals interactions were calculated with a switching function turned on at 1.0 nm and a cutoff at 1.2 nm. The time step was set to 2.0 fs with all the hydrogen-related bonds set as rigid (rigidBonds = all). Langevin dynamics were adopted at a temperature of 310 K. The Nosé−Hoover Langevin-piston method was used in the constant pressure simulations, with a targeted pressure at 1 atm. From the docked structures, after at least 10 000 steps of steepest-descent energy minimization, 1 ns equilibrations were carried out in the NVT ensemble at 310 K with constrained positions on both CYP3A4 and MEZ before the stepwise-restrained-molecular-dynamics simulations. Because of the limitation that molecular docking is unable to recover the Fe−N bond observed experimentally for the CYP3A4−triazole complex, stepwise restrained molecular dynamics were performed to recover the coordinate bond between the N3 from MEZ triazole ring and the Fe center in CYP3A4. The final restraint distance between N3 and Fe was set to 3.0 Å with 100 ps for each stage between 7.0 and 3.0 Å, which were separated by nine stages, with a force constant of 1 kcal/mol/Å2. The initial and final N3−Fe distances are listed in Table S1. After the restrained molecular dynamics, the final structure was equilibrated again for 60 ns in an NpT ensemble with an additional bond on N3−Fe restrained at 3.0 Å. 2.4. Free Energy Perturbation (FEP). We selected the binding structures on the basis of RMSD and the N3−Fe distance. We assumed that on the basis of the molecular resemblance between ketoconazole and MEZ, the final binding structure of CYP3A4 bound to MEZ would be similar to the structure of CYP3A4 bound to ketoconazole. Thus, three of the five structures from the MD simulations were picked to calculate the binding free energy with FEP methods based on the RMSDs of CYP3A4 compared with those of PDB 2V0M (Figure 1).24 All other simulation conditions were the same as those of the NpT equilibrations of the MD runs. The binding free energy was calculated following a thermodynamic cycle in which the obtainable free-energy changes were formations of MEZ in the CYP3A4 binding pocket (ΔGc) and formations of

crops against fungal diseases. We assign the configurations of the MEZ enantiomers by the combination of in vitro and in silico electronic circular dichroism (ECD). The underlying mechanism of enantioselective metabolism is deciphered by probing the molecular interaction of MEZ and CYP3A4 with the combination of an in vitro fluorescence binding assay and in silico modeling at the atomic level. The obtained results provide atomic insights for metabolism-based drug design and relevant screening of chemicals with adverse effects.

2. METHODS 2.1. Human-CYP3A4-Mediated Metabolism. The MEZ racemate was separated, and the cis-MEZ isomers were assigned their configurations by the use of ECD and density-functional theory (DFT, Texts S1−S3 and Figure S2). The metabolism of the racemate and enantiomers of MEZ by the human-CYP3A4 and NADPH system (BD Biosciences Company, Shanghai, China) was performed by following the manufacturer’s protocol. A solution of 10 μL of MEZ (5 μM), 50 μL of solution A, 10 μL of solution B, human CYP3A4 (0.1 pmol), and potassium phosphate buffer (110 mM, pH 7.4) was incubated in a 5 mL glass tube at 37 °C for 0−120 min (0, 30, 60, 90, or 120 min). Reactions were terminated using 1 mL of ice-cold methanol. The incubation assay was performed in triplicate. The enantiomer distribution of cis-MEZ was evaluated by the enantiomer fractions (EFs). An EF was calculated as the ratio of the relative peak area (RPA) of the enantiomer eluted first (peak 1) to the total concentration of both enantiomers (peak 1 + peak 2). The concentrations of the cis-MEZ isomers were evaluated by the RPA values obtained from HPLC/MSMS (Text S4). The Michaelis−Menten half-saturation constant (Km) and the maximum velocity of the metabolic reaction (Vmax) for the metabolism of MEZ were calculated by the Michaelis−Menten kinetic model (eq 1) using a substratedeletion method.18 V=

Vmax × [S] K m + [S]

(1)

where V is the velocity of metabolism, and S is the concentration of the substrates. The intrinsic clearance, CLint, was calculated as Vmax/Km. 2.2. Molecular Docking. The molecular docking method was used to generate initial guesses with the MEZ isomers in the active site of the CYP3A4 enzyme. The coordinates of CYP3A4 (PDB ID: 2V0M, Figure S3) were chosen as the starting conformations for the molecular docking. The original binding ligand, ketoconazole, was removed from the PDB file. The system was prepared with AutoDockTools and AutoDock Vina,19 with the searching box centered at the binding site of ketoconazole in PDB 2V0M. The top five candidates of both cis-RS and cis-SR MEZ were then selected for further investigations. 2.3. Molecular Dynamics Simulations. Following the molecular docking, missing atoms were restored for the top five conformations for both the cis-RS and cis-SR MEZ binding structures. We adopted a CHARMM36 force field for the CYP3A4 protein.20,21 MEZ force fields were generated using CGenFF.22 The system was solvated with a 1.5 nm TIP3P water shell, resulting in a 9.5 × 9.2 × 9.6 nm3 box with ∼80 000 atoms. The NaCl concentration was set to 0.1 M for all the systems. NAMD 2.923 was used as the MD package throughout

Figure 1. Flowchart of FEP calculations. B


Article


Figure 2. Stereoselective metabolism of cis-MEZ by the human CYP3A4 enzyme. (A) Time-dependent metabolism of the cis-SR and cis-RS isomers. (B) Measured EF value of cis-MEZ during the metabolic process.

