Modeling the transition state structure to probe a reaction mechanism on the oxidation of quinoline by quinoline 2-oxidoreductase
© The Author(s) 2016
Received: 20 May 2016
Accepted: 10 November 2016
Published: 24 November 2016
Quinoline 2-oxidoreductase (Qor) is a member of molybdenum hydroxylase which catalyzes the oxidation of quinoline (2, 3 benzopyridine) to 1-hydro-2-oxoquinoline. Qor has biological and medicinal significances. Qor is known to metabolize drugs produced from quinoline for the treatment of malaria, arthritis, and lupus for many years. However, the mechanistic action by which Qor oxidizes quinoline has not been investigated either experimentally or theoretically.
Purpose of the study
The present study was intended to determine the interaction site of quinoline, predict the transition state structure, and probe a plausible mechanistic route for the oxidative hydroxylation of quinoline in the reductive half-reaction active site of Qor.
Density functional theory calculations have been carried out in order to understand the events taking place during the oxidative hydroxylation of quinoline in the reductive half-reaction active site of Qor. The most electropositivity and the lowest percentage contribution to the HOMO are shown at C2 of quinoline compared to the other carbon atoms. The transition state structure of quinoline bound to the active site has been confirmed by one imaginary negative frequency of −104.500/s and −1.2365899E+06 transition state energies. The Muliken atomic charges, the bond distances, and the bond order profiles were determined to characterize the transition state structure and the reaction mechanism.
The results have shown that C2 is the preferred locus of interaction of quinoline to interact with the active site of Qor. The transition state structure of quinoline bound to the active site has been confirmed by one imaginary negative frequency. Moreover, the presence of partial negative charges on hydrogen at the transitions state suggested hydride transfer. Similarly, results obtained from total energy, iconicity and molecular orbital analyses supported a concerted reaction mechanism.
Quinoline 2-oxidoreductase is a member of molybdenum hydroxylases with a known three dimensional structure . It catalyzes the oxidative hydroxylation of quinoline (2, 3 benzopyridine) to 1-hydro-2-oxoquinoline. Qor is known to oxidatively hydroxylate carbon atoms of heterocyclic aromatic compounds, particularly quinoline and its derivatives. For instance, it catalyzes the first two steps in the degradation of quinoline in bacteria (Comamonas testosteroni 63) . Quinoline derivatives have been used in the treatments of malaria, arthritis, and lupus for many years . They are also used as a sole source of energy in bacteria , hepatocarcinogen in mice and rats, and several quinoline derivatives are mutagens . However, quinoline derivatives are known to represent one of the most successfully used classes of drugs, their therapeutic action is still not well understood. Remarkably, there is no clear catalytic mechanism known for the therapy of action of quinoline drugs . Therefore, the catalytic mechanism of Qor needs to be investigated in order to improve the use of quinoline in the drug design process.
Qor catalyzes similar substrates with the enzyme Xanthine oxidoreductase (XOR) . Quinoline, physiological substrates of Qor, and xanthine, physiological substrates of XOR, share some common features such as both are an aromatic compounds with two ring systems. Moreover, Qor and XOR are the members of molybdenum hydroxylases particularly xanthine oxidase family enzymes and hence basically they have similar redox active centers [7, 8]. For this reason the catalytic mechanisms of Qor is expected to be studied on the basis of the catalytic mechanisms of XOR . XOR from bovine milk is the most studied members of molybdenum hydroxylase. Consequently, it can be used as a bench mark to study the entire members of Mo hydroxylase such as Qor . Based on the currently accepted catalytic mechanisms of XOR , the catalytic mechanism of Qor is proposed in the study.
