A combined molecular docking and molecular structure in silico analysis on the substrate and product of leukotriene A4 hydrolase (LTA4H) was performed. The molecular structures of the substrate leukotriene A4 (LTA4) and product leukotirene B4 (LTB4) were studied through density functional theory (DFT) calculations at the B3LYP/6-31 + G(d) level of theory in both gas and condensed phases. The whole LTB4 molecule was divided into three fragments (hydrophobic tail, triene motif, and a polar acidic group) that were subjected to a full conformational study employing the most stable conformations of them to build conformers of the complete molecule and geometry optimize further. LTA4 conformers’ structures were modeled from the LTB4 minimum energy conformers. Both protonated and deprotonated species of LTA4 and LTB4 were analyzed according to pKa values found in the literature. Finally, a binding model of LTA4 with LTA4 hydrolase is proposed according to docking results that show intermolecular interactions that position the protonated and deprotonated ligand in the active site, in excellent agreement with the model suggested from LTA4H-inhibitors crystallographic data.
Leukotrienes (LTs) constitute a family of endogenous metabolites of arachidonic acid that are biosynthesized via the lipoxygenase pathway (1–3). These interesting compounds are a class of lipid mediators involved in the development and maintenance of inflammatory and allergic reactions (4–7). Leukotriene A4 (LTA4), an unstable alkyl epoxide formed from the immediate precursor 5-HPETE via 5-lipoxygenase (5-LO) (8), is converted to leukotriene B4 (5S,12R-dihydroxy-6Z,8E,10E,14Z-eicosatetraenoicacid; LTB4; Figure 1), by stereoselective hydratation of LTA4 hydrolase (LTA4H) (9). LTB4 is a potent pro-inflammatory mediator implicated in the pathogenesis of a number of diseases including inflammatory bowel disease (IBD), psoriasis, rheumatoid arthritis, and asthma and plays an important role in immunological responses, chronic obstructive pulmonary disease (COPD), and atherosclerosis (4,6,10–12).
On the other hand, LTA4H is a bifunctional zinc metalloenzyme that catalyzes the rate-limiting step in the production of LTB4 (13). The X-ray crystal structures of LTA4H in complex with different inhibitors have been obtained by several authors (14–16). According to several investigations, the metal site is located next to the putative active site bound to His295, His299, and Glu318 (14–18).
In this study, we show an analysis of the different interactions of LTA4 into LTA4H through molecular docking study. Molecular and electronic structure information of LTA4 and LTB4, obtained from an exhaustive conformational analysis, was used to set an initial conformation for the docking study.
There are few theoretical studies about the molecular structure of LTB4. One of them was performed by Brasseur et al. (19) using semiempirical methods. They present a computational description of the conformation of a pair of two isomeric molecules (6-cis and 6-trans-leukotriene B4) forming a complex with one calcium ion. Recently, Catoire et al. (20) have obtained a highly constrained seahorse conformation when it is bound to leukotriene receptor BLT2 by 1H-RMN. However, there has been no ab initio and/or DFT studies that analyze the conformation and electronic structure of LTA4 and LTB4.
Therefore, we believe that results of this work are of great contribution especially to elucidate: (i) the structural similitude between LTB4 and its precursor LTA4, (ii) the conformational changes of LTA4 in gas phase and condensed phase and when it is bounded to the enzyme, (iii) the active site of LTA4H where LTA4 binds and compare it with the interaction site of inhibitors (11–13) and thus, to obtain important information for future development of new inhibitors of LTB4 formation.
Material and Methods
Relaxed potential energy curves (PECs) were performed at HF/3-21G, HF/6-31 + G(d) and B3LYP/6-31 + G(d) levels of theory (21–23), while potential energy exploratory surfaces (PESs) were carried out at HF/3-21G level of theory. The conformers were then optimized at a higher level, and frequency calculations were carried out to confirm minimum energy conformers checking the absence of imaginary frequencies. Solvent effect was evaluated through geometry optimizations using the integral equation formalism polarizable continuum model (IEF-PCM) (24,25) at B3LYP/6-31 + G(d) level of theory. All calculations were performed using the gaussian 03 computational program (26).
Docking calculations were performed using autodock vina software (27) to propose a model of the LTA4-LTA4H complex. The crystal structure of LTA4H used in our study was obtained from Protein Data Bank (28) [PDB accession code: 3CHO (16)]. The original ligand, 2-amino-N-[4-(phenylmethoxy)phenyl]-acetamide, and water molecules in the crystal structure were removed. LTA4 initial conformation was obtained from the conformational study. A torsional restriction was applied for single bonds of the conjugated system of the triene motif fragment setting them as fixed, that is, not able to rotate. The box was defined to include completely the proposed binding site, and standard parameters for the docking calculation were used except for exhaustiveness for which a value of 100 was used.
The lowest binding energy docking complex was analyzed with NCIPlot program to identify and map non-covalent interactions using the promolecular densities and its derivatives. These interactions are non-local and manifest in real space as low-gradient isosurfaces with low densities, which are interpreted and colored according to the corresponding values of sign(λ2)ρ (29).