5, 10, 15, and 20 μM, the steady-state fluorescence intensities were monitored and recorded at 298 and 310 K using an excitation wavelength of 290 nm and emission wavelengths ranging from 300 to 500 nm. The integration time was 0.1 ns, and the entrance and exit slits were both set at 7 nm. The inner-filter effect was eliminated during the fluorescence-spectra analysis. All the spectroscopy measurements were carried out in triplicate.

MEZ in an aqueous solution (ΔGw). Thus, the binding free energy, ΔΔGbind, was calculated as ΔΔG bind = ΔGc − ΔGw

(2)

Note that signs are flipped if the free-energy change is defined as annihilation free energy. For simplicity, details can be found in Zhou et al.24 In addition, the soft-core potentials presented in Xia et al.25 were adopted for the Lennard-Jones interactions. The formation free energy of MEZ in each of the environments was calculated with the following equations: −1

ΔGλ = −β ln⟨exp( −β[V (λ + Δλ) − V (λ)])⟩λ

ΔG =

∑λ ΔGλ

3. RESULTS AND DISCUSSION 3.1. Assignment of the Absolute Configuration. The configurational assignment of chiral isomers has always been a bottleneck. MEZ has four isomers, two cis-isomers and two trans-isomers. Adiabatic time-dependent density-functional theory (TDDFT) together with ECD experiments is proved to produce high efficiencies and accuracies in the characterization of ECD spectra of chiral chemicals,26,27 which is the same approach applied here for obtaining the absolute configuration of MEZ. We separated the MEZ racemate using a SUPELCOSIL LC-18 HPLC column (250 × 4.6 mm, 5 μm, Text S1) and obtained four peaks, namely, peaks A−D, with retention times of 6.05, 6.49, 7.33, and 7.66 min, respectively (Figure S5). Peaks A−D account for 47.57, 47.58, 2.58, and 2.27% of the peak area, respectively, indicating that peaks A and B are the predominant ones. The combination of retention times and peak areas verifies that peaks A and B and peaks C and D are corresponding isomer pairs. The predominant enantiomers corresponding to peaks A and B were the ones mainly used in this study. We resolve the absolute configurations of the two enantiomers eluted from peaks A and B by a combination of an ECD experiment and TDDFT calculations at bvp86/6− 31+g(d,p) levels of theory (Texts S2 and S3). As shown in Figure S2, the experimentally determined ECD of the two enantiomers exhibit opposite bands at 198 and 222 nm. TDDFT regenerates the ECD spectra for these two enantiomers, in full accordance with those determined from the ECD experiment. The assignments of both the experimental and calculated ECD spectra show that the enantiomers eluted from peaks A and B correspond to cis-RS and cis-SR MEZ, respectively. With the assigned absolute conformations of MEZ, we are able to decipher the molecular mechanisms of the CYP3A4−MEZ interactions and MEZ metabolism mediated by CYP3A4 on the basis of the structure−activity relationship.

(3) (4)

where V(λ) = (1 − λ)V1 + λV2, V1 denotes the potential energy of the free state, and V2 denotes the potential energy of the bound state. λ is the free-energy-perturbation parameter of choice, and in this study, we used a 20 window scheme (0.000 01, 0.0001, 0.001, 0.01, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.95, 0.99, 0.999, 0.9999, 0.999 99, and 1). β is the Boltzmann factor. The averaged binding free energy was calculated with 5 independent FEP runs. The total time of the FEP calculations was 19 windows × 0.4 ns × 5 runs × 6 binding modes = 228 ns. 2.5. Inhibition of CYP3A4 Enzyme Activity. The inhibition of human-CYP3A4 activity was measured by a P450-Glo CYP3A4 Assay with Luciferin-IPA (Promega Corporation, Madison, WI). Solutions (12.5 μL) of metconazole enantiomers with concentrations ranging from 0.1 to 10 μM were added to each well of an opaque white 96-well plate and incubated with CYP3A4 (0.1 pmol) for 10 min at 37 °C. The reaction was activated by adding 25 μL of 2× NADPH regeneration reagents. After a 10 min incubation at 37 °C, 50 μL of the luciferin-detection reagent was added, and the mixtures were stabilized for 20 min. The luminescence signal was measured with an Infinite 200 PRO NanoQuant (Tecan Group Ltd., Männedorf, Switzerland). Ketoconazole was used as the positive control, and all measurements were performed in triplicate. 2.6. Fluorescence Spectroscopy. All steady-state fluorescence spectra were measured on a Fluomax-4 spectrofluorimeter (Horiba Jobin Yvon IBH, Glasgow, U.K.) equipped with a 1 mm quartz cuvette. Human CYP3A4 was purified (Text S5 and Figure S4) and stocked in 0.20 M Tris-HCl buffer (pH = 7.4, 0.1 M NaCl) with a final concentration of 0.2 μM. After the MEZ enantiomers were added at concentrations of 1, C


Article


Figure 3. Inhibitory effects of (A) ketoconazole, (B) the MEZ racemate, (C) cis-SR MEZ, and (D) cis-RS MEZ on human CYP3A4 enzyme activity. Pure CYP3A4 enzymes were exposed to MEZ and ketoconazole in a dose-dependent manner (0−10 μM) for 20 min. The data are presented as the mean changes in the enzymatic activities ± SD (n = 3).