The reaction mechanism is proposed to begin with the abstraction of the equatorial hydroxyl proton by the amino acid residue (Glu743). The neucleophile, oxy-anion of the hydroxyl group, attacks the electron deficient carbon center of the substrate and provides a tetrahedral species (tetrahedral intermediate or transition state). At the transition state hydrogen is transferred from the substrate carbon to the sulfido terminal of the active site . However, it not known whether oxidative hydroxylation of quinoline catalyzed by Qor is concerted or stepwise. In addition to that the mechanism of a catalytic reaction can be characterized in terms of the chemical events that take place during the reaction . However, several events that are expected to occur during the oxidation of quinoline such as formation of a bond between the equatorial oxygen and the quinoline carbon, cleavage of quinoline carbon-hydrogen bond, migration of hydrogen from quinoline carbon to the sulfido terminal of the active site, and conversion of quinoline to 1-hydro-2-oxoquinoline were neither known nor described. Moreover the nature of hydrogen transfer from the substrate carbon to the sulfido terminal of Qor is not known.
A density functional theory approach was designed to perform electronic structure calculations in order to investigate the catalytic mechanism and describe the events those are expected to take place during the catalytic oxidative hydroxylation of quinoline by Qor. The calculations were performed on the truncated active site model compound bound to quinoline. From the optimized structures several data such as total energies, Mulliken atomic charges, bond distance, bond order indices, and percentage contributions of the chemical constituents to the molecular orbitals were generated. These data were used to determine the interaction site of quinoline, model the transition state structure, and probe a plausible mechanistic route for the oxidative hydroxylation of quinoline in the reductive half-reaction active site of Quinoline 2-oxidoreductase.
The electronic structure calculations were performed with density functional theory method on the Gaussian® 03 W (version 6.0) program software package (Gaussian, Inc., Wallingford, CT, USA) . The DFT method employing the B3LYP level of theory  was applied on the model structures derived from the initial geometries of the crystal structures of Qor . The optimizations were carried out using the mixed basis set LANL2DZ for Mo which contains core potential (LanL2), and 6–31G (d1–p1) basis set for C, N, O and S .
The substrate quinoline and quinoline bound to the truncated reductive half-reaction active site of Qor at C2 and C4 position of quinoline were optimized in order to identify the interaction site of quinoline. The transition state structure was determined for the migration of substrate bound (HRH) from the substrate carbon (CRH) to the sulfido terminal (SMo). The linear transit scans were performed on the structure shown on Fig. 2.
The transition state structure was located by the presence of one imaginary negative frequency . The geometries from single point energy calculations were used for AOMix molecular analysis using AOMix 2011/2012 (reversion 6.6) software programs [17, 18]. The total energies and the Muliken atomic charges were generated from the optimized geometries of single point energy calculations. The total energies were normalized in order to profile the reaction coordinates.
Results and discussion
Probing the interaction site of quinoline
Moreover, the total energies obtained from optimization for C2-quinoline or C4-quinoline bound to the active site (Mo(+VI)–Oeq–C2–quinoline or Mo(+VI)–Oeq–C4–quinoline, respectively) are (−1.23661074E+06) and (−1.23661438E+06)kcal/mol, respectively. These results clearly show that the active site bound at C2 position of quinoline is destabilized by 3.64 kcal/mol relative to the active site bound at C4 position of quinoline. This indicates that the active site bound at C2 position of quinoline exhibits lower energy barrier to enter the transition state compared to the active site bound at C4 position of quinoline.
Therefore, the data from Mulliken atomic charge profile, % contribution on HOMO, and total energies are in favor of C2-pyridine as the preferred interaction site for quinoline. The result is consistent with the previous findings that quinoline becomes hydroxylated at C2 atom of the heterocyclic nitrogen containing ring .
