Empirical Valence Bond Simulations of the Hydride-Transfer Step in

Oct 13, 2016 - Department of Physical and Organic Chemistry, Jožef Stefan Institute ... Department of Cell and Molecular Biology, Uppsala Biomedical ...
0 downloads 0 Views 486KB Size
Subscriber access provided by Northern Illinois University

Article

Empirical Valence Bond Simulations of the Hydride Transfer Step in the Monoamine Oxidase A Catalyzed Metabolism of Noradrenaline Matic Poberznik, Miha Purg, Matej Repi#, Janez Mavri, and Robert Vianello J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.6b09011 • Publication Date (Web): 13 Oct 2016 Downloaded from http://pubs.acs.org on October 16, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Empirical Valence Bond Simulations of the Hydride Transfer Step in the Monoamine Oxidase A Catalyzed Metabolism of Noradrenaline Matic Poberžnik,1,† Miha Purg,2,† Matej Repič,3 Janez Mavri,3,* and Robert Vianello4,*

1

Department of Physical and Organic Chemistry, Jožef Stefan Institute, Jamova cesta 39, SI–1000 Ljubljana, Slovenia.

2

Department of Cell and Molecular Biology, Uppsala Biomedical Centre, Husargatan 3, S–75124 Uppsala, Sweden.

3

Laboratory for Biocomputing and Bioinformatics, National Institute of Chemistry, Hajdrihova ulica 19, SI–1000 Ljubljana, Slovenia. Phone: +386 1 4760309. E-mail: [email protected] 4

Computational Organic Chemistry and Biochemistry Group, Ruđer Bošković Institute, Bijenička cesta 54, HR–10000 Zagreb, Croatia. Phone: +385 1 4561117. E-mail: [email protected]



These authors contributed equally to this work

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

ABSTRACT: Monoamine oxidases (MAOs) A and B are flavoenzymes responsible for the metabolism of biogenic amines such as dopamine, serotonin and noradrenaline, which is why they have been extensively implicated in the etiology and course of various neurodegenerative disorders, and, accordingly, used as primary pharmacological targets to treat these debilitating cognitive diseases. The precise chemical mechanism through which MAOs regulate the amine concentration, which is vital for the development of novel inhibitors, is still not unambiguously determined in the literature. In this work, we present atomistic empirical valence bond simulations of the rate-limiting step of the MAO A catalyzed noradrenaline (norepinephrine) degradation involving the hydride transfer from the substrate α–methylene group to the flavin moiety of the FAD prosthetic group, employing the full dimensionality and thermal fluctuations of the hydrated enzyme with extensive configurational sampling. We show that MAO A lowers the free-energy of activation by 14.3 kcal mol–1 relative to the same reaction in aqueous solution, while the calculated activation free energy of ΔG‡ = 20.3 ± 1.6 kcal mol–1 is found in reasonable agreement with the correlated experimental value of 16.5 kcal mol–1. The results presented here offer a strong support that both MAO A and MAO B isoforms function by the same hydride transfer mechanism. We also considered few point mutations of the "aromatic cage" tyrosine residue (Tyr444Phe, Tyr444Leu, Tyr444Trp, Tyr444His, Tyr444His and Tyr444Glu), and the calculated changes in reaction barriers are in agreement with experiments, thus providing further support to the proposed mechanism.

2 ACS Paragon Plus Environment

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

INTRODUCTION Monoamine oxidases (MAOs) are flavoenzymes that metabolize biogenic and dietary amines including the monoamine neurotransmitters noradrenaline, dopamine, serotonin and some histamine metabolites in various parts of the brain.1,2 MAOs convert amines to the corresponding imines that leave the enzyme active site and are non-enzymatically hydrolyzed to the corresponding aldehyde and ammonia.3 The enzyme is regenerated to its active form by molecular oxygen, producing hydrogen peroxide (H2O2) that is a precursor of the reactive oxygen species (ROS). This is why MAO catalytic activity is extensively implicated in neurodegenerative processes.4 Although H2O2 is usually a slow two-electron oxidizer and is rather stable, in the presence of a metal catalyst or heme, it can act as a very rapid and indiscriminate oxidant. The full oxidizing strength of H2O2 can be harnessed if it either reacts with reduced metal ions (most notably, Cu+ and Fe2+) or is single-electron reduced, which both give OH•, being one of the most potent oxidizing agents known to chemistry.4 OH• reacts at diffusion limited rates with almost everything found in the cell: as such, its toxicity is nonselective and its diffusion distance is, therefore, very short. Reactive radical moieties chemically decompose bilayer membranes leading to their leakage, being responsible for neuron dishomeostasis and death, giving rise to neurodegeneration. The principal role of MAO is the regulation of the neurotransmitter levels in the central and peripheral nervous systems, having a major impact on cardiac output, blood pressure, sleep, mood, cognition, and movement.5 MAO is found in two isoforms, MAO A and MAO B, which share around 70% sequence identity, yet they differ in tissue distribution and selectivity.6 Inhibitors that mainly act on MAO A are used in the treatment of depression due to their ability to raise serotonin and noradrenaline concentrations, while MAO B inhibitors decrease dopamine degradation and are used to treat Parkinson disease.7 Furthermore, MAO inhibition has a notable neuroprotective effect,

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

since, as mentioned, MAO catalyzed reactions yield neurotoxic products such as hydrogen peroxide and aldehydes.4,8,9 Despite this enormous functional significance, the precise molecular mechanism by which MAOs achieve their activity, which is vital for the development of novel and more effective inhibitors, is still not unambiguously determined.10–15 The availability of the high resolution crystal structures of MAO A16 and MAO B17 has opened the possibility to study the catalytic reaction by computational approaches.18 Recently, our group performed the first quantum mechanical (QM) study that convincingly demonstrated the prevailing feasibility of the two-step direct hydride transfer mechanism over alternative pathways for dopamine degradation using a QM-only cluster model of MAO B.14 This study was later extended to include the full enzyme through the EVB approach, which gave an activation free energy of 16.1 kcal mol– 1 15

, in excellent agreement with the experimentally determined value of 16.5 kcal mol–1.19 Our

mechanistic proposal is already gaining some affirmation in the literature,20–22 although, for example, Kästner and co-workers performed QM/MM study on human MAO B and suggested that there is a substantial electron transfer from the substrate to flavin before the reaction and attributed this in favor of the polar nucleophilic mechanism13,23 together with some earlier reports.24 However, we must emphasize that we did not observe any such electron transfer,14 and that Kästner and coworkers failed to obtain the substrate-flavin adduct, which was originally proposed to facilitate the proton transfer.25 Hydride transfer formally involves the transfer of two electrons and a proton. We are aware that both processes may not be fully concerted, which might depend on the nature of the substrate, its orientation and distance from the flavin co-factor moiety, as well as on the electrostatic potential produced by the protein environment. We are convinced that, what might appear as somewhat different reaction mechanisms, the one proposed by Kästner and co-workers and the direct hydride transfer mechanism, suggested by us, may be rationalized in this way.