3.2. Enantioselective Metabolism Mediated by Human CYP3A4. As the predominant isoform of the human-CYP450 family, CYP3A4 mediates the metabolism of various chemicals.28−31 To explore how CYP3A4 mediates the metabolism of MEZ, the cis-MEZ racemate comprising the cisSR and cis-RS enantiomers at 5 μM was incubated with the human CYP3A4 enzyme (0.1 pmol) for 0, 30, 60, 90, and 120 min. As the positive control, triadimefon, a known triazole fungicide metabolized by the human-liver microsome (HLM) and CYP3A4 (0.1 pmol) was chosen for the experiment (Figure S6). The degradation rate of triadimefon at 1 h is around 40%, in line with the previous reported ∼40% metabolic rate mediated by HLM.32 After the incubation with CYP3A4 for 120 min, the degradation rate of triadimefon is up to 80%. During the process of the metabolism, the concentrations of the MEZ isomers were measured. The plot of their concentrations versus the metabolic velocity shows that the metabolism of MEZ by human CYP3A4 follows first-order kinetics (Figure S7). The key kinetic parameters, including Vmax and Km, were obtained (Table S2). The cis-RS isomer has a Vmax of 12.325 μM/min and a Km of 3.127 μM, showing the very fast metabolism of MEZ by human CYP3A4. In contrast, cis-SR has a lower Vmax of 10.422 μM/min and a higher Km of 53.873 μM, indicating that cis-SR has a lower potential of being metabolized by CYP3A4. The RPA of MEZ decreases in a time-dependent manner with increasing incubation time (Figures 2 and S8), which shows the gradual metabolism of MEZ. Compared with those of the control racemate, which was not incubated with CYP3A4, the metabolic rates of cis-RS and cis-SR MEZ after the 120 min incubation were 80 and 40%, respectively (Figure 2A). The metabolic rates indicate that human CYP3A4 predominantly metabolizes cis-RS. The incubation of the racemate with CYP3A4 shows that cis-RS and cis-SR have

metabolic rates of 80 and 35% after a 120 min incubation, respectively (Figure S8), consistently confirming the enantioselective metabolism. To monitor this enantioselectivity, the variation of the EFs for cis-SR and cis-RS MEZ was calculated as the ratio of the RPA of cis-SR MEZ over the sum of the RPAs of cis-SR and cisRS MEZ. The EF value for the cis-MEZ racemate is close to 0.5 before the incubation and rapidly increases to 0.82 after the 120 min incubation (Figure 2B). The changing EF value indicates the different metabolic rates of the two enantiomers according to different incubation times. The EF value of 0.82 shows the predominant metabolism of cis-RS by CYP3A4. 3.3. Stereoselective Inhibition of Human CYP3A4 Activity. The inhibition or induction of the liver CYP enzyme by azole fungicides is one of the key events in their tumorigenesis. The carcinogenic and noncarcinogenic azole fungicides reportedly showed different activities toward the CYP450 of medaka fish.14 To investigate the effects of MEZ enantiomers on the enzymatic activity of CYP3A4, ketoconazole (Figure S1), a well-known strong, reversible inhibitor of CYP3A4 is used as the positive control. Our P450-Glo CYP3A4 bioassay (Luciferin-IPA) shows the changes in the enzymatic activity of CYP3A4 after an incubation with ketoconazole, the MEZ racemate, and cis-enantiomers with concentrations ranging from 0 to 10 μM (Figure 3). The percentage of the enzymatic activity of CYP3A4 significantly decreases after the incubation, suggesting the inhibition of ketoconazole and MEZ. This inhibition occurs in a concentration-dependent manner. Our determined IC50 of ketoconazole is 0.189 μM, close to the reported IC50 value of 0.18 μM.33 The IC50 of the MEZ racemate (0.213 μM) is close to that of ketoconazole (Figure 3A,B), indicating that MEZ is a strong inhibitor of CYP3A4, comparable with ketoconazole. For cis-MEZ, the IC50 values in order are cis-SR (0.15 μM) < the racemate (0.213 μM) < cis-RS D


Article

The Journal of Physical Chemistry B (0.29 μM) (Figure 3C,D). The cis-SR isomer is a comparatively stronger inhibitor than ketoconazole, and it is more potent than the cis-RS isomer, showing significant enantioselective inhibition. The minor stereochemical variation between the enantiomers causes significant differences in both the inhibition of CYP3A4 activity and the enantiomers’ metabolism by CYP3A4. Compared with their metabolism by CYP3A4, cis-SR inhibits more enzymatic activity of CYP3A4 and thus affects the enzyme kinetic behavior more severely, causing the reduced metabolic potency of CYP3A4 toward cis-SR. In contrast, the less potent inhibition by cis-RS results in the higher metabolic rate, Vmax, and smaller Km. 3.4. Free Energy Perturbation (FEP) Calculation of the Binding Free Energies. Triazole-related chemicals were previously reported as having type II binding with CYP3A4 at its active site and direct heme binding.2 We modeled the tertiary structure of CYP3A4 in a complex with the cis-MEZ enantiomers on the crystal structure of CYP3A4−ketoconazole (Figure S3, PDB ID: 2V0M) and on the basis of molecular docking. Because docking methods do not capture the coordinate bond between the nitrogen atoms (N3) in the MEZ triazole ring and the heme iron, stepwise-restrained-MD simulations were adopted to form the N3−Fe bond. The N3− Fe bond was subsequently added as an additional bond in the 60 ns equilibrations. The N3−Fe distances from the docking results (initial N3−Fe) and stepwise-restrained-MD simulations (final N3−Fe) are listed in Table S1. Clearly, no N3−Fe bonds are constructed in any docked structures, which verifies how the subsequent MD simulations are necessary to recover this coordinate bond. RMSDs with the reference of the crystal structure (PDB ID: 2V0M) are listed in Table S1. Trajectories with smaller RMSDs were chosen for further free-energy calculations with the FEP method (Figure 1). The calculated binding free energies are listed in Table 1. The binding modes with the lowest binding free energy for