Prediction and characterization of transition state structure
Mulliken atomic charges for selected elements from linear transit scan calculations
The Mulliken atomic charges on Mo are 0.616, 0.584, and 0.414 respectively, for the substrate bound intermediate, transition state, and product bound intermediate. This reflects a decrease in the partial positive charge on Mo ion as HRH migrates from CRH to SMo. The decrease in charge on Mo indicates the development of negatively charged particles on it. This is consistent with the reduction of Mo as the substrate bound active site (Mo(+VI)) is converted to the product bound active site (Mo(+IV)). Unlike Mo ion, the Mulliken atomic charge on substrate carbon (CRH) was shown to increase as HRH migrates from CRH to SMo. The charges on CRH are 0.084, 0.198, and 0.330, respectively, at the substrate bound intermediate, transition state, and product bound intermediate. The profile reveals that the partial positive charges on CRH was shown to increase by a factor of two as HRH moves from the substrate bound carbon to the transition states and further increased by 66.3% as HRH moves to the product bound intermediate. The increase in partial positive charge on CRH is due to the partial transfer of electrons away from it. The increase and decrease in the partial negative charges on Mo and CRH is consistent with the assumption that Mo is reduced from (Mo(+VI)) to (Mo(+IV)) in the course of the reaction, due to the transfer of electrons from CRH the molybdenum center. Although the changes in magnitude are not comparable, the charge on the equatorial oxygen (Oeq) shows the same trend as CRH. The atomic charge values on Oeq are −0.576, −0.544, and −0.469 when HRH is at the substrate bound carbon, transition state, and product bound sulfido terminal, respectively. The decrease in the partial negatively charged particles on Oeq might be due to the increase in the attraction of bonding electrons (Oeq–CRH) by CRH. On the other hand, the electropositivity of the substrate hydrogen (HRH) decreases as it moves from the substrate bound carbon to the product bound sulfido terminal. This indicates that the accumulation of negatively charged particles, on HRH, is high when it is found at the sulfido terminal compared to the substrate bound. Unlike all the other inorganic ligands coordinated to Mo, the atomic charge distribution on the apical oxygen shows no more significant variation as HRH moves from CRH to SMo. As a result, it can be reasonably concluded that the apical oxo plays a “spectator” role in the reaction. In previous works, it was reported that the apical oxo may play an important role in the stabilization of the intermediate states of the catalytic cycle by increasing the Mo = O strength by “spectator oxo effect” though it is not directly participated in catalysis . The charge distribution on HRH at CRH–HRH, TS, and SMo–HRH are 0.142, 0.048, and 0.041, respectively. This result shows that the electropositivity of HRH is decreased by 66.3% as HRH move from CRH to the transition state and further decreased by 76.1% at SMo compared to the transition state. The rapid decrease in electropositivity or rapid increase in electronegativity of HRH, as it migrates from CRH to SMo, is due to the development of partial positive charges on HRH. This result supported hydride transfer from CRH to SMo which is consistent with recent investigations . The partial negative charge distributions on the sulfido terminal (SMo) are −0.626, −0.444, and −0.391 as HRH is found at CRH, transition state, and SMo in the respective order. This result shows the increase in the electropositivity of SMo as HRH moves from CRH to SMo itself. This might be due to the transfer of partial negatively charged electrons from the π-type electrons between apical oxygen and molybdenum (Mo = O) to the empty dxy orbitals of Mo. Finally, the atomic charge distributions on the dithiolene sulfurs slightly increase as HRH moves from CRH to SMo. The result shows that the partial negatively charged particles are increased by 0.019 and 0.040 for Sα and Sβ, respectively. The increase in electronegativity might be due to the back donation of electrons from the dxy orbital’s of Mo to the pz orbitals of the dithiolene sulfur atoms. It implies that electrons from the Mo center passes to the other redox centers through the dithiolene sulfurs. The change in electronegativity of Sβ is higher than Sα by 0.021. Sβ is at about 150.134° angle from the equatorial oxygen which implies that Sβ is almost trance to the equatorial oxygen. For this reason, Sβ, which carried the partial negatively charged particles, would have a trance effect on the equatorial oxygen which is a leaving group in the course of the reaction.
In summary, results obtained from the bond lengths possibly predicts that the events which are proposed to takes place at the transition state such as bond formation (CRH–Oeq and SMo–HRH) and bond cleavage (Mo–Oeq and CRH–HRH) inherit the characteristics of the substrate bound. Moreover, the lengthening of bond lengths predicts the cleavage of Mo–Oeq and CRH–HRH bonds while the shortening of bond lengths predicts the formation of CRH–Oeq and CRH–HRH bonds during the oxidative hydroxylation of quinoline in the reductive half-reaction active site of Qor.