4 ACS Paragon Plus Environment

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Noradrenaline is a central nervous system neurotransmitter and neuromodulator whose distribution and decomposition kinetics are of prime importance for understanding its physiological role and allowance for therapy. There are estimates that the noradrenaline system consists of just a few thousand neurons located primarily in the locus coeruleus, solitary nucleus and caudal ventrolateral nucleus parts of the brain. In spite of their small number, when activated noradrenaline neurons play major roles, mostly to mobilize both the brain and the body for various actions,26 including increasing arousal and alertness, promoting vigilance, enhancing the formation and retrieval of memory, and focusing attention. Noradrenaline is responsible for tonic and reflexive changes in the cardiovascular system and is used as an injectable drug for the treatment of critically low blood pressure. Few noradrenaline reuptake inhibitors are in clinical use for the treatment of depression, panic disorder, attention deficit hyperactivity disorder, narcolepsy and obesity.27 Noradrenergic input from the locus coeruleus controls memory processing, learning and behavior,28 while Heneka and co-workers demonstrated its complex role in the pathology of Alzheimer disease,29 which is all why noradrenaline levels have to be very carefully are precisely regulated. Tipton and co-workers studied the dynamics of noradrenaline decomposition by using various homogenized parts of human brain tissue at 37°C in oxygen saturated solutions,30 and reported the MAO A Vmax value of 561 ± 42 pmol mgprotein–1 min–1, which gives the activation free energy of 22.8 kcal mol–1 (see Supporting Information). Although no recombinant MAO was available at that time, the authors also measured the Vmax values for dopamine (680 ± 123), phenylethylamine (20 ± 8) and serotonin (228 ± 31), which relate to activation free energies of 22.7, 24.8, and 23.3 kcal mol–1. These seem highly overestimated in the context of very recent measurements by various groups,19,31–35 which all cluster somewhere between 16.0–17.5 kcal mol–1. Relating the results by Tipton and co-workers with the averages of the mentioned recent measurements for each substrate we obtained a linear correlation, from which we extrapolated that noradrenaline activation free energy should be around ΔG‡(exp) =

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

16.5 kcal mol–1 (for details see Supporting Information). The latter value will be employed here as the reference point for computational data. The purpose of the current article is to present the first computational molecular simulations on the catalytic activity of the MAO A isoform in order to evaluate its role in the metabolism of noradrenaline using a full enzyme model, and to obtain reaction free energies employing a QM/MM treatment of the enzyme that allows for the extensive configurational sampling. Specifically, the reactive subsystem was described with the empirical valence bond (EVB) method,36 allowing for a direct evaluation of the MAO A catalytic effect by comparing the activation free energy in the enzyme to the corresponding reference reactions in the gas phase or aqueous solution.37,38 It will also provide the first computational analysis of the effect of active-site point mutations on the kinetics of the hydride transfer step of amine decomposition in MAO A. The results presented here will serve as a basis to study the metabolism of other biogenic amines and provide further support in favor of the conclusion that both MAO isoforms operate by the same direct hydride transfer mechanism that we postulated for MAO B earlier.14,15 This is particularly important since some authors proposed that, despite their high similarity and practically identical active sites, both MAO A and B isoforms catalyze the same conversion of amines to imines through different mechanisms.12

COMPUTATIONAL METHODS Reference Reaction. For an accurate determination of the enzyme catalytic effect, the kinetic and thermodynamic parameters of an appropriate reference reaction are necessary. Usually the corresponding transformation in aqueous solution is considered, but the uncatalyzed degradation of

6 ACS Paragon Plus Environment

Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

noradrenaline is an extremely slow process with a net reaction ΔGr > 0, and, to the best of our knowledge, no experimentally measured parameters are available. Hence, we decided to model the reaction in the gas-phase with quantum mechanical methods using the Gaussian09 program package.39 Specifically, we used the DFT based M06–2X functional developed by Truhlar and coworkers40 together with the 6–311++G(2df,2pd) basis set to obtain total electronic energies on geometries optimized with a smaller 6–31+G(d,p) basis set, giving rise to the M06–2X/6– 311++G(2df,2pd)//M06–2X/6–31+G(d,p) model employed here. The stationary points were characterized with frequency calculations and the transition state was confirmed by following the steepest decent path on the potential energy surface along the intrinsic reaction coordinate. Thermal corrections from the frequency analysis were added to total electronic energies without the application of the scaling factors to give Gibbs free energies reported here. Although, as mentioned, the most appropriate reference reaction would be the reaction in aqueous solution, we modeled the reaction as a direct hydride transfer in the gas phase, because the selected quantum mechanical methodology offers more accurate results in the gas phase compared to those obtained using various implicit solvation models. This approach has proven as very successful and accurate in our previous studies.15 Structure and parameterization of MAO A. The protein atomic coordinates were taken from the MAO A high-resolution (2.2 Å) crystal structure in complex with harmine,41 as obtained from the Protein Data Bank (access code 2Z5X). The protein structure was modified for the EVB simulations by removing harmine and water molecules, and only side chains with the highest occupancies were considered where multiple conformations were present in the crystal structure. The missing hydrogen atoms were generated using the MOLARIS program package,42 which was also employed for all of the subsequent molecular dynamics (MD) simulations and EVB calculations. By default all histidine residues in MOLARIS are in the Nδ–H tautomeric form (π tautomer) and we employed this setup in