stable than SR-r0. This can be explained by the binding poses of MEZ in the CYP3A4 binding pocket and will be discussed in detail in the next section. 3.5. Binding Specificity of cis-MEZ to Human CYP3A4. The most favorable binding structures of cis-RS (RS-r0) and cisSR (SR-r0) MEZ obtained from our MD simulations are shown in Figures 4 and S9. The X-ray crystal structure of CYP3A4 in a complex with ketoconazole (PDB ID: 2V0M, 2.8 Å) is shown as a reference. The residues within 4.5 Å of the binding ligands are shown in Figure 4. All residues contributing to the binding of ketoconazole are shown as references: F57, D76, R105, F108, M114, S119, L210, L211, T224, F241, I301, F304, A305, T309, A370, M371, R372, E374, C442, G481, L482, of which 12 residues are hydrophobic, 2 residues are negatively charged, 2 residues are positively charged, and 3 residues are hydrophilic. For the RS-r0-binding mode, (R105, F108, S119, I301, A305), G306, (T309), I369, (A370, M371), G480, G481, (L482) are in close contact with cis-RS MEZ, of which 7 residues are hydrophobic, 1 residue is positively charged, and 2 residues are hydrophilic. The 9 residues in parentheses are the common residues between the bindings of ketoconazole and cis-RS MEZ. For the SR-r0-binding mode, (R105, F108), V111, (M114), A117, (S119), I120, (L211, F241), I300, (I301, F304, A305), I369, (A370), L373 are revealed to be in close contact with cisSR MEZ, of which 13 residues are hydrophobic, 1 residue is positively charged, and 1 residue is hydrophilic. The listed 10 residues in parentheses are the common residues between the bindings of ketoconazole and cis-SR MEZ. Because of the overall-hydrophobic binding pocket of CYP3A4, the polar MEZ molecule is mainly stabilized by backbone interactions and S119. S119 is one of the key residues involved in ligand binding, and the formed hydrogen bonds are the driving force for complex formation. Park et al. revealed that the stable hydrogen bonds of S119 with the ligands metyrapone and progesterone served as the significant binding force to stabilize the ligands at the active site of CYP3A4.1 Overall, the hydrophobic surroundings put a penalty on the binding of cis-MEZ, on the basis of our FEP calculations (Table 1). Specifically, there are two reasons why cis-RS MEZ is more stable in the CYP3A4 binding pocket than cis-SR MEZ: (1) with fewer surrounding hydrophobic residues, the hydrophobic penalty on cis-RS MEZ binding is reduced, and (2) the conformation of cis-RS MEZ ensures its adoption of a more extended structure than cis-SR MEZ, so that cis-RS MEZ is more exposed to the solvent, whereas cis-SR MEZ sits inside the hydrophobic pocket (Figures 4 and S9). 3.6. Distinct Conformational Changes Induced by cisMEZ Enantiomers. The steady-state fluorescence spectra of purified CYP3A4 (0.2 μM, Figure S4) in the presence or absence of the cis-MEZ isomers were monitored at 298 and 310 K. After the elimination of the inner-filter effect, the emission spectrum of CYP3A4 with the excitation wavelength of CYP3A4 at 290 nm exhibits a maximum peak at 340 nm (Figure 5), similar to the reported fluorescence-spectrum maximum peak at λem = 330 nm.34,35 The fluorescence-spectral intensity of CYP3A4 changes in a concentration-dependent manner with the increasing titration concentration of the cisMEZ enantiomers from 0 to 20 μM (steps of 0, 1, 5, 10, 15, and 20 μM). Compared with the spectrum of the ligand-free CYP3A4 (Figure 5, curve a), cis-SR highly enhances the fluorescencespectral intensity, resulting in a slight red shift of the peak maximum of CYP3A4. In contrast, cis-RS significantly quenches

Table 1. FEP Results for the Chosen MD Runsa structureb RS-r0 RS-r2 RS-r5 SR-r0 SR-r1 SR-r2

ΔΔG −6.4 −2.7 −4.0 −4.6 −2.7 −1.9

(1.3) (2.6) (1.9) (2.1) (1.0) (1.5)

ΔΔElec −11.9 −11.4 −12.7 −12.2 −12.8 −12.4

(1.2) (1.2) (1.1) (1.7) (0.9) (1.1)

ΔΔVDW 5.0 5.9 5.2 6.8 7.7 7.1

(1.1) (1.4) (0.8) (1.9) (0.9) (1.2)

coupling term 0.5 2.8 3.5 0.7 2.4 3.4

(1.0) (1.2) (2.0) (1.2) (1.8) (1.3)

