German Cavelier

Role of Fluctuations in Quinone Reductase Hydride Transfer: a Combined Quantum Mechanics and Molecular Dynamics Study By German Cavelier* † and L. Mario Amzel * ¶ * Department of Biophysics and Biophysical Chemistry Johns Hopkins University School of Medicine, Baltimore, MD 21205 † Present address: Office of Interdisciplinary Research and Scientific Technology, Division of Neuroscience and Basic Behavioral Science, National Institute of Mental Health, National Institute of Health, Bethesda, MD 20892 ¶ Corresponding Author. E-mail:- Abstract: Quinone Reductase is a cytosolic FAD-containing enzyme that carries out the obligatory twoelectron reduction of quinones to hydroquinones. The first step in the mechanism consists of the reduction of the FAD by NAD(P)H via direct a hydride transfer. Combined QM/MM calculations show that the protein accelerates this step by a combination of effects that include charge stabilization and distortion. The calculations also show that dynamic effects play an important role in QR catalysis: the distance between the donor and the acceptor atoms of the hydride transfer, which is too long for transfer in the static structure, becomes shorter than 3 Å 25% of the time due to motions of the protein and the cofactors. Keywords: Density Functional, Enzymes, Flavoproteins, DT-diaphorase, isoalloxazine, Quantum Chemistry, Protein fluctuations, Molecular Dynamics INTRODUCTION Quinone reductase (QR1; NQO1), a cytosolic phase 2 detoxification flavoenzyme, catalyzes the obligatory two-electron reduction of quinones to hydroquinones using either NADH or NADPH as the electron donor. The catalytic cycle of QR1 involves reduction of bound FAD by NAD(P)H via a hydride transfer from C4 of the nicotinamide to N5 of the flavin. In a previous paper1 the energetics of charge stabilization by QR1 following this hydride transfer was estimated using an ab initio DFT calculation. However, other important aspects of the mechanism of H- transfer remain poorly understood. For example, in the x-ray structure of the complex of NADP+ with oxidized QR1, the distance between C4 of the nicotinamide and N5 of the flavin (4.2 Å), is probably too long for aLINE direct(BELOW) hydride TO transfer. Since theON hydride transfer is fast (i.e. isPAPER not rate CREDIT BE INSERTED THE FIRST PAGE OF EACH CP851, From Physics to Biology; BIFI 2006 II International Congress, edited by J. Clemente-Gallardo, Y. Moreno, J. F. Sáenz Lorenzo, and A. Velázquez-Campoy © 2006 American Institute of Physics-/06/$23.00 1 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp limiting in QR1) questions remain about what effects contribute to the acceleration of this step. It is becoming increasingly clear that enzymes are not rigid “lock and key” devices: molecular motions may make critical contributions to rate acceleration in enzymatic catalysis2. Enzymes can bring their substrates closer to the transition state geometry not only by static stabilization, but also by means of conformational fluctuations, dynamic preorganization, and other dynamical effects. Here we analyze the energetics of these effects in the hydride transfer step of QR1 by ab initio quantum mechanical calculations combined with molecular mechanics/dynamics (QM/MM). METHODS Computer Programs. Cofactors within the context of the protein were modeled using the programs QUANTA and UNICHEM on SGI workstations. The initial optimizations and transition state search were carried out with the MNDO 94 program (AM1 Hamiltonian), accessed through UNICHEM, and run on an SV1 CRAY at the Advanced Biomedical Computing Center, National Cancer Institute, Frederick, MD. Ab initio calculations were done as single point energy calculations at the B3LYP/6-31G(d) level of theory, as implemented in GAUSSIAN3 98 on the same SV1 CRAY supercomputer. Visualizations were done with UNICHEM, QUANTA, MATLAB, and EXCEL. Molecular Dynamics calculations were performed with CHARMM on a Power Challenger SGI. Residues included in the quantum mechanical calculations. In addition to the cofactors, the following residues, determined to be important in the catalytic mechanism of QR1, were included in the QM calculations: Trp 105, Phe 