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 31

all enzyme simulations. This choice was checked by inspecting the hydrogen bonding patterns in the vicinity of each His residue, which then provided further validity to such selection. The protonation state of ionizable protein residues was determined through the PROPKA 3.0 server,43 a choice which is further justified by the fact that thus obtained pKa values agree well with our earlier estimates for the MAO A active site residues44 calculated using the semi-microscopic version of the protein dipoles Langevin dipoles model, which treats the protein relaxation in the microscopic framework of the linear response approximation.45 Gauging thus obtained data against the physiological pH = 7.4, it turned out that negatively charged residues were Asp64, Asp339, Glu400 and Glu436, positively charged residues were Arg45, Arg47, Arg51, Lys218, Lys341 and Lys395, while the charge on the flavin moiety was –2. The initial set of atomic charges for noradrenaline and the co-factor moieties was obtained with the Merz-Kollman population analysis at the B3LYP/6–31G(d) level utilizing the conductor-like polarized continuum (CPCM) solvent reaction field model with all parameters for pure water. To address the conformational flexibility of these molecules, the obtained charges were then used in 60 ps molecular dynamics simulations, where a snapshot structure was recorded for every picosecond of the last 10 ps. The same population analysis was then calculated for the obtained ten conformers, and the calculated average charge distribution (see Supporting Information) was employed in all subsequent EVB simulations. This procedure was performed for noradrenaline and either a truncated lumiflavin (used in the gas-phase and water simulations) or the full FAD co-factor (used in protein simulations) in both the reactant and product states. Noradrenaline was manually placed within the enzyme active site in its neutral uncharged form (Figure 1).

8 ACS Paragon Plus Environment

Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 1. Position of the manually placed noradrenaline (NA) within the MAO A active site, with the transferring hydride from the substrate α–CH group and the accepting flavin co-factor N5 atom given in spheres. The subsequently mutated residue Tyr444 is indicated in orange.

EVB Simulation Parameters. EVB calculations were performed in the gas phase (the reference state calibrated to the gas-phase quantum mechanical calculations), as well as in an all-atom representation of the aqueous solution and within the MAO A active site, employing the MOLARIS program together with the ENZYMIX force field.42 The EVB activation barriers were obtained using the standard free energy perturbation/umbrella sampling (FEP/US) approach described in details elsewhere.46 In all cases, the EVB regions were identical and contained the lumiflavin moiety of the FAD co-factor and L-noradrenaline. Two EVB adiabatic states were applied, corresponding to reactants (Michaelis complex) and products of the hydride transfer reaction. The solvent was represented as an all atom surface-constrained model (SCAAS) using a recommended spherical cutoff of 20 Å47 measured from the centroid of the reacting atoms. The all atom system was placed within a 3 Å cubic grid of Langevin dipoles, beyond which the solvent was represented with the water dielectric constant. Harmonic potentials with the force constants of 0.5 kcal mol–1 Å–2 in the

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

protein and 1.0 kcal mol–1 Å–2 in both the gas-phase and aqueous solution were used as positional restraints on the initial position of the each EVB atom to prevent the substrate from leaving the vicinity of the co-factor. To avoid the substrate drift in the free energy perturbation calculations, we introduced a restraint of 3.0 kcal mol–1 Å–2 for distances between the substrate α–carbon atom and the flavin N5 atom greater than 3 Å, in accordance with recommended values,48 and consistent with our previous simulations.15 To assess the conformational flexibility of the substrate within the MAO A active site, we performed a short relaxation and gradual warming molecular dynamics runs, during which the whole system was kept in the initial adiabatic state, employing a three-step procedure: running 20 ps at 30 K, 40 ps at 100 K, and, lastly, 310 ps at 300K. At each picosecond of the final 10 ps of the longest simulation, the geometric coordinates were stored and used for the subsequent reaction simulations, to reduce the error of the manual substrate placement, thus producing 10 independent free energy profiles. The rate-limiting step, corresponding to the substrate–flavin hydride abstraction was simulated by dividing the reaction coordinate to 51 bins (where the ratio of adiabatic state 2 in state 1 is different for each bin) and MD simulations were carried out for a total of 20 ps in each bin, giving rise to a total simulation time of 1020 ps. This simulation protocol was the same for evaluating reaction parameters in all phases (gas-phase, aqueous solution and MAO A) and for determining the effect of the point mutations within the enzyme. In Silico Mutagenesis. In order to calculate the effect of point mutations on the reaction energy profile, the initial atomic positions of one of the relaxed structures was taken and the mutants were generated using UCSF Chimera program, where the highest probability rotamer of Phe, Leu, Trp, Glu and His (in the Nδ–H tautomeric form) was chosen as a replacement for the Tyr444 residue in the wild-type MAO A. A positional restraint was placed on the EVB region, while the rest of the enzyme was allowed to relax using the same three-step procedure as already described. The reaction was then simulated using the same protocol as for the wild-type MAO A.

10 ACS Paragon Plus Environment

Page 11 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

RESULTS AND DISCUSSION QM Energetics of the Gas-Phase Reference Reaction and the EVB Parameterization. We initiated our analysis by performing the quantum-mechanical DFT analysis of the model reaction between the isolated lumiflavin of the FAD co-factor (Figure 2, R = Me, Enz = H) and noradrenaline in the gas-phase with the aim of obtaining reliable thermodynamic parameters to which the EVB Hamiltonian would be parameterized. We have shown in our previous study15 that the initial ratelimiting step of the amine degradation is associated with the transfer of a hydride anion from the α– CH2 group of the substrate to the flavin N5 atom (Figure 2). To subsequently model the hydride transfer step with EVB, we need to know the activation free energy (ΔG‡gas) and the reaction free energy (ΔG°gas) of this reference process. While it is straightforward to calculate the former, the calculation of ΔG°gas is complicated by the fact that, in the gas phase, there is a covalent adduct formation following the H– transfer between the newly created cationic substrate and anionic FADH– co-factor at the C(α)–N(5) distance of 1.6 Å (Figure 3) in agreement with our previous results.14,18 The absence of a stable transient intermediate, which represents the EVB product of the hydride transfer, can be explained by the high energetic penalty of separating the charged species in the gas phase, which is also reflected in the very high endergonicity of this process of 25.2 kcal mol–1 (Figure 2), being almost identical to 24.9 kcal mol–1 found for dopamine.15 This high endergonicity reflects a significant SN1 character of the reaction, which, to a large extent, vanishes by the inclusion of the polar environment. To approximate the relative energy of the transient intermediate we, therefore, selected the point on the IRC profile where the structure of the FADH– group matches the geometry of the gas-phase optimized isolated FADH–. In addition, the IRC profile shows a marked change in the potential energy surface around the same point (Figure 2), indicating a viable transient intermediate. It is worth emphasizing that the calculated activation free energy is insensitive to the choice of the exact energy of the intermediate transient intermediate.15

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

Figure 2. Rate limiting step of the hydride transfer mechanism (left) and the quantum mechanical gas-phase free-energy profile (right, M06–2X/6–311++G(2df,2pd)//M06–2X/6–31+G(d,p) results), with the indicated stationary points corresponding to reactants (R), transition state (TS), transient intermediate (TI) and adduct (AD). The obtained data were later used for the EVB calibration.