Results and Discussions
Conformational study for LTB4
To obtain the structure of LTA4 for the docking study, we previously performed a conformational analysis on LTB4, which optimized structure was used to propose the initial conformation of LTA4 for the molecular docking study.
Several papers agree that LTB4 is a cis-trans (or E-Z) isomers (5S,12R-dihydroxy-6Z,8E,10E,14Z-eicosatetraenoicacid) (19,20,30) and that LTB4 shows conformational restrictions in the double bonds configuration of LTB4 (6Z,8E,10E,14Z). Despite the restrictions of double bonds, the molecule of LTB4 has 13 rotational bonds, and if these dihedrals were affected by systematic 120° changes, more than 1.500 × 103 conformations could be obtained. To avoid this large number of possible conformations, a systematic analysis was realized in a stepwise manner on three different important fragments of the LTB4 easily recognizable: (i) hydrophobic tail (ii) a triene motif, and (iii) a polar group (Figure 2). Our systematic study includes a severe analysis of each fragment that was selected based on its chemical characteristics. Molecular fragmentation has been used in previous work of our group given excellent results in the study of large molecules (31,32).
This fragment is composed by a chain of methylene units, which is bound to the triene fragment through a double bond. Conformational study of this fragment consisted in the analysis of the four dihedral angles presented in Figure 2A. Rotating dihedrals φ1 and φ4 two potential energy curves were obtained in Figure 3A,B. The plots present the global energy minimum around 180° for φ1 and two local minima next to 120° and −120° with the same energy for φ4. Potential energy exploratory surfaces of the type E = f(φ2, φ3) shows a global minimum with dihedral values of about φ2 =180° and φ3 =180°. This is show in Supporting Information, Figure S1.
We have found that the hydrophobic tail fragment has high symmetry and an elongated minimum energy conformation with dihedral values next to 180° for φ1, φ2, φ3, and 120° or −120° for φ4. However, the low rotational barriers (<3 kcal/mol) indicate that this fragment has a high conformational flexibility.
Although the conjugated triene system confers the fragment a high rigidity, there are dihedral angles that deserve to be analyzed: φ5, φ6 and φ7 (Figure 2). In contrast to these dihedrals angles, the values of dihedral φ8 depend largely on the fragment with the acidic function. So, we believe that the analysis of this dihedral has no practical meaning at this point.
Dihedral angle φ5 was studied through a PEC (Figure 3C). The global energy minimum has the double bond in opposite position with respect to the OH group, next to 120°, where no interaction occurs, but probably torsional tension is relieved. Potential energy exploratory surfaces of the type E = f(φ6, φ7) shown in Figure S1 presents the global minimum on the region of φ6 ≈180° and φ7 ≈−120°. However, there are local minima with rotational barriers of about 3 kcal/mol and energy values close to global energy minimum. The low rotational barrier confers a relative flexibility that would only be restrained in the complete LTB4 molecule by steric repulsion.
Polar (acidic) group
This fragment is characterized by an OH group over a quiral carbon and a carboxylic group, both separated by three methylene groups. Owing to the methylene chain, the fragment presents a high flexibility in the dihedral angles φ9, φ10, φ11 and φ12. A similar analysis to the others fragment was performed to the dihedral angles of the present fragment, briefly: PECs were made for dihedrals φ9 and φ12, while dihedrals φ10 and φ11 were analyzed through PES.
In Figure 3D is identified the zone of the highest energy (φ9 ≈0°, repulsion zone) and zones of minimum energy (φ9 ≈180°, global minimum energy, and φ9 ≈60°, ΔE = 0.1 kcal/mol). Potential energy exploratory surfaces obtained from φ10 and φ11, plotted in Figure S1, shows that the global energy minimum is localized next to 60° for both φ10 and φ11. The low rotational barriers between any of the two minima indicate the high flexibility of the fragment.
Furthermore, rotation of dihedrals φ12 and φ13 produces a series of conformers with intramolecular hydrogen bond interactions. Consequently, a NBO analysis was performed on these conformations. In addition, taking into consideration the values of pKa for leukotrienes range from 3 to 5 (33,34), deprotonated species should also be studied. Table 1 presents the three potential conformations for acidic group in both species obtained from DFT calculations. The total energy indicates that both species of the first conformer are the most stable conformations. The interaction energy E(2) between donor and acceptor orbitals shows an effective interaction between the oxygen lone pairs (LP) and the sigma antibonding orbitals (σ*) of O–H bond, which is higher in the deprotonated species (no charge transfer interaction was found in the last conformer for the anionic species). Although the third acidic conformer presents higher charge transfer stabilization energy, steric restraints could disfavor this conformation.