106, Gly 149, Gly 150, Tyr 155, and His 161 4-7. All residues and residue pairs were terminated at their carboxy termini with methyl groups. Identification of the transition state. The system considered here comprises approximately 200 atoms from FAD, NAD, and the interacting amino acid residues. It is not practical to perform a saddle point optimization for a system of this size8-19. Furthermore, as the reaction coordinate comprises movements of cofactors and of residues around the active site of the enzyme, finding the transition state requires optimization combined with molecular dynamics of the whole protein, i.e., combined QM/MM methodology20-33. To overcome these computational difficulties, we carried out our calculations in two stages: (1) identification of an approximate transition state, and (2) optimization of the system by successive approximations. Starting with the optimized structure a series of grid calculations was made in points around the straight line from C4N to the transition state, and around the straight line from the transition state to N5F. Optimization at each grid point was performed allowing the cofactors to relax, with the exception of their links to the sugar-adenine portion, which were kept fixed at the x-ray coordinates. (This restriction mimics the protein environment, where the cofactors are anchored firmly in place.) These calculations were carried out using the MNDO 94 program in UNICHEM, with the AM1 Hamiltonian. 2 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp Relaxation of the active site plus the whole protein. To take into account the effects of relaxation of the protein, the cofactors and the substrate in the protein environment, quantum mechanical calculations of the active site were combined with molecular dynamics calculations of the complete system (protein plus cofactors) using procedures implemented in CHARMM20,25. For these combined calculations the results from the quantum chemical DFT B3LYP/6-31G(d) calculations (geometry and Mulliken charges of the cofactors and of residues around the active site of the enzyme) were input as part of the CHARMM parameter and topology files 34. The structures including these modifications were minimized with CHARMM and used as the initial models in whole enzyme molecular dynamics calculations of the different species involved in the catalytic mechanism: QR1 with reduced N-methyl nicotinamide and oxidized FAD, QR1 with the transition state, QR1 with oxidized N-methyl nicotinamide and reduced FADH, and the final structure after completion of the charge relay. Molecular dynamics calculations were run starting with the minimized structures, using an integration step of 1 fs. The molecule was heated from 0 to 300 K, increasing the temperature by 10 K every 100 steps (total 3,100 steps; 3.1 ps), equilibrated for 1,000 steps (1 ps), and run for 30 or more picoseconds. The time series generated by the dynamics simulation was analyzed for the following critical atom-to-atom distances: (a) the distance between C4N (C4 of nicotinamide) and N5F (N5 of flavin), i.e., the hydride transfer path; (b) the distance between FO2 (O2 of flavin) and the O of Tyr-155, important for the first step of the charge relay performed by the enzyme; and (c) the distance between the Nε of His-161 and the hydroxyl O in Tyr155, involved in the last step of the charge relay mechanism. RESULTS and DISCUSSION Transition state identification and relaxation of the cofactors. In the initial search, only movements of the hydride were allowed, while keeping all other atoms fixed. These calculations included cofactors and selected amino acid residues important in the QR1 mechanism. The sensitivity of the transition state to movement of the cofactors was analyzed by permitting relaxation of the coordinates of the carboxyamide portion of the nicotinamide. It was found that once the hydride leaves the nicotinamide, the companion hydrogen on C4N begins to enter the plane of the nicotinamide ring and collides with one of the hydrogens of the carboxyamide. The result is that in the transition state the carboxyamide plane is rotated with respect to the plane of the ring such that the NH2 moves towards the