All numbers are averages of five independent FEP runs. Standard deviations are shown in parentheses. bRS and SR stand for the different enantiomers of cis-MEZ, and -rn (n = 0, 1, 2, 3, or 5) indicates the independent MD run. a

each stereoisomer are RS-r0 (−6.4 kcal/mol) and SR-r0 (−4.6 kcal/mol), respectively. Here, we obtain a 1.8 kcal/mol difference from the FEP calculations. The experimental Km values differ by ∼17-fold, corresponding to the ∼1.7 kcal/mol difference. Thus, we have a plausible agreement between the experiments and the computational studies. The main contribution of the binding free energy to both cisRS MEZ and cis-SR MEZ is attributed to the electrostatic interactions. However, the difference between the electrostatic contributions of RS-r0 and SR-r0 is a mere −0.3 kcal/mol. The main difference is from the van der Waals (VDW) interactions, which contribute 1.8 kcal/mol, with RS-r0 significantly more E


Article


Figure 4. Stable binding modes and contact residues (within the 4.5 Å radius of MEZ). From left to right: ketoconazole as the binding ligand (PDB ID: 2V0M), the favorable cis-RS MEZ binding mode, and the favorable cis-SR MEZ binding mode. The residues are drawn with surface representation and colored in picture: red, positive; blue, negative; green, hydrophilic; white, hydrophobic. The water molecules surrounding the ligands are drawn with ball-and-stick models (red: oxygen, white: hydrogen). Cl atoms in the ligands are highlighted with green spheres.

Figure 5. Induced conformational changes of the human CYP3A4 enzyme by the binding of the MEZ enantiomers at the concentrations ranging from 0 to 20 μM (steps of 0, 1, 5, 10, 15, and 20 μM; a−f). The fluorescence of CYP3A4 was monitored with λem of 340 nm and λex of 290 nm.

Figure 6. Structural changes in CYP3A4. (A) Ketoconazole-bound structure (PDB ID: 2V0M), (B) cis-RS MEZ bound structure, and (C) cis-SR MEZ bound structure superimposed on ligand-free forms (PDB ID: 1TQN). F-G helix and linker regions are shown in red for the ligand-free forms, brown for the ketoconazole-binding mode (A), black for the cis-RS MEZ binding mode (B), and orange for the cis-SR MEZ binding mode (C). The C-termini of the ligand-free forms are shown in light blue for the ligand-free modes and dark blue for the ligand-binding modes. The H helices are shown in magenta. The I helices are shown in green.

F


Article


Figure 7. Hydrophobic cluster (Y53, F57, F108, F213, F215, F219, F220, F241, and F304) representations for (A) ligand-free CYP3A4 (PDB ID: 1TQN), (B) ketoconazole-bound CYP3A4 (PDB ID: 2V0M), (C) cis-RS MEZ bound CYP3A4, and (D) cis-SR MEZ bound CYP3A4. HEME complexes are shown at the bottom. The two ketoconazole molecules are shown in yellow and brown (B), cis-RS MEZ is shown in black (C), and cis-SR MEZ is shown in orange (D).

the fluorescence intensity of CYP3A4 and causes a blue shift of peak maximum at λem = 340 nm. With the temperature increasing from 298 to 310 K, the fluorescence intensity increases slightly; however, the overall shape of the fluorescence spectrum of CYP3A4 remains unchanged. The distinct changes in the fluorescence intensity indicate quite different conformational changes of CYP3A4 induced by the binding of the cisMEZ enantiomers. To the best of our knowledge, this is the first report on the distinct conformational changes of CYP3A4 induced by the enantioselective binding of triazole enantiomers. At the excitation wavelength of 290 nm, the fluorescence intensity is contributed to mainly by the tryptophan residues (W72, W126, and W408). The increase and decrease of the steady-state fluorescence intensity of CYP3A4 caused by cis-SR and cis-RS MEZ suggest quite different disruptions of the microenvironment of the tryptophan residues upon the binding of cis-SR and cis-RS MEZ. During the processes of moving outward toward the water for cis-RS MEZ and reaching inward to the active site of CYP3A4 of cis-SR MEZ, the enantiomers interact with different residues of CYP3A4. These distinct interactions induce conformational changes in the active site, potentially disrupting the microenvironment of the tryptophan residues. Williams et al. revealed that conformational changes in CYP3A4 might occur upon ligand binding and revealed that progesterone and metyrapone induced little conformational changes.5 The binding of ketoconazole reportedly induced dramatic conformational changes in CYP3A4, causing shifts in the positions of the secondary-structure elements relative to their positions in the structure of ligand-free CYP3A4.3 We are able to identify the distinct conformational changes of CYP3A4 due to cis-RS MEZ and cis-SR MEZ binding using MD simulations (Figure 6). Compared with the upward shift of the F-G helix and linker region (residues 202−260) in the ketoconazole-binding structure (Figure 6A), leftward shifts are seen in both of the MEZ-enantiomer-binding structures (Figure 6B,C). The relative position of the H helix (residues 271−279) drifts away in the cis-RS MEZ binding structure (Figure 6B), which is different from the positions seen in the ketoconazolebinding structure, cis-SR MEZ binding structure, and ligand-free structure. The relative positions of the C-termini (residues 464−496) of all three ligand-binding structures are similar. Surprisingly, however, a uniquely formed β-sheet in the Cterminus of the cis-RS MEZ binding structure is found. Structural differences found here may directly correlate with the aforementioned fluorescence-intensity changes.

