Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 115 (41), 11248-55

Membrane Position of Ibuprofen Agrees With Suggested Access Path Entrance to Cytochrome P450 2C9 Active Site

Affiliations

Membrane Position of Ibuprofen Agrees With Suggested Access Path Entrance to Cytochrome P450 2C9 Active Site

Karel Berka et al. J Phys Chem A.

Abstract

Cytochrome P450 2C9 (CYP2C9) is a membrane-anchored human microsomal protein involved in the drug metabolism in liver. CYP2C9 consists of an N-terminal transmembrane anchor and a catalytic cytoplasmic domain. While the structure of the catalytic domain is well-known from X-ray experiments, the complete structure and its incorporation into the membrane remains unsolved. We constructed an atomistic model of complete CYP2C9 in a dioleoylphosphatidylcholine membrane and evolved it by molecular dynamics simulations in explicit water on a 100+ ns time-scale. The model agrees well with known experimental data about membrane positioning of cytochromes P450. The entry to the substrate access channel is proposed to be facing the membrane interior while the exit of the product egress channel is situated above the interface pointing toward the water phase. The positions of openings of the substrate access and product egress channels correspond to free energy minima of CYP2C9 substrate ibuprofen and its metabolite in the membrane, respectively.

Figures

Figure 1
Figure 1
In this work we discuss several structural aspects of CYP2C9 anchored to the membrane at atomic resolution. The scheme shows that it is hypothesized that hydrophobic substrates (S) enter the CYP enzyme active site (AS) via an access channel (ACh). The substrate is oxidized in the active site to a more hydrophilic product (P) and leaves the active site via an egress channel (ECh) to cytosol. The oxidation involves heme (HEME) and requires electrons, which can be provided by cytochrome P450 reductase (CPR).
Figure 2
Figure 2
Snapshot of the structure of CYP2C9 in DOPC membrane taken at 50 ns. The protein is shown in green, while parts determined by epitope screening as not being exposed to water (i.e., in the membrane or inside the protein core) are shown in violet. Parts predicted to be accessible to water are shown in blue. Heme is shown by red spheres. The DOPC membrane is shown using the transparent stick representation with orange spheres at the positions of phosphorus atoms in the lipid beads. The image was prepared with Pymol 0.99rc6..
Figure 3
Figure 3
Left panel shows B-factors mapped onto the structure calculated from individual 100 ns long simulations of CYP2C9 in membrane. The B-factors are colored by spectrum from dark blue for rigid regions to red for the most flexible regions. Heme is shown in magenta and membrane lipids are shown in gray with orange spheres for phosphoric atoms. Right panel shows RMSF per residue for simulations of wt CYP2C9 in membrane (black line), wt CYP2C9 in water (green line), CYP2C9 based on crystal structure 1OG2 (without N-terminal anchor) in water (red line), as well as RMSF calculated from the B-factors of crystals structures 1OG2 and 1OG5 (cyan and blue lines, respectively). The figure shows that the most flexible parts of the molecule are the N-terminal anchor (FR1), G/H loop (FR5), K/L loop (FR9), and C-terminus (FR10). Interestingly, the F/G loop, which is flexible in some CYPs, is rather rigid. Also, the central part of the protein close to the catalytic site is relatively rigid. The simulation of CYP2C9 in water shows highly enhanced flexibility of the N-terminal part (FR1) and of the F/G-loop (FR6) in comparison with simulation of CYP2C9 anchored to the membrane. The profiles of RMSF calculated from all simulations have similar trends, which also agree well with published data..
Figure 4
Figure 4
Proximal side of CYP2C9 anchored to DOPC membrane is accessible to water and CYP reductase (CPR). These two views of CYP2C9 show amino acids responsible for binding to CPR (K121, R125, R132, F134, M136, K138, K432, and G442 shown in orange and heme shown in red). These amino acids are present on the CYP proximal side surface. CPR-binding amino acids are accessible to water in our model.