flavin. This conformation of the carboxyamide in the transition state was confirmed by performing optimizations with different dihedral angle values for the rotation of the carboxyamide plane, moving the NH2 both towards and away from the flavin. Calculations including the charge relay mechanism. The ensuing step in the proposed mechanism of QR1 involves a charge relay. The proposed charge relay consists of the migration of a proton from Tyr-155 to O2F of the flavin, followed by a second proton 3 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp transfer from His-161 to Tyr-155. It is not known whether the proton transfers take place as part of the formation of the transition state or after the hydride transfer has already occurred. To test both possibilities the energy of the structures with the transition state and either one or both protons transferred was calculated as described in Methods. Results of these calculations show that transference of one or both protons does not stabilize the transition state suggesting that the transition state is formed before either of the two protons is transferred. Relaxation of the intermediate and final structures. To analyze the effect of cofactor relaxation the energies were recalculated allowing the cofactors to relax by optimizing their geometries with the semi-empirical MNDO 94 program in UNICHEM, with the AM1 Hamiltonian. The sugar and adenine portions of both cofactors and of the side chains included in the QM calculations were kept fixed at their x-ray positions. Although grid calculations performed allowing only movement of the hydride show a clear saddle point (transition state), the potential energy surface is asymmetrical, indicating the development of unbalanced forces upon movement of the hydride (Fig. 1). FIGURE 1. Energy surface obtained with single point energy calculations. The points of the 3D grid are along and around the lines joining the position of the hydride at the transition state with its positions when bound to the donor (C4 of nicotinamide) and the acceptor (N5 of the flavin). 4 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp FIGURE 2. Energy surface obtained allowing relaxation of the cofactors. The points are defined as those in Fig. 1. Moreover, the estimated activation energy is 80 kcal/mol, about one order of magnitude too large to account for the experimental rate of reduction of the enzyme. To reduce this strain, an optimization was performed at each grid point, allowing the cofactors to relax but fixing their sugar-adenine portions, considered to be anchored in place by the protein. The resulting energy surface (Fig. 2) clearly shows that the unbalanced forces are a consequence of restricting the movement of the cofactors: once they are allowed to relax, the resulting energy surface is symmetrical.35-38 An additional optimization allowing the active site side chains to move without constraints, lowered the transition state energy by about 30 kcal/mol. However, this approximation is not realistic because the side chains move to positions where the rest of the protein might not permit them to move. Allowing side chain movement by combined QM/MM methodology was then used to include the relaxation by molecular mechanics of the whole protein around the quantum mechanically-treated active site. Enzymatic rate enhancement by relaxation of the whole protein. For these calculations the reference state was the enzyme before the reaction, with oxidized FAD and the reduced NADH lying above the FAD. Exploratory molecular dynamics runs showed that after heating to 300 K (3.1 ps), early equilibration (1.0 ps) and molecular dynamics (70.0 5 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp ps), the difference in potential energy between the initial and the transition states stabilizes once the simulation was run for at least 65 picoseconds (Fig. 3). FIGURE 3. Energies before and during the transition state (TS). (a) Total energy. (b) Potential energy. (c)Potential energy difference between the initial sate and the transition state. The steady state value of this difference in potential energy indicates that relaxation of the protein during the reaction lowers the transition state