A previous study found that ketoconazole was able to disrupt the phenylalanine-rich hydrophobic cluster in the binding pocket including F108, F213, F215, F219, F220, F241, and F304.2 Figure 7 shows a zoomed-in view of the hydrophobic cluster in different forms of CYP3A4: ligand-free CYP3A4 (Figure 7A), ketoconazole-bound CYP3A4 (Figure 7B), cis-RS MEZ bound CYP3A4 (Figure 7C), and cis-SR MEZ bound CYP3A4 (Figure 7D). We can clearly see that the hydrophobic chlorobenzene ring interacts with F108 in the cis-RS MEZ binding structure and with both F108 and F304 in the cis-SR MEZ binding structure because of the different positionings of the cis-MEZ. This in turn affects the arrangement of the surrounding residues of the binding pocket, resulting in the overall-distinct binding structures of the cis-MEZ enantiomers bound to CYP3A4.

4. CONCLUSION Many studies of chiral chemicals interacting with CYP P450s lacked information on the binding characteristics at the atomic level because of missing information on the absolute configurations of the enantiomers. Here, we successfully used a stepwise-restrained-MD method to model the molecular interactions of MEZ enantiomers with the human CYP3A4 enzyme. There are distinct binding modes for the two isomers to CYP3A4. cis-RS MEZ adopts an extended structure in the CYP3A4 binding pocket, in which the Cl atom is exposed to the solvent. However, cis-SR MEZ sits within the hydrophobic core of the CYP3A4 binding pocket, resulting in a larger freeenergy penalty. Our FEP simulations reveal that the bindingfree-energy difference mainly comes from the VDW interactions, which result from the hydrophobic nature of the binding pocket. Moreover, the binding-free-energy difference from our simulations is in good agreement with the experimental Michaelis−Menten-constant (Km) measurements. The binding specificities of the MEZ isomers result in the difference in their molecular recognition processes toward the human CYP3A4 enzyme. They share the same active site and type II binding. Additionally, they distinctly induce conformational changes in the active site, through which cis-SR MEZ causes an increase in the fluorescence intensity, and cis-RS MEZ causes a decrease in the fluorescence intensity. The RMSDs between the MEZ-binding modes and ketoconazole-binding mode indicate that the overall binding conformations are similar. However, the F-G helix and linker region and the H helix of CYP3A4 show distinct arrangements in the cis-MEZenantiomer-binding structures. Moreover, the residues from G


Article


21621005), the Major Research Plan of the National Natural Science Foundation of China (no. 91643107), and the Fundamental Research Funds for the Central Universities. R.Z. acknowledges financial support from the IBM Blue Gene Science Program.

CYP3A4 that stabilize ligand binding differ for ketoconazole, cis-RS MEZ, and cis-SR MEZ. The binding of MEZ enantioselectively induces conformational changes of CYP3A4, consequently affecting the metabolic activity of CYP3A4 toward the MEZ enantiomers. Compared with their metabolism by CYP3A4, cis-SR inhibits enzymatic activity of CYP3A4 more and thus affects enzyme kinetic behavior more severely, causing reduced metabolic potency of CYP3A4 toward cis-SR. In contrast, the less potent inhibition by cis-RS results in a higher metabolic rate, Vmax, and a smaller Km. Our study reveals how the binding specificity enantioselectively affects the metabolism of MEZ, and it deciphers their molecular recognition of human CYP3A4, providing atomic insights for metabolism-based drug design and relevant screening of chemicals with adverse effects.

■

ABBREVIATIONS

■

REFERENCES

MD, molecular dynamics; MEZ, metconazole; CYP3A4, cytochrome P450 3A4; ECD, electronic circular dichroism; DFT, density-functional theory; EFs, enantiomer fractions; RPA, relative peak area; Km, Michaelis−Menten half-saturation constant; Vmax, maximum velocity of a metabolic reaction; PME, particle-mesh Ewald; FEP, free-energy perturbation; TDDFT, time-dependent density-functional theory; HLM, human-liver microsome; VDW, van der Waals