Figure 3. Schematic representation (left) and 3D structure (right) of the covalent adduct that is formed between noradrenaline and the flavin co-factor in the gas phase following the hydride transfer (marked AD on the energy profile in Figure 2).

The results we obtained indicate that the transition state and adduct are 34.1 and 10.8 kcal mol–1 higher in energy than the reactants, respectively (Figure 2). The transition state is characterized with one imaginary frequency of υ = 820i cm–1, and N(5)–H and C(α)–H distances of 1.119 and 1.570 Å, respectively. As stated before, the transient intermediate cannot be fully optimized due to the

12 ACS Paragon Plus Environment

Page 13 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

subsequent adduct formation during the optimization, but its energy is approximated at 25.2 kcal mol–1 above the reactants. Therefore, in the absence of the aqueous-phase experimental data, the EVB gas-phase shift (α0) and coupling parameter (Hij) were parameterized to reproduce ΔG‡gas = 34.1 kcal mol–1 and ΔG°gas = 25.2 kcal mol–1, which show high similarity with the values calculated for dopamine within MAO B of 32.8 and 24.9 kcal mol–1, in the same order.15 Thus obtained calibration parameters were α0 = 298.7 kcal mol–1 and Hij = 99.4 kcal mol–1. The system was subsequently moved first to the aqueous solution and then to the MAO A active site using the same parameter set, which is a valid approximation, due to the demonstrated phase-independence of the EVB off-diagonal (Hij) coupling term.49 This approach allows us to (1) explore the effect of changing the environment, and (2) to estimate the enzyme catalytic enhancement relative to the aqueous solution. We would like to stress that, in the enzyme and water, the energies of the TI and AD structures are expected to become close to each other due to the effect of the electrostatic screening of the corresponding environment.

Figure 4. Free-energy profiles for the rate-limiting noradrenaline hydride abstraction in MAO A, gas-phase and aqueous solution, each obtained from 10 independent EVB runs after 1020 ps of simulations. The minima on each graph correspond to EVB diabatic states shown in Figure 2.

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

The Effect of the Aqueous Solution and MAO A Enzyme Environment. The results of EVB simulations are presented in Table 1 and Figure 4. The choice of the point mutations was guided by the experimental work of Edmondson and co-workers50 for qualitative comparison (results are described as discussed in the next section).

Table 1. Activation and reaction free energies for the rate-limiting hydride transfer step of the noradrenaline degradation together with standard errors (in round brackets) obtained by performing 1020 ps of EVB simulations starting from 10 independent initial configurations. System

Transition state (ΔG‡/kcal mol–1) Transient intermediate (ΔG°/kcal mol–1)

Gas

34.1 (1.5)

25.2 (1.3)

Water

33.0 (3.3)

20.3 (2.0)

WT MAO A

18.7 (1.6)

3.8 (0.8)

Tyr444Phe

18.2 (1.4)

1.2 (1.5)

Tyr444Leu

19.3 (0.9)

5.8 (1.7)

Tyr444Trp

20.2 (2.7)

–0.7 (1.7)

Tyr444His(0)a

25.7 (2.5)

6.6 (1.4)

Tyr444His(+)

34.9 (2.6)

16.5 (1.8)

Tyr444Glu(0)

23.8 (1.3)

5.7 (2.7)

Tyr444Glu(–)

21.7 (1.3)

–1.5 (2.6)

a

Nδ–H tautomeric form of the neutral His residue was used

It turns out that the reaction in the aqueous solution is associated with the EVB free energy barrier of 33.0 kcal mol–1, which is only slightly lower compared to the corresponding gas-phase value of 34.1 kcal mol–1. This reduction of only 1.1 kcal mol–1 is significantly smaller than that observed for dopamine, where it is 6.3 kcal mol–1,15 but the absolute value of the calculated activation free energy

14 ACS Paragon Plus Environment

Page 15 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

itself seems to fall in the right range. Specifically, studies of the hydride transfer on a series of NAD+ analogues in the aqueous solution have shown that the typical free energy barrier for hydride transfer is around 20 kcal mol–1.51,52 However, while flavin and NAD+ analogues are both very good hydride ion acceptors, NADH is significantly a better hydride donor than the α–CH2 group of noradrenaline.53 On this basis, we expect the barrier for the hydride transfer between aliphatic amines and FAD in water must be considerably higher than 20 kcal mol–1, which is the case here. Taking the reaction into the MAO A environment, there is a further reduction of the barrier to 18.7 kcal mol–1, while the reaction is still endergonic with the reaction free energy of 3.8 kcal mol–1 (Table 1, Figure 5). We have to mention that several experimental studies54,55 and our previous computational results56 have agreed in showing that MAO substrates are likely bound to the active site in their protonated monocationic forms, which is usually the most abundant form of monoamines under physiological conditions, but the neutral form of the substrate is mandatory for the hydride abstraction reaction.14,54,55,57 This suggests that a substrate deprotonation within the enzyme active site must take place prior to the reaction, which could be accomplished by several water molecules present in the enzyme, and which is associated with some energy cost to arrive at a substrate reactive form. In our previous study that focused on the pKa values of the active site residues before and after the substrate binding,56 we have demonstrated that the pKa value of the dopamine bound within MAO B active site (pKa = 8.8) is practically unchanged compared to its value in the aqueous solution (pKa = 8.9).58 This led us to safely estimate that the matching noradrenaline pKa value within the MAO active site would closely resemble its value in water (pKa = 8.6),58 implying that it would require 1.6 kcal mol–1 at the physiological pH of 7.4 to deprotonate monocationic noradrenaline to its uncharged reactive form, as calculated using ΔGdeprotonation = ln(10)·kBT·(pKa – pH). This value, when added to the calculated reaction barrier of 18.7 ± 1.6 kcal mol–1, gives the total activation free energy of 20.3 ± 1.6 kcal mol–1. The latter value is in reasonably good agreement with the extrapolated

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

experimental value of 16.5 kcal mol–1, thus providing some evidence in favor of the proposed hydride transfer mechanism.

Figure 5. Relevant structures selected to most closely correspond to the corresponding stationary points during a single EVB run. The transferring hydride anion is shown as a white sphere.