barrier by an average of 70 kcal/mol. Since the activation energy estimated without protein relaxation is ~80 kcal/mol, allowing for relaxation of the protein lowers the activation energy to ~10 kcal/mol (Fig. 4), a range compatible with the experimental data. 6 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp FIGURE 4. Effect of the relaxation of the protein on the energy barrier of the reaction. The energies are shown using the energy before the reaction as the reference state (zero value). The reduction in the activation energy (~70 Kcal/mol) is the result of allowing the protein to relax by MD simulations. It must be noted that the molecular dynamics energies fluctuate, and thus the actual lowering of the barrier at any given instant may be more or less than the average of 70 kcal/mol. This effect is similar to the rate-promoting vibrations discussed in the analysis of the reaction of horse liver alcohol dehydrogenase39, and to the network of coupled motions40 and the preorganization and protein dynamics effect2 found in hydride transfer in dihydrofolate reductase. It also similar to the contribution to catalysis of conformational fluctuations found in a study of in HIV-1 protease32, and by acyl carrier protein reductase from Mycobacterium tuberculosis33. Molecular dynamics studies allowing relaxation of the whole protein highlight the contribution of protein flexibility to enzymatic catalysis2,39,41. For example, in some cases, the crystallographic distance between two atoms proposed to interact during catalysis is too large40,41, but this difficulty may be artifactual: protein motions may bring the two atoms closer together a significant fraction of the time41, and allow the reaction to occur. The procedure used in this study allows exploration of this type of dynamic contribution of the protein to the enzymatic mechanism of QR. The dynamic relaxation that lowers the energy of the transition state also lowers the mean atom-to-atom distances between reacting groups, favoring the overall reaction. The distance between C4N and N5F (the hydride transfer path) before the reaction occurs (red curve in Fig. 5a) is approximately 3.5 Å, goes up to more than 4.0 Å, and then returns to around 3.5 Å. However, at the transition state (blue curve in Fig. 5a) the distance between C4N and N5F becomes shorter. It is less than 3.1 Å 40% of the time of the simulation, and shorter than 2.9 Å for 15% of the time (Fig. 6a). This will no doubt facilitate the hydride transfer, because the electronic charge seems to only partially migrate at the transition state. At the transition state the hydride only has a negative charge of -0.35 rather than the expected 1.00, with the rest of the electronic charge still with the NAD. The close proximity of C4N and N5F will facilitate migration of the charge, probably through a HOMO (High Occupied Molecular Orbital) shared between NAD, the hydride and FAD. 7 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp FIGURE 5. Distances between critical atoms of the cofactors and the protein side-chains during the MD simulations. (a) Distance between nicotinamide C4 and flavin N5. (b) Distance between His 161 Nε and Tyr 155 OH. (c) Distance between Tyr 155 OH and flavin O2. Distances in diagrams (b) and (c) correspond to the proton donors and acceptors of the charge relay. Curves corresponding to the simulation before the reaction are in red, and those in the transition state are in blue. The short distance between C4N and N5F may even allow tunneling2,36-38. The distance between FO2 and the O in Tyr-155 before the reaction (red curve, Fig. 5c) begins at ~3.5 8 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp Å and increases to ~3.8 Å. At the transition state (blue curve, Fig. 5c) it begins at ~3.8 Å and eventually reaches close to hydrogen-bonding distance (~3.5 Å). This close apposition will facilitate the proton transfer from Tyr-155 to FO2, and thus facilitate the charge relay. FIGURE 6. Percentage of the time spent closer than the indicated distance by critical atoms of the cofactors and the protein side-chains during the MD simulations. (a) Distance between nicotinamide C4 and flavin N5. (b) Distance between His 161 