ASSOCIATED CONTENT

S Supporting Information *

(1) Park, H.; Lee, S.; Suh, J. Structural and Dynamical Basis of Broad Substrate Specificity, Catalytic Mechanism, and Inhibition of Cytochrome P450 3A4. J. Am. Chem. Soc. 2005, 127, 13634−13642. (2) Ekroos, M.; Sjögren, T. Structural Basis for Ligand Promiscuity in Cytochrome P450 3A4. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 13682−13687. (3) Sevrioukova, I. F.; Poulos, T. L. Structure and Mechanism of the Complex between Cytochrome P4503A4 and Ritonavir. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 18422−18427. (4) Guengerich, F. P. Cytochrome P-450 3A4: Regulation and Role in Drug Metabolism. Annu. Rev. Pharmacol. Toxicol. 1999, 39, 1−17. (5) Williams, P. A.; Cosme, J.; Vinković, D. M.; Ward, A.; Angove, H. C.; Day, P. J.; Vonrhein, C.; Tickle, I. J.; Jhoti, H. Crystal Structures of Human Cytochrome P450 3A4 Bound to Metyrapone and Progesterone. Science 2004, 305, 683−686. (6) Krishnamoorthy, N.; Gajendrarao, P.; Thangapandian, S.; Lee, Y.; Lee, K. W. Probing Possible Egress Channels for Multiple Ligands in Human CYP3A4: A Molecular Modeling Study. J. Mol. Model. 2010, 16, 607−614. (7) Fishelovitch, D.; Shaik, S.; Wolfson, H. J.; Nussinov, R. Theoretical Characterization of Substrate Access/Exit Channels in the Human Cytochrome P450 3A4 Enzyme: Involvement of Phenylalanine Residues in the Gating Mechanism. J. Phys. Chem. B 2009, 113, 13018−13025. (8) Mazur, C. S.; Kenneke, J. F. Cross-Species Comparison of Conazole Fungicide Metabolites Using Rat and Rainbow Trout (Onchorhynchus mykiss) Hepatic Microsomes and Purified Human CYP 3A4. Environ. Sci. Technol. 2008, 42, 947−954. (9) Kunze, K. L.; Nelson, W. L.; Kharasch, E. D.; Thummel, K. E.; Isoherranen, N. Stereochemical Aspects of Itraconazole Metabolism in Vitro and in Vivo. Drug Metab. Dispos. 2006, 34, 583−590. (10) Peng, C.-C.; Shi, W.; Lutz, J. D.; Kunze, K. L.; Liu, J. O.; Nelson, W. L.; Isoherranen, N. Stereospecific Metabolism of Itraconazole by CYP3A4: Dioxolane Ring Scission of Azole Antifungals. Drug Metab. Dispos. 2012, 40, 426−435. (11) Lv, X.; Pan, L.; Wang, J.; Lu, L.; Yan, W.; Zhu, Y.; Xu, Y.; Guo, M.; Zhuang, S. Effects of Triazole Fungicides on Androgenic Disruption and CYP3A4 Enzyme Activity. Environ. Pollut. 2017, 222, 504−512. (12) Pan, L.; Lu, L.; Wang, J.; Zheng, C.; Fu, Y.; Xiao, S.; Jin, Y.; Zhuang, S. The Fungicide Difenoconazole Alters mRNA Expression Levels of Human CYP3A4 in HepG2 Cells. Environ. Chem. Lett. 2017, 15, 673−678. (13) Mercadante, R.; Polledri, E.; Scurati, S.; Moretto, A.; Fustinoni, S. Identification and Quantification of Metabolites of the Fungicide Tebuconazole in Human Urine. Chem. Res. Toxicol. 2014, 27, 1943− 1949. (14) Lin, C.-H.; Chou, P.-H.; Chen, P.-J. Two Azole Fungicides (Carcinogenic Triadimefon and Non-carcinogenic Myclobutanil)

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.7b11170. Texts: preparation of the metconazole enantiomers, electronic-circular-dichroism spectroscopy, calculated electronic circular dichroism, HPLC-MS/MS operation, expression and purification of human CYP3A4. Tables: backbone RMSD of CYP3A4 with regard to PDB 2V0M and the N3−Fe distance before and after the stepwiserestrained-molecular-dynamics simulations, parameters for Michaelis−Menten kinetics. Figures: molecular structures of the four isomers of MEZ; structure and configuration assignment of cis-metconazole (cis-MEZ); starting molecular-docking template made from the X-ray crystal structure of CYP3A4 in complex with ketoconazole (PDB ID: 2V0M); Western blot determination of CYP3A4; SDS-PAGE analysis of purified CYP3A4; HPLC separation of the four isomers; metabolism of triadimefon by HLM and the human CYP3A4 enzyme; curve of reaction velocity versus substrate concentrations; degradation rate of the cis-MEZ racemate incubated with the human CYP3A4 enzyme for 0, 30, 60, 90, or 120 min; stereoselective binding characteristics of cis-MEZ to the human CYP3A4 enzyme; backbone RMSD measurements from the equilibration of the CYP3A4−MEZ complexes after the stepwise-restrainedmolecular-dynamics simulations (PDF)

■

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (W.L.). *E-mail: [email protected] (R.Z.). ORCID

Shulin Zhuang: 0000-0002-7774-7239 Joseph A. Morrone: 0000-0001-8550-4854 Weiping Liu: 0000-0002-1173-892X Ruhong Zhou: 0000-0001-8624-5591 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research is supported by grants from the National Natural Science Foundation of China (nos. 21477113, 21427815, and H