In the present study we did not consider the nuclear quantum effects associated with the hydride transfer13,23,59 that are inherent to experimental activation free energies. Experimentally measured values for the H/D kinetic isotope effect between 6–13 for various monoamines in MAO A25 indicate that the quantum nature of the nuclear motion is relevant for this enzyme.59 We thus expect that the quantization of the nuclear motion in simulations would give a somewhat lower barrier and a potentially better agreement with the extrapolated experimental activation free energy would be found. Reproduction of the MAO A kinetic isotope effect by the path integration methodology59 or implicit quantization60 remains a challenge for future. In concluding this section, let us mention that the reduction in the activation barrier compared to the aqueous solution is 14.3 kcal mol–1, which corresponds to a rate-enhancement of over 10 orders

16 ACS Paragon Plus Environment

Page 17 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

of magnitude. This implies that MAO A is more efficient in catalyzing noradrenaline degradation than MAO B is with dopamine, where a rate-enhancement of around 9 orders of magnitude was observed,15 but a larger activation free energy of 20.3 ± 1.6 kcal mol–1, compared to 16.1 kcal mol–1 for MAO B, suggests MAO A isoform is, overall, a slower enzyme in metabolizing noradrenaline than MAO B is for dopamine. The difference between experimental and calculated activation free energies of 3.8 kcal mol–1 deserves some discussion. Firstly, we would like to emphasize that there is some space for improvements in the determination of the experimental data, since the existing values were not obtained from the temperature-dependent experiments, which would give rise to Arrhenius plots, from which the activation parameters would be obtained in a straightforward and unambiguous way. At the computational side, the main drawback is the lack of experimental data for the reference reaction in aqueous solution, which led us to proceed with the quantum-chemical study of the reference reaction in the gas phase. Having used a generally quite accurate DFT functional,40 we feel that the calculated gas phase barrier is still associated with around 2 kcal mol–1 of uncertainty. Imperfect force fields and very long reorientational correlation times of the reacting species along with inherent errors in both the initial substrate placement and positional constraints on the reactive site atoms, could lead to probably not completely converged free energy calculations, and are an additional possible source of error in the calculated activation free energies within enzymes. The sampling of the intermolecular degrees of freedom between the co-factor moiety and the substrate is especially problematic. Demanding statistical sampling is inherent to both ab initio- and EVBQM/MM calculations of enzyme reaction profiles.61 Still, it is worth to emphasize that the applied EVB computational methodology represents a state-of-the-art approach and that ab initio QM/MM approaches are subject to somewhat worse converged free energy reaction profiles because of usually much shorter simulation times.

17 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

In Silico Mutagenesis Studies. To further validate our reaction mechanism, we decided to proceed with the calculation of the reaction parameters with several mutant enzymes at the position of the "aromatic cage" Tyr444 residue (Figure 1), which was subsequently mutated to His, Phe, Leu, Trp and Glu side-chains and the effect of these modifications on the reaction energies evaluated (Table 1). The choice of the investigated point-mutations was motivated by the experimental work of Edmondson and co-workers,50 who successfully expressed the corresponding mutant enzymes of MAO B in Pichia pastoris and measured kcat values for two benzylamine- and three phenylethylamine substrates. The same attempt to do so for MAO A resulted in enzymes found to be unstable upon membrane extraction. Initial placement of the substrate in the studied mutants is shown in Supporting Information Figure S1. Our analysis confirmed the functional importance of the probed tyrosine residue, since mutations to Leu, Trp, His and Glu produced enzymes that are slower, as being associated with higher activation free energies (Table 1), which is in good agreement with experiments.50 On the other hand, Tyr444Phe mutation slightly lowered the activation free-energy, which seems to indicate that the catalytic role of the Tyr444 residue is predominantly achieved through its aromatic benzene moiety, rather than through the hydroxyl –OH group, which is a significant observation. This trend also agrees with the mentioned measurements,50 which revealed the increased kcat values for p-CF3- and p-NO2-benzylamines and p-NO2-phenylethylamine in the analogous Tyr435Phe MAO B mutant relative to the wild-type enzyme, although, at the same time, somewhat lower kcat values were reported for the unsubstituted benzylamine and phenylethylamine. Overall, the efficiency of the wild-type MAO A and its Tyr444 mutants, as evidenced from our results with the noradrenaline substrate is WT ≈ Tyr444Phe > Tyr444Leu > Tyr444Trp > Tyr444His, which could be closely related to the mentioned experimental results in MAO B with phenylethylamine50 that read WT > Tyr435Phe ≈ Tyr435Leu > Tyr435His > Tyr435Trp. This qualitative agreement is even more remarkable knowing that experimental steady-state

18 ACS Paragon Plus Environment

Page 19 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

measurements in MAO B include not only the hydride transfer, a reductive half-reaction, but also the process of flavin co-factor regeneration to its reduced form assisted by molecular oxygen (FADH2 → FAD), an oxidative half-reaction, which was shown to have some influence on the overall MAO B rate.62 Another interesting aspect is provided with His and Glu mutants regarding their protonation forms. It turns out that Tyr444His(0) mutant, where His residue is unionized, gives an enzyme with 7 kcal mol–1 higher activation free-energy than WT, which is even further increased by another 9 kcal mol–1 to ΔG‡ = 34.9 kcal mol–1 in Tyr444His(+) upon protonating the histidine residue, thus even exceeding the value for the aqueous solution (Table 1). This trend is a result of a positive charge build-up on the substrate in the transition state, thus agreeing with the idea that during reaction the departing hydrogen is abstracted as a hydride anion. Therefore, positively charged species near the active site have an anti-catalytic effect as indicated by our results. It remains a challenge to experimentally check if the Tyr444His mutant has a very low turnover under acidic pH values. On the other hand, we also investigated Tyr444Glu mutants with either unionized or negatively charged glutamic acid residues (Table 1). For the neutral Tyr444Glu(0), the barrier is higher than for WT, being ΔG‡ = 23.8 kcal mol–1, but is significantly reduced to ΔG‡ = 21.7 kcal mol–1, when Glu is deprotonated to –CH2–CH2–COO–. The induced negative charge stabilizes the cationic transition state, which lowers the barrier. Taken all together, the results for all mutant enzymes studied in this work are consistent with the proposed hydride transfer and provide further confirmation that both MAO A and MAO B enzymes operate through the same reaction mechanism.