The percentage contribution of the molecular orbital fragments (Modxy) to the highest occupied molecular orbitals (HOMOs) of Qor at the substrate bound CRH-HRH, transition state, and SMo–HRH are 2.17, 21.67 and 80.57, respectively. The result shows that the metallic character increase as HRH moves from C2 of quinoline to SMo. The increase in metallic character depicts that electrons are transferred from C2 of quinoline to the Mo center and hence the reduction of Mo(+VI) to Mo(+IV) during the oxidative hydroxylation of quinoline by Qor.
Probing a reaction mechanism for the oxidation of quinoline
After the transition state structure was located, various geometries (Scheme 2) were optimized in order to understand the events which take place during the catalytic conversion of quinoline to 1-hydro-2-oxoquinoline and probe a plausible mechanistic route for the oxidative hydroxylation of quinoline in the reductive half-reaction active site of Qor. In this reaction mechanism the equatorial oxygen is proposed to nucleophilically attack the electron deficient carbon (C2) to form structure (b) after the deprotonation of the equatorial hydroxyl group of the active site by Glu713. The possible inorganic ligands that might be considered for the nucleophilic attack on C2 of quinoline are the equatorial oxo (Oeq), apical oxo (Ooxo) and sulfido terminal (SMo).
The Mulliken atomic charges for selected elements from geometry optimization for the structures shown in Scheme 2
It is already described that the reaction mechanism can be proceed through nucleophilic attack by the equatorial oxygen on C2 and hydride transfer is taking place during the oxidative hydroxylation of quinoline. But, further description is requited whether the reaction mechanism is concerted or stepwise process.
In addition to that, there is no significant change in the percentage contribution of Modxy to the HOMO as structure (a, 2.96) is converted to (b, 2.11). On the contrary, the conversion of structures (a) to (c, 20.96) or (c) to (d, 80.54) is takes placed with dramatic increase in the percentage contribution of Modxy to the HOMO which assures the inexistence of structure (b) in the reaction mechanism. Similarly, the HOMOs in Fig. 10 show that there is no significant change in the electron densities distribution between structures (a) and (b). If structure (b) is existed in the reaction mechanism, there should be a change in the electrons densities distribution from structures (a) to (b) as the change shown from structures (a) to (c) and structures (c) to (d) in Fig. 10.
Once again, this result predicts that structure (b) is not existed in the reaction mechanism. Consequently the nucleophilic attack on the substrate carbon by the equatorial oxygen and the hydride transfer from the substrate carbon to the sulfido terminal of the active site are proposed to be concerted for the oxidative hydroxylation reaction mechanism of quinoline in the active site of Qor. This finding is consistent with theoretical and isotopic experimental results that a concerted (one step) mechanism by the deprotonated active site is the most plausible for reactions catalyzed by molybdenum hydroxylases .
Moreover, CRH–Oeq and CRH–HRH bond lengths are changed from 1.452 to 3.137 and 1.201 to 1.091, respectively as HRH migrates from the substrate bound [structure (b)] to the transition state (TS-c). This result indicates that the formation of CRH–Oeq bond is much higher (about 15 times) than the cleavage of CRH–HRH bond. It implies that nucleophilic attack (CRH–Oeq) is faster than hydride transfer (CRH–HRH). Hence, hydride transfer is the rate limiting step in the catalysis stage of the oxidative hydroxylation of quinoline in the reductive half-reaction active site of Qor. This result is consistent with previous findings that hydride transfer is the rate determining step in the concerted reaction mechanism unlike the stepwise mechanism in which the nucleophilic attack is the rate determining step .