Article

The Journal of Physical Chemistry B Exhibit Different Hepatic Cytochrome P450 Activities in Medaka Fish. J. Hazard. Mater. 2014, 277, 150−158. (15) Rösch, A.; Anliker, S.; Hollender, J. How Biotransformation Influences Toxicokinetics of Azole Fungicides in the Aquatic Invertebrate Gammarus Pulex. Environ. Sci. Technol. 2016, 50, 7175−7188. (16) Dong, F.; Li, J.; Chankvetadze, B.; Cheng, Y.; Xu, J.; Liu, X.; Li, Y.; Chen, X.; Bertucci, C.; Tedesco, D.; Zanasi, R.; Zheng, Y. Chiral Triazole Fungicide Difenoconazole: Absolute Stereochemistry, Stereoselective Bioactivity, Aquatic Toxicity, and Environmental Behavior in Vegetables and Soil. Environ. Sci. Technol. 2013, 47, 3386−3394. (17) Shi, W.; Nacev, B. A.; Bhat, S.; Liu, J. O. Impact of Absolute Stereochemistry on the Antiangiogenic and Antifungal Activities of Itraconazole. ACS Med. Chem. Lett. 2010, 1, 155−159. (18) Obach, R. S.; Reed-Hagen, A. E. Measurement of Michaelis Constants for Cytochrome P450-Mediated Biotransformation Reactions Using a Substrate Depletion Approach. Drug Metab. Dispos. 2002, 30, 831−837. (19) Trott, O.; Olson, A. J. AutoDock Vina: Improving the Speed and Accuracy of Docking with A New Scoring Function, Efficient Optimization, and Multithreading. J. Comput. Chem. 2010, 31, 455− 461. (20) Huang, J.; MacKerell, A. D. CHARMM36 All-Atom Additive Protein Force Field: Validation Based on Comparison to NMR Data. J. Comput. Chem. 2013, 34, 2135−2145. (21) Best, R. B.; Zhu, X.; Shim, J.; Lopes, P. E. M.; Mittal, J.; Feig, M.; MacKerell, A. D. Optimization of the Additive CHARMM AllAtom Protein Force Field Targeting Improved Sampling of the Backbone ϕ, ψ and Side-Chain χ1 and χ2 Dihedral Angles. J. Chem. Theory Comput. 2012, 8, 3257−3273. (22) Vanommeslaeghe, K.; Hatcher, E.; Acharya, C.; Kundu, S.; Zhong, S.; Shim, J.; Darian, E.; Guvench, O.; Lopes, P.; Vorobyov, I.; Mackerell, A. D. CHARMM General Force Field: A Force Field for Drug-Like Molecules Compatible with the CHARMM All-Atom Additive Biological Force Fields. J. Comput. Chem. 2010, 31, 671−690. (23) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kalé, L.; Schulten, K. Scalable Molecular Dynamics with NAMD. J. Comput. Chem. 2005, 26, 1781− 1802. (24) Zhou, R.; Das, P.; Royyuru, A. K. Single Mutation Induced H3N2 Hemagglutinin Antibody Neutralization: A Free Energy Perturbation Study. J. Phys. Chem. B 2008, 112, 15813−15820. (25) Xia, Z.; Huynh, T.; Kang, S.-g.; Zhou, R. Free-Energy Simulations Reveal that Both Hydrophobic and Polar Interactions Are Important for Influenza Hemagglutinin Antibody Binding. Biophys. J. 2012, 102, 1453−1461. (26) Furche, F.; Ahlrichs, R.; Wachsmann, C.; Weber, E.; Sobanski, A.; Vö gtle, F.; Grimme, S. Circular Dichroism of Helicenes Investigated by Time-Dependent Density Functional Theory. J. Am. Chem. Soc. 2000, 122, 1717−1724. (27) McCann, D. M.; Stephens, P. J. Determination of Absolute Configuration Using Density Functional Theory Calculations of Optical Rotation and Electronic Circular Dichroism: Chiral Alkenes. J. Org. Chem. 2006, 71, 6074−6098. (28) Cheng, Q.; Sohl, C. D.; Yoshimoto, F. K.; Guengerich, F. P. Oxidation of Dihydrotestosterone by Human Cytochromes P450 19A1 and 3A4. J. Biol. Chem. 2012, 287, 29554−29567. (29) Sevrioukova, I. F.; Poulos, T. L. Structural and Mechanistic Insights into the Interaction of Cytochrome P4503A4 with Bromoergocryptine, a Type I Ligand. J. Biol. Chem. 2012, 287, 3510−3517. (30) Sevrioukova, I. F.; Poulos, T. L. Structural Basis for Regiospecific Midazolam Oxidation by Human Cytochrome P450 3A4. Proc. Natl. Acad. Sci. U. S. A. 2017, 114, 486−491. (31) Zhuang, S.; Lv, X.; Pan, L.; Lu, L.; Ge, Z.; Wang, J.; Wang, J.; Liu, J.; Liu, W.; Zhang, C. Benzotriazole UV 328 and UV-P Showed Distinct Antiandrogenic Activity upon Human CYP3A4-Mediated Biotransformation. Environ. Pollut. 2017, 220, 616−624.

(32) Barton, H. A.; Tang, J.; Sey, Y. M.; Stanko, J. P.; Murrell, R. N.; Rockett, J. C.; Dix, D. J. Metabolism of Myclobutanil and Triadimefon by Human and Rat Cytochrome P450 Enzymes and Liver Microsomes. Xenobiotica 2006, 36, 793−806. (33) Koe, X. F.; Lim, E. L.; Seah, T. C.; Amanah, A.; Wahab, H. A.; Adenan, M. I.; Sulaiman, S. F.; Tan, M. L. Evaluation of in Vitro Cytochrome P450 Induction and Inhibition Activity of Deoxyelephantopin, a Sesquiterpene Lactone from Elephantopus Scaber L. Food Chem. Toxicol. 2013, 60, 98−108. (34) Davydov, D. R.; Rumfeldt, J. A. O.; Sineva, E. V.; Fernando, H.; Davydova, N. Y.; Halpert, J. R. Peripheral Ligand-Binding Site in Cytochrome P450 3A4 Located with Fluorescence Resonance Energy Transfer (FRET). J. Biol. Chem. 2012, 287, 6797−6809. (35) Lampe, J. N.; Atkins, W. M. Time-Resolved Fluorescence Studies of Heterotropic Ligand Binding to Cytochrome P450 3A4. Biochemistry 2006, 45, 12204−12215.

I


Binding Specificity Determines the Cytochrome ... - ACS Publications

Recommend Documents