19 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 31

CONCLUSIONS By using the empirical valence bond QM/MM method, we demonstrate that MAO A catalyzes noradrenaline degradation by lowering the activation barrier for the rate-limiting hydride transfer from the substrate α–CH2 group to the N(5) atom of its flavin co-factor. The enzyme reaction represents a rate-enhancement of over 10 orders of magnitude compared to the corresponding reaction in the aqueous solution (a reduction of the corresponding barrier by 14.3 kcal mol–1). The calculated activation free energy in the enzyme of ΔG‡ = 20.3 ± 1.6 kcal mol–1 is in a reasonable qualitative agreement with the extrapolated experimental value of 16.5 kcal mol–1, which together with our previous results on the MAO B catalyzed dopamine decomposition (ΔG‡CALC = 16.1 kcal mol–1, ΔG‡EXP = 16.5 kcal mol–1)15 indicates that both MAO isoforms function by the same direct two-step hydride transfer mechanism.14,18 In silico mutagenesis emphasized the functional importance of the "aromatic cage" Tyr444 residue and suggested that its hydroxyl group is less important than the benzene moiety for its role in catalysis. It remains a challenge to link MAO point mutations with neurological diseases and changes in the efficiency of the MAO inhibition.63 In conjunction with previous experimental and computational work, the data presented here improve the understanding of the mechanism of the catalytic activity of the MAO family of enzymes, which can allow for the design of novel and improved MAO inhibitors for neuroprotective use, such as transition-state analogues.

20 ACS Paragon Plus Environment

Page 21 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

ASSOCIATED CONTENT Supporting Information Available. Determination of the reliable experimental activation free energy for the noradrenaline degradation within MAO A enzyme from the various values reported in the literature. Parameterization charges, bonding and van der Waals parameters of the FAD cofactor, lumiflavin moiety and noradrenaline molecule for EVB simulations. Figure S1 showing the initial placement of the substrate in the studied mutants. This material is available free of charge via the Internet.

AUTHOR INFORMATION Corresponding Authors * E-mail: [email protected] (Robert Vianello) * E-mail: [email protected] (Janez Mavri) Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT R.V. gratefully acknowledges the European Commission for an individual FP7 Marie Curie Career Integration Grant (contract number PCIG12–GA–2012–334493). J.M. would like to thank the Slovenian Research Agency for financial support within the framework of the Program Group P10012.

21 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 31

REFERENCES

1

Harsing, L. G. In Handbook of Neurochemisty and Molecular Neurobiology: Neurotransmitter Systems; Lajtha, A., Vizi, S. E., Eds.; 3rd Ed.; Springer: New York, 2008; pp 151–170.

2

Ramsay, R. R. Molecular Aspects of Monoamine Oxidase B. Prog. Neuropsychopharmacol. Biol. Psychiatry. 2016, 69, 81–89 and references cited therein.

3

Husain, M.; Edmondson, D. E.; Singer, T. P. Kinetic Studies on the Catalytic Mechanism of Liver Monoamine Oxidase. Biochemistry 1982, 21, 595–600.

4

Pavlin, M.; Repič, M.; Vianello, R.; Mavri, J. The Chemistry of Neurodegeneration: Kinetic Data and their Implications. Mol. Neurobiol. 2016, 53, 3400–3415.

5

Strolin Benedetti, M.; Tipton, K. F. Monoamine Oxidases and Related Amine Oxidases as Phase I Enzymes in the Metabolism of Xenobiotics. J. Neural Transm. 1998, 52, 149–171.

6

Youdim, M. B. H.; Edmondson, D. E.; Tipton, K. F. The Therapeutic Potential of Monoamine Oxidase Inhibitors. Nat. Rev. Neurosci. 2006, 7, 295–309.

7

Fišar, Z. Drugs Related to Monoamine Oxidase Activity. Prog. Neuropsychopharmacol. Biol. Psychiatry. 2016, 69, 112–124.

8

Gadda, G. Oxygen activation in Flavoprotein Oxidases: the Importance of Being Positive. Biochemistry 2012, 51, 2662–2669.

9

Edmondson, D. E. Hydrogen Peroxide Produced by Mitochondrial Monoamine Oxidase Catalysis: Biological Implications. Curr. Pharm. Des. 2014, 20, 155–160.

22 ACS Paragon Plus Environment

Page 23 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

10

The Journal of Physical Chemistry

Akyüz, M. A.; Erdem, S. S.; Edmondson, D. E. The Aromatic Cage in the Active Site of Monoamine Oxidase B: Effect on the Structural and Electronic Properties of Bound Benzylamine and p-Nitrobenzylamine. J. Neural Transm. 2007, 114, 693–698.

11

Gaweska, H.; Fitzpatrick, P. F. Structures and Mechanism of the Monoamine Oxidase Family. Biomol. Concepts 2011, 2, 365–377.

12

Orru, R.; Aldeco, M.; Edmondson, D. E. Do MAO A and MAO B Utilize the same Mechanism for the C–H Bond Cleavage Step in Catalysis? Evidence Suggesting Differing Mechanisms. J. Neural Transm. 2013, 120, 847–851.

13

Abad, E.; Zenn, R. K.; Kästner, J. Reaction Mechanism of Monoamine Oxidase From QM/MM Calculations. J. Phys. Chem. B 2013, 117, 14238–14246.

14

Vianello, R.; Repič, M.; Mavri, J. How are Biogenic Amines Metabolized by Monoamine Oxidases? Eur. J. Org. Chem. 2012, 7057–7065.

15

Repič, M.; Vianello, R.; Purg, M.; Duarte, F.; Bauer, P.; Kamerlin, S. C. L.; Mavri, J. Empirical Valence Bond Simulations of the Hydride Transfer Step in the Monoamine Oxidase B Catalyzed Metabolism of Dopamine. Proteins: Struct., Funct., Bioinf. 2014, 82, 3347–3355.

16

De Colibus, L.; Li, M.; Binda, C.; Lustig, A.; Edmondson, D. E.; Mattevi, A. ThreeDimensional Structure of Human Monoamine Oxidase A (MAO A): Relation to the Structures of Rat MAO A and Human MAO B. P. Natl. Acad. Sci. U. S. A. 2005, 102, 12684–12689.

17

Binda, C.; Hubalek, F.; Li, M.; Edmondson, D. E.; Mattevi, A. Crystal Structure of Human Monoamine Oxidase B, a Drug Target Enzyme Monotopically Inserted into the Mitochondrial Outer Membrane. FEBS Lett. 2004, 564, 225–228.