Nε and Tyr 155 OH. (c) Distance between Tyr 155 OH and flavin O2. Diagrams (a) and (b) correspond to the distances between the proton donors and acceptors of the charge relay. Curves corresponding to the simulation before the reaction are in red, and those in the transition state are in black. A similar relaxation was observed in calculations performed for a similar enzymatic reaction, the deacylation in Class A Beta-Lactamases41. When the structure was allowed to relax, calculations showed distance shortening and energy barrier lowering for 9 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp different steps of the proposed reaction mechanism that should greatly facilitate the reaction. In particular, the crystallographic coordinates show a large distance between the catalytic glutamate and serine residues, excluding direct proton transfer. However, upon relaxation by QM/MM, the proposed mechanism becomes energetically feasible and the distances for proton transfer are within reasonable limits. Before the reaction, the distance between the Nε of His-161 and the O of Tyr-155 (red curve in Fig. 5b) is around 4.0 Å, and increases steadily to close to 5.0 Å. However, at the transition state (blue curve in Fig. 5b), the simulation shows that these two atoms come within hydrogen bonding distance, thus facilitating the second step in the charge relay. The set of motions found here, which results in distance shortening and energy barrier lowering in quinone reductase, is similar to the network of coupled motions described in the hydride transfer reaction of dihydrofolate reductase40. Taken together all these results show that relaxation of the entire protein during the catalytic reaction can bring interacting atoms closer together and significantly lower the transition state energy, greatly facilitating the enzymatic reaction, both in energetic and in dynamic terms40,42-44. Extensive non-statistical dynamics have been shown to be important in gas-phase SN2 reactions45. The molecular dynamics simulations performed here with QM-derived atomic charges extend this type of analysis to enzyme systems and provide non-statistical details that are not accessible from studies based on (static) potential energy surfaces analyzed with transition state theory. In particular, motions of protein atoms that participate in the reaction mechanism provide an important contribution to catalysis because they result in transient close proximity of reacting groups. An experimental evaluation of the contribution of enzyme dynamics to catalysis has been carried out for conformational fluctuations of dihydrofolate reductase in the micro- to mili-second range using nuclear magnetic resonance relaxation methods46. It was found that the time scales of protein dynamics coincide with those of substrate turnover, but the details of the motions during catalysis cannot be observed by this method46. Molecular dynamics methods such as those used here, based on quantum chemical and molecular dynamics calculation before and after the reaction, provide a simulation tool to study in atomic detail the contribution of atomic movements to catalysis40. SUMMARY AND CONCLUSIONS Contributions to catalysis. The factors that increase the enzymatic reaction rate and result in rate-enhancement by QR1 can be dissected by referring to general mechanisms known to be important for rate enhancement in enzyme catalysis: Approximation, Charge Relay, Electrostatic Catalysis, and Strain or Distortion47,48. This study shows that in QR these contributions are amplified by dynamic effects, giving a high overall rate enhancement41,49. Approximation (effective concentration) plays an important role in catalysis by QR. The enzyme provides an optimal initial alignment of the substrate and flavin cofactor, such 10 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp that the hydride needs to cross a distance of only 2.5 Å (dC4-N5=3.5 Å). Dynamic effects (Fig. 5a, blue trajectory) reduce this distance by an average to 0.5 Å. (Fig. 6a shows that the NC4-FN5 distance is below 3 Å approximately 25% of the time.) The distance between the Nε of His-161 and the OH oxygen of Tyr-155 is also diminished dynamically (Fig. 5b, blue