Table 1. Conformational and electronic data obtained by rotation of dihedrals φ12 and φ13 for polar fragment. Dihedrals are in degrees
Molecular structure for LTB4
Using the minimum energy conformers of the three fragments, we proposed an initial molecular structure for LTB4. However, its stable conformation could be the same or different to the structure proposed. Figure 4 shows the superposition of the most stable conformer and molecular structure obtained from the fragmentary study. The principal difference of both structures is the values of dihedral angle φ8. The value proposed for φ8 in the initial structure was of 180° (Figure 4, in dark gray), while full optimized value is −125° (Figure 4, in light gray), giving rise to a ΔE between both structures of about 35 kcal/mol. This result is expected because dihedral φ8 was not analyzed in the fragmentary analysis. Consequently, dihedral value was modified when the whole molecule was fully optimized.
Molecular structure for LTA4
LTB4 shows a high structural similarity to its precursor the leukotriene A4. The most important structural difference resides in the presence of an epoxy group in LTA4 molecule. This epoxy group is hydrolyzed and converted into an OH group in the enzymatic step that involves the formation of LTB4 (13–15,17,18).
To obtain a stable conformation for LTA4, which can be used in molecular docking study, we suggested an initial conformation using the dihedral values of LTB4. Hydrophobic tail dihedral initial values were taken from the most stable conformation of LTB4. In addition, literature suggests the 7E, 9E, 11Z, 14Z configurations for the unsaturated part of LTA4 (8). However, dihedrals θ1 and θ2 of LTA4 (Figure 5), are involved in the orientation of the epoxy group that plays a key role in enzyme binding. Consequently, to perform a complete analysis, these important dihedrals were studied.
A full optimization carried out for the four possible conformers gave low energy differences between all conformers (Figure S2). Solvent effect analysis indicates that the stable conformations obtained in vacuo are not altered by the presence of the solvent (Table 2).
Table 2. Energy and structural data of LTA4 conformers. Dihedrals are in degrees
HF/6-31 + G(d)
B3LYP/6-31 + G(d)
IEF-PCM/B3LYP/6-31 + G(d)
A molecular docking study was performed to propose a binding model and interpret the different interactions between the ligand and the enzyme. Although the reaction mechanism proposed by Thunnissen et al. (15) involves the deprotonated species of LTA4, both the acidic and anionic species of LTA4 were subjected to docking studies.
Nine possible binding complexes were obtained for the anionic species of LTA4. The most stable one, that is, the one with lowest ΔG of binding (about −9.5 kcal/mol), is shown in Figure 6. The model reveals that the carboxylate group interacts with Arg563 and is close to Lys565; the epoxide is positioned with the oxygen atom coordinating with Zn2+ cation and forming a square-based pyramidal metal complex with His 295, His 299, and Glu 318 (15); the hydrophobic tail is buried in a hydrophobic pocket; and Asp375 is oriented toward the triene group where it would control the position of a water molecule and, therefore, the stereospecific insertion of the 12R-hydroxyl group of LTB4 as is proposed in the literature according to the reaction mechanism presented (14,15,17,18).
The most stable docking complex of the enzyme and the acidic species of LTA4 presents the substrate located in the same cavity and with almost the same orientation that was found for the anionic one (Figure 6). Moreover, ΔG of binding is only 0.1 kcal/mol higher.
Figure 6 also shows a complex web of weak non-covalent interactions between the substrate and the active site of the enzyme (including the Zn2+ cation) where van der Waals and hydrophobic interactions dominate by far (light green surfaces). Moreover, interaction analysis reveals a hydrogen bond between the carboxylate group of LTA4 and Arg563. In the proposed binding model, LTA4 does not interact with Lys565. However, we expect that the dynamic behavior of both the substrate and the enzyme leads to the formation of a hydrogen bond (or more probably a salt bridge) between LTA4 and the positively charged residue Lys565 because it is close to Arg563 in the binding site.
LTA4 minimum energy conformation from quantum analysis shows differences with enzyme-bounded conformation (Figure 7). However, it is well known that flexible molecules change their conformation upon binding to a protein (35); thus, it is not surprising in our finding if we take into account the high degree of flexibility of LTA4 molecule.
We obtained from conformational analysis the stable structures for the substrate and product of the enzyme LTA4H using Hartree Fock and DFT calculations. Most stable conformation of LTB4 shows an elongated structure with an intramolecular interaction in the polar fragment analyzed through NBO analysis. Based on the structural data of LTB4, four possible conformations of LTA4 were proposed. We found that conformer IV is the most stable in gas and solvated phases with a ΔE of 1 kcal/mol less than the other conformers.
Molecular docking results provided an interaction model of LTA4 with LTA4H in excellent correlation with the information that has been inferred from enzyme-inhibitors crystal structures (14,15): The substrate fits in the active site of the enzyme arranging its functional groups according to the reaction mechanism proposed by several authors (14,17,18). Both species of LTA4, protonated and deprotonated, bind to the same site with proper orientation of the key groups for the enzymatic reaction. The substrate is held in the active site by multiple weak attractive interactions along its surface and a hydrogen bond to Arg563.
Thus, the in silico model of LTA4-LTA4H complex obtained from molecular docking in combination with a conformation analysis of substrate and product provides insights into the development of new and most efficient inhibitors of LTB4 formation.
This work was supported by Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) project PIP11220100100151 and Universidad Nacional de San Luis (UNSL).