After the product bound [structure (d)] is formed, it is further dissociated into various structures either through one or two electron transfer process to give the most stable product [structure (g)]. There are four possible paths (I, II, III and IV) for the dissociation of structure (d) into structure g (Fig. 9). Path (III) [(a), (c), (d), (f), and (g)] and path (IV) [(a), (c), (d), (e), (f), and (g)] are passed through the complex (f) which has 65.436 kcal/mol energy barrier from the transition state. Hence, path (III) and (IV) can be ruled out due to the highest energy barrier relative to path (I) and (II). Path (II) [(a), (c), (d) and (g)] has 39.801 kcal/mol energy barrier between the transition state [structure (c)] and the product bound [structure (d)]. On the other hand Path (I) [(a), (c), and (g)] is passed through the transition state and directly converted to the product (structure g). Due to this higher energy barrier (39.801 kcal/mol) relative to path (I), the reaction is not expected to pass through path (II). Therefore, the formation of the product [structure (g)] through path (II), (III), and (IV) will be retarded by 39.801, 65.436, and 65.436 kcal/mol respectively relative to path (I). In path (I), the product is formed with minimum energy relative to the other paths. Hence, path (I) is preferred for the product release stage for the oxidative hydroxylation of quinoline in the reductive half-reaction active site of Qor.
In summary, the results obtained from energy, charges, bond length, and percentage contribution of the chemical fragments to the HOMOs, and molecular orbital analysis supported concerted reaction mechanism for the oxidation of quinoline to 1-hydro-2-oxoquinoline on the in the reductive half-reaction active site of Qor.
Density functional theory methods of electronic structures calculation was used for the study. Based on the data obtained from Mulliken atomic charge profile, % contribution on HOMO, and total energies, it is theoretically probed that C2 is the interaction site of quinoline.
The SMo–HRH bond distance for the model transition state structures of quinoline is found to be 1.960Å. The transition state structure was confirmed with one imaginary negative frequency of −104.5. The transition state total energy of quinoline is found to be −1.2365899E+06 kcal/mol.
The increase and the decrease in the partial positive charges on Mo and C2 of quinoline shows that molybdenum is reduced from Mo(+VI) to Mo(+IV) in the course of the reaction due to the transfer of electrons from C2 of quinoline to the molybdenum center. Likewise, the partial negative charge on Oeq is decreased due to the withdrawal of bonding electrons (Oeq–CRH) away from it. On the other hand, the electropositivity of the substrate hydrogen (HRH) is decreased due to the accumulation of negatively charged particles on it. The apical oxo plays a “spectator” role in the reaction as it shows insignificant charge variations. Moreover, the equatorial oxygen is a better nucleophile relative to the apical oxo since the accumulation of partial negative charge on the equatorial oxygen is higher than the apical oxo.
The increase and the decrease in the bond lengths predicted the cleavage of Mo–Oeq and CRH-HRH and formation of CRH–Oeq and SMo–HRH bonds at the transition state, respectively. The increase in metallic character of molybdenum revealed that electrons are transferred from C2 of quinoline to Mo center and hence the reduction of Mo(+VI) to Mo(+IV) during the oxidative hydroxylation of quinoline by Qor.
From the Mulliken atomic charge changes, it is reasonably predicted that the equatorial oxygen is a better nucleophile in the oxidative hydroxylation of quinoline. The decrease and the increase in the partially negatively charged particles on Mo and C2, respectively assured the transfer of electrons from C2 of quinoline to the Mo center. The accumulation of partial negative charges on the hydrogen atom at the product bound relative to the substrate bound, possibly predicted that hydrogen is transferred in the form of hydride (H + 2e−) from C2 to SMo. Eventually, it is reasonably concluded that the oxidative hydroxylation of quinoline in the reductive half-reaction active site of Qor are concerted.
EAB carried out all the computional calculations, analyzed and interpreted the data. The author prepared, read and approved the final manuscript.
The author declares that he has no competing interests.
- Bonin I, Martins BM, Purvanov V, Fetzner S, Huber R, Dobbek H (2004) Active site geometry and substrate recognition of the molybdenum hydroxylase quinoline 2-oxidoreductase. Structure (London) 12:1425–1435View ArticleGoogle Scholar
- Schach S, Tshisuaka B, Fetzner S, Lingens F (1995) Quinoline 2-oxidoreductase and 2-oxo-1, 2-dihydroquinoline 5-6, dioxygenase from Comamonas testosterone 63: the first two enzymes in quinoline and 3-methylquinoline degradation. Eur J Biochem 232:536–544View ArticleGoogle Scholar
- Graves PR, Kwiek JJ, Fadden P, Ray R, Hardeman H, Coley AM, Foley M, Haystead TAJ (2002) Discovery of novel targets of quinoline drugs in the human purine binding proteome. Am Soc Pharmacol Exp Therapuetics Mol Pharmacol 62:1364–1372Google Scholar
- Jianlong W, Liping H, Hanchang S, Yi Q (2001) Biodegradation of quinoline by gel immobilized Burkholderia sp. Chemosphere 44:1041–1046View ArticleGoogle Scholar
- Schwarz G (2005) Molybdenum cofactor biosynthesis and deficiency. Cell Mol Life Sci 62:2792–2810View ArticleGoogle Scholar
- Stephan I, Tshisuaka B, Fetzner F, Ltngbns F (1996) Quinaldine 4-oxidase from Arthrobactor sp. Ru41a, a versatile prokaryotic molybdenum-containing hydroxylase active towards N-containing heterocyclic compounds and aromatic aldehydes. Eur J Biochem 236:155–162View ArticleGoogle Scholar
- Deeken UF, Goldenstedt B, Janßen RG, Kapp R, Huttermann J, Fetzner S (2003) Functional expression of quinoline 2-oxidoreductase genes (qorMLS) in Pseudomonas putida KT2440 pUF1 and in P. putida 86-1 Dqor pUF1 and analysis of the Qor proteins. Eur J Biochem 270:1567–1577View ArticleGoogle Scholar
- Hille R (1996) The mononuclear molybdenum enzymes. Chem Rev 96:2757–2816View ArticleGoogle Scholar
- Okamoto K, Matsumoto K, Hille R, Eger BT, Pai EF, Nishino T (2004) The crystal structure of Xanthine oxidoreductase during catalysis: implications for reaction mechanism and enzyme inhibition. Proc Nat Acad Sci. 101:7931–7936View ArticleGoogle Scholar
- Hille R (2006) Structure and function of Xanthine oxidoreductase. Eur J Inorg Chem 10:1913–1926View ArticleGoogle Scholar
- Schwarz G, Mendel RR, Ribbe MW (2009) Molybdenum cofactors, enzymes and pathways. Nature 460:839–847View ArticleGoogle Scholar
- Luisa CM, Eleonora E, Soledad GE, Barbara H, Alejandro TE (2011) The reaction electronic flux in chemical reactions. Sci China Chem 54(12):1982–1988View ArticleGoogle Scholar
- People JA et al (2004) Gaussian 03, revision C02 Gaussian, Inc. CT, WallingfordGoogle Scholar
- Becke AD (1993) Density-functional thermochemistry. III. The role of exact exchange. J Chem Phys. 98:5648–5652View ArticleGoogle Scholar
- Hay PJ, Wadt WR (1985) Ab initio effective core potentials for molecular calculations. Potentials for the transition metal atoms Sc to Hg. J Chem Phys. 82:270–283View ArticleGoogle Scholar
- Contreras JG, Gerli LA (2008) Conformational preference in 4, 6-dimethyl-1, 3-thioxane. J Chil Chem Sci. 53:1400–1402Google Scholar
- Gorelsky SI (2009). AOMix: Program for molecular orbital analysis; University of Ottawa http://www.sg-chem.net/
- Gorelsky SI, Lever ABP (2001) Electronic structure and spectra of ruthenium diimine complexes by density functional theory and INDO/S. Comparison of the two methods. J Organomet Chem 635:187–196View ArticleGoogle Scholar
- Zhang XH, Wu YD (2005) A theoretical study on the mechanism of the reductive half-reaction of Xanthine oxiddase. Inorg Chem 44(5):1466–1471View ArticleGoogle Scholar
- Amano T, Ochi N, Sato H, Sakaki S (2007) Oxidation reaction by Xanthine oxidase. Theoretical study of reaction mechanism. J Am Chem Soc 129:8131–8138View ArticleGoogle Scholar