trajectory; and Fig. 6b) by 0.3-0.8 Å. In the transition state this distance is shorter than 3.5 Å 25% of the time. A similar shortening is observed for the distance between the oxygen in Tyr-155 and the FO2 in the isoalloxazine ring (Fig. 5c, blue trajectory; and Fig. 6c). With atoms at these short distances a significant fraction of the time, the transfer of the hydride and the proton should be highly facilitated. If the rate enhancement by dynamic proximity is greater than the inverse of the fraction of the time spent at that distance, the net effect will be an increase of the rate. It is obvious that Electrostatic Catalysis plays an important role in rate enhancement by QR. That this contribution is enhanced by dynamic effects can be inferred from Fig. 3. Dynamic electrostatic rearrangements in and around the active site upon going from the initial state to the transition state significantly lower the energy of the transition state. A final effect important in rate enhancement by QR is Strain or Distortion. The optimizations and energy analysis indicate that the protein holds the reduced flavin in its strained planar conformation, destabilizing the ground state of the back hydride transfer reaction (from the isoalloxazine ring of the flavin to the quinone/substrate) by 2 to 4 kcal/mol. This makes the flavin a better reductant for the wide variety of quinones that it is known to reduce. The present study also shows that, as part of these contributions, dynamic effects play an important role in catalysis by QR. During the hydride transfer from NADH to the FAD the hydride donor (C4 of nicotinamide) and acceptor (N5 of FAD) spend part of their time at distances that favor the formation of the transition state of the transfer. Hydride transfer takes place when, under normal thermal motion, the protein adopts one of the conformations that favors the transfer. Since these conformations are present a significant fraction of the time, transfer when the protein adopts them must represent one of the important paths in the mechanism of the reaction. Using the procedures described here it will be straightforward to make predictions about relative rate enhancements for different quinone pro-drug substrates in the design of cancer chemotherapeutic drugs 4-6. For example, calculations may be used to redesign drugs known to be activated by quinone reductase that are reduced more effectively by the rat enzyme than by human QR50. ACKNOWLEDGMENTS Supported by Grant GM 51362 of the National Institute of General Medical Sciences. We thank the Supercomputing Center of the National Cancer Institute, Frederick, Md. 11 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Cavelier G, Amzel LM. Mechanism of NAD(P)H : quinone reductase: Ab initio studies of reduced flavin. Proteins-Structure Function and Genetics 2001;43(4):420-432. Rajagopalan PT, Benkovic SJ. Preorganization and protein dynamics in enzyme catalysis. Chem Rec 2002;2(1):24-36. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Zakrzewski VG, Montgomery JA, Stratmann RE, Burant JC, Dapprich S, Millam JM, Daniels AD, Kudin KN, Strain MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R, Mennucci B, Pomelli C, Adamo C, Clifford S, Ochterski J, Petersson GA, Ayala PY, Cui Q, Morokuma K, D. K. Malick, Rabuck AD, Raghavachari K, Foresman JB, Cioslowski J, J. V. Ortiz, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I, R. Gomperts, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A, Gonzalez C, Challacombe M, Gill PMW, Johnson BG, Chen W, Wong MW, Andres JL, Head-Gordon M, an ESR, Pople JA. Gaussian 98. Pittsburgh PA: Gaussian, Inc.;- p. p. Amzel LM. Structure-based drug design. Current Opinion in Biotechnology 1998;9(4):366-369. Faig M, Bianchet MA, Talalay P, Chen S, Winski S, Ross D, Amzel LM. Structures of recombinant human and mouse NAD(P)H : quinone oxidoreductases: Species comparison and structural changes with substrate binding and release. Proceedings of the National Academy of Sciences of the United States of America 2000;97(7):-. Faig M, Bianchet MA, Winski S, Hargreaves R, Moody CJ, Hudnott AR, Ross D, Amzel LM. Structure-based development of anticancer drugs: Complexes of NAD(P)H : quinone