23 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

18

Page 24 of 31

Vianello, R.; Domene, C.; Mavri, J. The Use of Multiscale Molecular Simulations in Understanding a Relationship between the Structure and Function of Biological Systems of the Brain: The Application to Monoamine Oxidase Enzymes. Front. Neurosci. 2016, 10, 327.

19

Edmondson, D. E.; Binda, C.; Wang, J.; Upadhyay, A. K.; Mattevi, A. Molecular and Mechanistic Properties of the Membrane-Bound Mitochondrial Monoamine Oxidases. Biochemistry 2009, 48, 4220–4230.

20

Akyüz, M. A.; Erdem, S. S. Computational Modeling of the Direct Hydride Transfer Mechanism for the MAO Catalyzed Oxidation of Phenethylamine and Benzylamine: ONIOM (QM/QM) Calculations. J. Neural Transm. 2013, 120, 937–945.

21

Atalay, V. E.; Erdem, S. S. A Comparative Computational Investigation on the Proton and Hydride Transfer Mechanisms of Monoamine Oxidase Using Model Molecules. Comp. Biol. Chem. 2013, 47, 181–191.

22

Zapata-Torres, G.; Fierro, A.; Barriga-González; G., Salgado; J. C., Celis-Barros, C. Revealing Monoamine Oxidase B Catalytic Mechanisms by Means of the Quantum Chemical Cluster Approach. J. Chem. Inf. Model. 2015, 55, 1349–1360.

23

Zenn, K. R.; Abad, E.; Kästner, J. Influence of the Environment on the Oxidative Deamination of p-Substituted Benzylamines in Monoamine Oxidase. J. Phys. Chem. B 2015, 119, 3678– 3686.

24

Erdem, S. S.; Karahan, O.; Yildiz, I.; Yelekçi, K. A Computational Study on the AmineOxidation Mechanism of Monoamine Oxidase: Insight into the Polar Nucleophilic Mechanism. Org. Biomol. Chem. 2016, 4, 646–658.

24 ACS Paragon Plus Environment

Page 25 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

25

The Journal of Physical Chemistry

Miller, J. R.; Edmondson, D. E. Structure-Activity Relationships in the Oxidation of paraSubstituted Benzylamine Analogues by Recombinant Human Liver Monoamine Oxidase A. Biochemistry 1999, 38, 13670–13683.

26

Kuhar, M. J.; Couceyro, P. R.; Lambert, P. D. In Catecholamines in Basic Neurochemistry: Molecular, Cellular and Medical Aspects; Siegel, G. J., Agranoff, B. W., Albers, R. W., Fisher, S. K., Uhler, M. D., Eds.; 6th Ed.; Lippincott-Raven: Philadelphia, 1999.

27

Gasior, M.; Bergman, J.; Kallman, M. J.; Paronis, C. A. Evaluation of the Reinforcing Effects of Monoamine Reuptake Inhibitors under a Concurrent Schedule of Food and I.V. Drug Delivery in Rhesus Monkeys. Neuropsychopharmacology 2005, 30, 758–764.

28

Sara, S. J. Locus Coeruleus in Time with the Making of Memories. Curr. Opin. Neurobiol. 2015, 35, 87–94.

29

Heneka, M. T.; Nadrigny, F.; Regen, T.; Martinez-Hernandez, A.; Dumitrescu-Ozimek, L.; Terwel, D.; Jardanhazi-Kurutz, D.; Walter, J.; Kirchhoff, F.; Hanisch, U. K., et al. Locus Ceruleus Controls Alzheimer's Disease Pathology by Modulating Microglial Functions Through Norepinephrine. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 6058–6063.

30

O'Carroll, A. M.; Bardsley, M. E.; Tipton, K. F. The Oxidation of Adrenaline and Noradrenaline by the Two Forms of Monoamine Oxidase from Human and Rat Brain. Neurochem. Int. 1986, 8, 493–500.

31

Cruz, F.; Edmondson, D. E. Kinetic Properties of Recombinant MAO-A on Incorporation into Phospholipid Nanodisks. J. Neural Transm. 2007, 114, 699–702.

25 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

32

Page 26 of 31

Wang, J.; Edmondson, D. E. Do Monomeric vs. Dimeric Forms of MAO-A Make a Difference? A Direct Comparison of the Catalytic Properties of Rat and Human MAO-A's. J. Neural Transm. 2007, 114, 721–724.

33

Vintém, A. P. B.; Price, N. T.; Silverman, R. B.; Ramsay, R. R. Mutation of Surface Cysteine 374 to Alanine in Monoamine Oxidase A Alters Substrate Turnover and Inactivation by Cyclopropylamines. Bioorg. Med. Chem. 2005, 13, 3487–3495.

34

Li, M.; Hubálek, F.; Newton-Vinson, P.; Edmondson, D. E. High-Level Expression of Human Liver Monoamine Oxidase A in Pichia Pastoris: Comparison with the Enzyme Expressed in Saccharomyces Cerevisiae. Protein Expr. Purif. 2002, 24, 152–162.

35

Nandigama, R. K.; Edmondson, D. E. Structure-Activity Relations in the Oxidation of Phenethylamine Analogues by Recombinant Human Liver Monoamine Oxidase A. Biochemistry 2000, 39, 15258–15265.

36

Warshel, A.; Weiss, R. M. An Empirical Valence Bond Approach for Comparing Reactions in Solutions and in Enzymes. J. Am. Chem. Soc. 1980, 102, 6218–6226.

37

Warshel, A.; Weiss, R. M. Empirical Valence Bond Calculations of Enzyme Catalysis. Ann. N. Y. Acad. Sci. 1982, 367, 370–382.

38

Shurki, A.; Derat, E.; Barrozo, A.; Kamerlin, S. C. L. How Valence Bond Theory Can Help You Understand Your (Bio)Chemical Reaction. Chem. Soc. Rev. 2015, 44, 1037–1052.

39

Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A., et al. Gaussian 09, Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009.

26 ACS Paragon Plus Environment

Page 27 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

40

The Journal of Physical Chemistry

Zhao, Y.; Truhlar, D. G. The M06 Suite of Density Functionals for Main Group Thermochemistry, Thermochemical Kinetics, Noncovalent Interactions, Excited States, and Transition Elements: Two New Functionals and Systematic Testing of Four M06-Class Functionals and 12 Other Functionals. Theor. Chem. Acc. 2008, 120, 215–241.