Figure 5
Figure 5
Comparison of models of embedding the CYP2C9 protein into the membrane. Planes of membrane upper layer (approximately at the glycerol level) are colored as follows: This work (orange), Williams et al. (yellow), Wade et al. (cyan), Zhao et al. (red), Lomize et al. PPM 2.0 protein membrane positioning approach for 1OG5 from OPM database http://opm.phar.umich.edu/(23) (blue), and the same PPM 2.0 approach applied to on our wt CYP2C9 model with the N-terminal anchor present (white). As the OPM database shows the planes of positions for the hydrophobic slab, PPM planes were shifted by 10 Å upward to be consistent with other planes at the glycerol level.
Figure 6
Figure 6
Positions of access/egress channels within membrane anchored CYP2C9. Left panel shows the model of CYP2C9 in DOPC membrane. Protein is shown in dark gray cartoon representation. Heme is shown by red balls. DOPC membrane is shown in wires. The most widely open active site access channel is solvent channel (blue). The mouth opening of the solvent channel is slightly above the membrane-water interface. Channels from 2x family are shown in an order following their opening during the simulation (see Figure S5 in Supporting Information) as follows: 2c (green), 2ac (cyan), 2b (yellow), 2e (magenta), 2f (orange), and 2a (red). All of the openings of these channels are pointing into the membrane; however, the 2b and 2c channel openings are pointing into the headgroup region interface. Access channels were analyzed by MOLE with Pymol 0.99rc6. Right panel shows schematical description of the positions of above-mentioned channels of our model.
Figure 7
Figure 7
Positions of CYP2C9 substrate ibuprofen and its metabolite on membrane. The upper left part shows partial densities of selected groups on the DOPC membrane/water boundary, while the lower part shows free energy profiles for ibuprofen and 3-hydroxyibuprofen. Free energies were calculated by the potential of mean force imposed on the distance in the z-axis between the center-of-mass of the drug and membrane. Zero free energy was arbitrarily selected in the distance of 3.5 nm from the center of the membrane (i.e., in water). The free energy profile for uncharged ibuprofen was shifted by the free energy calculated from the ibuprofen acidity constant in water and for pH 7 by 3.5 kcal/mol (14.6 kJ/mol). The scheme of protonated ibuprofen highlighted with an arrow shows the oxygenation site at position 3 and approximate positions of the opening for 2x and solvent channels are shown by magenta and blues arrows, respectively. The structures on the right side were selected from the window taken from umbrella sampling with minimal free energy. Note that while neutral ibuprofen is in the center of the membrane without any contact with the water environment, both charged molecules are in contact with water molecules mediated by their charged carboxylic group (shown in red). The more polar 3-hydroxyibuprofen is however more exposed to water by its hydroxyl group as shown by black arrow. The figure was prepared in Pymol 0.99rc6..
Figure 8
Figure 8
Proposed mechanism of CYP2C9 substrate access and product release. Ibuprofen is accumulated in the membrane, from which it can be effectively “hoovered” by access channels leading to the CYP active site, and following the enzymatic reaction the resulting metabolite may be released via egress channels leading to the cytosol.

Similar articles

See all similar articles

Cited by 45 PubMed Central articles

See all "Cited by" articles

References

    1. Ortiz de Montellano P. R.. Cytochrome P450: Structure, Mechanism, and Biochemistry, 3rd ed.; Kluwer Academic/Plenum Publishers: New York, 2005.
    1. Anzenbacher P.; Anzenbacherova E. Cell. Mol. Life Sci. 2001, 58 (5–6), 737–47. - PubMed
    1. Coon M. J. Annu. Rev. Pharmacol. Toxicol. 2005, 45, 1–25. - PubMed
    1. Graham S. E.; Peterson J. A. Arch. Biochem. Biophys. 1999, 369 (1), 24–9. - PubMed
    1. Cojocaru V.; Winn P. J.; Wade R. C. Biochim. Biophys. Acta 2007, 1770 (3), 390–401. - PubMed

Publication types

Feedback