oxidoreductase 1 with chemotherapeutic quinones. Structure 2001;9(8):659-667. Li R, Bianchet MA, Talalay P, Amzel LM. The three-dimensional structure of NAD(P)H:quinone reductase, a flavoprotein involved in cancer chemoprotection and chemotherapy: Mechanism of the two-electron reduction. Proc Natl Acad Sci USA 1995;92:-. Abashkin Y, Russo N, Toscano M. Transition-State Localization By a DensityFunctional Method - Applications to Isomerization and Symmetry-Forbidden Reactions. Theoretica Chimica Acta 1995;91(3-4):179-186. Alhambra C, Corchado JC, Sanchez ML, Gao JL, Truhlar DG. Quantum dynamics of hydride transfer in enzyme catalysis. J Am Chem Soc 2000;122(34):-. Andres J, Moliner V, Safont VS, Domingo LR, Picher MT. On Transition Structures for Hydride Transfer Step in Enzyme Catalysis. A Comparative Study on Models of Glutathione Reductase Derived from Semiempirical, HF, and DFT Methods. J Org Chem 1996;61(22):-. Andres J, Moliner V, Safont VS, Aullo JM, Diaz W, Tapia O. Transition structures for hydride transfer reactions in vacuo and their role in enzyme catalysis. Theochem-Journal of Molecular Structure 1996;371:299-312. Hurley MM, HammesSchiffer S. Development of a potential surface for simulation of proton and hydride transfer reactions in solution: Application to 12 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. NADH hydride transfer. Journal of Physical Chemistry a 1997;101(21):-. Diaz W, Aullo JM, Paulino M, Tapia O. Transition structure and reactive complexes for hydride transfer in an isoalloxazine-nicotinamide complex. On the catalytic mechanism of glutathione reductase. An ab initio MO SCF study. Chemical Physics 1996;204(2-3):195-203. Ferenczy GG, Naray-Szabo G, Varnai P. Quantum mechanical study of the hydride shift step in the xylose isomerase catalytic reaction with the fragment self- consistent field method. International Journal of Quantum Chemistry 1999;75(3):215-222. Mestres J, Duran M, Bertran J. Characterization of the Transition State for the Hydride Transfer in a Model of the Flavoprotein Reductase Class of Enzymes. Bioorg Chem 1996;24:69-80. Nishimoto K, Higashimura K, Asada T. Ab initio molecular orbital study of the flavin-catalyzed dehydrogenation reaction of glycine - protein transport channel driving hydride-transfer mechanism. Theoretical Chemistry Accounts 1999;102(1-6):355-365. Park BK, Doh ST, Son GS, Kim JM, Lee GY. Mo Study of Hydride Transfer Between Nadh and Flavin Nucleotides. Bulletin of the Korean Chemical Society 1994;15(4):291-293. Webb SP, Agarwal PK, Hammes-Schiffer S. Combining electronic structure methods with the calculation of hydrogen vibrational wavefunctions: Application to hydride transfer in liver alcohol dehydrogenase. Journal of Physical Chemistry B 2000;104(37):-. Tapia O, Andrés J, Safont VS. Transition structures in vacuo and the theory of enzyme catalysis. Rubisco's catalytic mechanism: a paradigmatic case? Journal of Molecular Structure: THEOCHEM 1995;342(1-3):131-140. Brooks BR, Bruccoleri RE, Olafson BD, States DJ, Swaminathan S, Karplus M. Charmm - a Program For Macromolecular Energy, Minimization, and Dynamics Calculations. Journal of Computational Chemistry 1983;4(2):187-217. Cui Q, Karplus M. Molecular properties from combined QM/MM methods. I. Analytical second derivative and vibrational calculations. J Chem Phys 2000;112(3):-. Cummins PL, Gready JE. Molecular dynamics and free energy perturbation study of hydride-ion transfer step in dihydrofolate reductase using combined quantum and molecular mechanical model. Journal of Computational Chemistry 1998;19(8):977-988. Liu H, Müller-Plathe F, Gunsteren WFv. A Combined Quantum/Classical Molecular Dynamics Study of the Catalytic Mechanism of HIV Protease. Journal of Molecular Biology 1996;261(3):454-469. Sheppard DW, Burton NA, Hillier IH. Ab initio hybrid quantum mechanical/molecular mechanical studies of the mechanisms of the enzymes protein kinase and thymidine phosphorylase. Journal of Molecular StructureTheochem 2000;506:35-44. Cui Q, Karplus M. Triosephosphate isomerase: a theoretical comparison of alternative pathways. J Am Chem Soc 2001;123(10):-. 