41

Son, S. Y.; Ma, J.; Kondou, Y.; Yoshimura, M.; Yamashita, E.; Tsukihara, T. Structure of Human Monoamine Oxidase A at 2.2 Å Resolution: The Control of Opening the Entry for Substrates/Inhibitors. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 5739–5744.

42

Lee, F. S.; Chu, Z. T.; Warshel, A. Microscopic and Semimicroscopic Calculations of Electrostatic Energies in Proteins by the POLARIS and ENZYMIX Programs. J. Comput. Chem. 1993, 14, 161–185.

43

Li, H.; Robertson, A. D.; Jensen, J. H. Very Fast Empirical Prediction and Rationalization of Protein pKa Values. Proteins: Struct., Funct., Bioinf. 2005, 61, 704–721.

44

Repič, M.; Purg, M.; Vianello, R.; Mavri, J. Examining Electrostatic Preorganization in Monoamine Oxidases A and B by Structural Comparison and pKa Calculations. J. Phys. Chem. B 2014, 118, 4326–4332.

45

Sham, Y. Y.; Chu, Z. T.; Warshel, A. Consistent Calculations of pKa’s of Ionizable Residues in Proteins: Semi-Microscopic and Microscopic Approaches. J. Phys. Chem. B 1997, 101, 4458–4472.

46

Kamerlin, S. C. L.; Warshel, A. The Empirical Valence Bond Model: Theory and Applications. Wiley Interdiscip. Rev. Comput. Mol. Sci. 2011, 1, 30–45.

27 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

47

Page 28 of 31

Warshel, A.; King, G. Polarization Constraints in Molecular Dynamics Simulation of Aqueous Solutions: The Surface Constraint All Atom Solvent (SCAAS) Model. Chem. Phys. Lett. 1985, 121, 124–129.

48

Gumbart, J. C.; Roux, B.; Chipot, C. Efficient Determination of Protein–Protein Standard Binding Free Energies from First Principles. J. Chem. Theory Comput. 2013, 9, 3789–3798.

49

Hong, G.; Rosta, E.; Warshel, A. Using the Constrained DFT Approach in Generating Diabatic Surfaces and Off Diagonal Empirical Valence Bond Terms for Modeling Reactions in Condensed Phases. J. Phys. Chem. B 2006, 110, 19570–19574.

50

Li, M.; Binda, C.; Mattevi, A.; Edmondson, D. E. Functional Role of the "Aromatic Cage" in Human Monoamine Oxidase B: Structures and Catalytic Properties of Tyr435 Mutant Proteins. Biochemistry 2006, 45, 4775–4784.

51

Kong, Y. S.; Warshel, A. Linear Free-Energy Relationships with Quantum-Mechanical Corrections - Classical and Quantum-Mechanical Rate Constants for Hydride Transfer Between NAD(+) Analogs in Solutions. J. Am. Chem. Soc. 1995, 117, 6234–6242.

52

Kreevoy, M. M.; Ostovic, D.; Lee, I. S. H.; Binder, D. A.; King, G. W. Structure Sensitivity of the Marcus λ for Hydride Transfer between NAD+ Analogues. J. Am. Chem. Soc. 1988, 110, 524–530.

53

Horn, M.; Schappele, L. H.; Lang-Wittkowski, G.; Mayr, H.; Ofial, A. R. Towards a Comprehensive Hydride Donor Ability Scale. Chem. Eur. J. 2013, 19, 249–263.

28 ACS Paragon Plus Environment

Page 29 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

54

The Journal of Physical Chemistry

Jones, T. Z.; Balsa, D.; Unzeta, M.; Ramsay, R. R. Variations in Activity and Inhibition with pH: The Protonated Amine is the Substrate for Monoamine Oxidase, but Uncharged Inhibitors Bind Better. J. Neural Transm. 2007, 114, 707–712.

55

Dunn, R. V.; Marshall, K. R.; Munro, A. W.; Scrutton, N. S. The pH Dependence of Kinetic Isotope Effects in Monoamine Oxidase A Indicates Stabilization of the Neutral Amine in the Enzyme-Substrate Complex. FEBS J. 2008, 275, 3850–3858.

56

Borštnar, R.; Repič, M.; Kamerlin, S. C. L.; Vianello, R.; Mavri, J. Computational Study of the pKa Values of Potential Catalytic Residues in the Active Site of Monoamine Oxidase B. J. Chem. Theory Comput. 2012, 8, 3864–3870.

57

Tan, A. K.; Ramsay, R. R. Substrate-Specific Enhancement of the Oxidative Half-Reaction of Monoamine Oxidase. Biochemistry 1993, 32, 2137–2143.

58

Williams, D. A.; Lemke, T. L. Foye's Principles of Medicinal Chemistry; 6th Ed.; Lippincott Williams & Wilkins: Philadelphia, 2008.

59

Mavri, J.; Matute, R. A.; Chu, Z. T.; Vianello, R. Path Integral Simulation of the H/D Kinetic Isotope Effect in Monoamine Oxidase B Catalyzed Decomposition of Dopamine. J. Phys. Chem. B 2016, 120, 3488−3492.

60

Kržan, M.; Vianello, R.; Maršavelski, A.; Repič, M.; Zakšek, M.; Kotnik, K.; Fijan, E.; Mavri, J. The Quantum Nature of Drug-Receptor Interactions: Deuteration Changes Binding Affinities for Histamine Receptor Ligands. PLoS ONE 2016, 11, e0154002.

61

Bolhuis, P. G.; Chandler, D.; Dellago, C.; Geissler, P. L. Transition Path Sampling: Throwing Ropes over Rough Mountain Passes, in the Dark. Annu. Rev. Phys. Chem. 2002, 53, 291–318.

29 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

62

Page 30 of 31

Reid McDonald, G.; Olivieri, A.; Ramsay, R. R.; Holt, A. On the Formation and Nature of the Imidazoline I2 Binding Site on Human Monoamine Oxidase-B. Pharm. Res. 2010, 62, 475– 488.

63

Geha, R. M.; Rebrin, I.; Chen, K.; Shih, J. C. Substrate and Inhibitor Specificities for Human Monoamine Oxidase A and B are Influenced by a Single Amino Acid. J. Biol. Chem. 2011, 276, 9877–9882.

30 ACS Paragon Plus Environment

Page 31 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Table of Contents Entry

31 ACS Paragon Plus Environment