13 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. Ramos MJ, Melo A, Henriques ES, Gomes J, Reuter N, Maigret B, Floriano WB, Nascimento MAC. Modeling enzyme-inhibitor interactions in serine proteases. International Journal of Quantum Chemistry 1999;74(3):299-314. Varnai P, Warshel A. Computer simulation studies of the catalytic mechanism of human aldose reductase. J Am Chem Soc 2000;122(16):-. Bentzien J, Muller RP, Florian J, Warshel A. Hybrid ab initio quantum mechanics molecular mechanics calculations of free energy surfaces for enzymatic reactions: The nucleophilic attack in subtilisin. Journal of Physical Chemistry B 1998;102(12):-. Kong YS, Warshel A. Linear Free-Energy Relationships With QuantumMechanical Corrections - Classical and Quantum-Mechanical Rate Constants For Hydride Transfer Between Nad(+) Analogs in Solutions. J Am Chem Soc 1995;117(23):-. Yadav A, Jackson RM, Holbrook JJ, Warshel A. Role of Solvent Reorganization Energies in the Catalytic Activity of Enzymes. J Am Chem Soc 1991;113(13):-. Cavalli A, Carloni P. Enzymatic GTP hydrolysis: insights from an ab initio molecular dynamics study. J Am Chem Soc 2002;124(14):-. Piana S, Carloni P, Parrinello M. Role of conformational fluctuations in the enzymatic reaction of HIV-1 protease. J Mol Biol 2002;319(2):567-583. Pantano S, Alber F, Lamba D, Carloni P. NADH interactions with WT- and S94A-acyl carrier protein reductase from Mycobacterium tuberculosis: an ab initio study. Proteins 2002;47(1):62-68. Pavelites JJ, Gao JL, Bash PA, Mackerell AD. A molecular mechanics force field for NAD(+), NADH, and the pyrophosphate groups of nucleotides. Journal of Computational Chemistry 1997;18(2):221-239. Foresman JB, Frisch A. Exploring Chemistry with Electronic Structure Methods. Pittsburgh, PA: Gaussian, Inc.;- p. p. Ringe D, Petsko GA. Tunnel vision. Nature 1999;399(3 June 1999):417-418. Kohen A, Cannio R, Bartolucci S, Klinman JP. Enzyme dynamics and hydrogen tunneling in a thermophilic alcohol dehydrogenase. Nature 1999;399(3 June 1999):496-499. Kohen A, Klinman JP. Enzyme Catalysis: Beyond Classical Paradigms. Accounts Chem Res 1998;31:397-404. Caratzoulas S, Mincer JS, Schwartz SD. Identification of a protein-promoting vibration in the reaction catalyzed by horse liver alcohol dehydrogenase. J Am Chem Soc 2002;124(13):-. Agarwal PK, Billeter SR, Rajagopalan PT, Benkovic SJ, Hammes-Schiffer S. Network of coupled promoting motions in enzyme catalysis. Proc Natl Acad Sci U S A 2002;99(5):-. Castillo R, Silla E, Tunon I. Role of protein flexibility in enzymatic catalysis: quantum mechanical- molecular mechanical study of the deacylation reaction in class A beta- lactamases. J Am Chem Soc 2002;124(8):-. Bahar I, Erman B, Jernigan RL, Atilgan AR, Covell DG. Collective motions in HIV-1 reverse transcriptase: examination of flexibility and enzyme function. J Mol Biol 1999;285(3):-. 14 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp 43. 44. 45. 46. 47. 48. 49. 50. Keskin O, Jernigan RL, Bahar I. Proteins with similar architecture exhibit similar large-scale dynamic behavior. Biophys J 2000;78(4):-. Mesecar AD, Stoddard BL, Koshland DE, Jr. Orbital steering in the catalytic power of enzymes: small structural changes with large catalytic consequences. Science 1997;-):202-206. Sun L, Song K, Hase WL. A SN2 reaction that avoids its deep potential energy minimum. Science 2002;-):875-878. Eisenmesser EZ, Bosco DA, Akke M, Kern D. Enzyme dynamics during catalysis. Science 2002;-):-. Garcia-Viloca M, Gao J, Karplus M, Truhlar DG. How enzymes work: analysis by modern rate theory and computer simulations. Science 2004;-):186195. Strajbl M, Shurki A, Kato M, Warshel A. Apparent NAC effect in chorismate mutase reflects electrostatic transition state stabilization. J Am Chem Soc 2003;125(34):-. Sulpizi M, Schelling P, Folkers G, Carloni P, Scapozza L. The rational of catalytic activity of herpes simplex virus thymidine kinase. a combined biochemical and quantum chemical study. J Biol Chem 2001;276(24):-. Chen S, Knox R, Wu K, Deng PS, Zhou D, Bianchet MA, Amzel LM. Molecular basis of the catalytic differences among DT-diaphorase of human, rat, and mouse. J Biol Chem 1997;272(3):-. 15 Downloaded 09 Feb 2007 to-. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp