Subscriber access provided by UNIV OF MISSISSIPPI

Article

Insights into the Lactonase Mechanism of Serum Paraoxonase 1 (PON1): Experimental and Quantum Mechanics/Molecular Mechanics (QM/MM) Studies Quang Anh Tuan Le, Seonghoon Hwan Kim, Rakwoo Chang, and Yong Hwan Kim J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.5b03184 • Publication Date (Web): 06 Jul 2015 Downloaded from http://pubs.acs.org on July 11, 2015

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 38

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

Insights into the Lactonase Mechanism of Serum Paraoxonase 1 (PON1): Experimental and Quantum Mechanics/Molecular Mechanics (QM/MM) Studies

Quang Anh Tuan Le1, Seonghoon Kim2, Rakwoo Chang2*and Yong Hwan Kim1* 1

Department of Chemical Engineering, Kwangwoon University, Seoul 139-701, Republic of Korea 2

Department of Chemistry, Kwangwoon University, Seoul 139-701, Republic of Korea

*Corresponding authors: 1

Yong Hwan Kim

Tel.: + 82-2- 940-5675; Fax +82-2-941-1785 E-mail address: [email protected] 2

Rakwoo Chang

Tel.: + 82-2- 940-5243 E-mail address: [email protected]

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

Abstract Serum paraoxonase 1 (PON1) is a versatile enzyme for the hydrolysis of various substrates (e.g., lactones, phosphotriesters) and for the formation of a promising chemical platform γ-valerolactone. Elucidation of the PON1-catalyzed lactonase reaction mechanism is very important for understanding the enzyme function and for engineering this enzyme for specific applications. Kinetic study and hybrid quantum mechanics/molecular mechanics (QM/MM) method were used to investigate the PON1-catalyzed lactonase reaction of γbutyrolactone (GBL) and (R)-γ-valerolactone (GVL). The activation energies obtained from the QM/MM calculations were in good agreements with the experiments. Interestingly, the QM/MM energy barriers at MP2/3-21G(d,p) level for the lactonase of GVL and GBL were respectively 14.3–16.2 and 11.5–13.1 kcal/mol, consistent with the experimental values (15.57 and 14.73 kcal/mol derived from respective kcat values of 36.62 and 147.21 s-1). The QM/MM energy barriers at MP2/6-31G(d) and MP2/6-31G(d,p) levels were also in relatively good agreements with the experiments. Importantly, the difference in the QM/MM energy barriers at MP2 level with all investigated basis sets for the lactonase of GVL and GBL were in excellent agreement with the experiments (0.9–3.1 and 0.8 kcal/mol, respectively). A detailed mechanism for the PON1-catalyzed lactonase reaction was also proposed in this study.

Key words: ab initio Quantum Mechanics/Molecular Mechanics, serum paraxonase 1, PON1, lactonase, γ-valerolactone, γ-butyrolactone

2

ACS Paragon Plus Environment

Page 2 of 38

Page 3 of 38

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

1.

Introduction Serum paraoxonase 1 (PON1) is a mammalian enzyme that catalyzes the hydrolysis of a

wide range of substrates such as thiolactones, lactones, esters, and phosphotriesters. The name of the enzyme was originally derived from paraoxon, a metabolite of pesticide parathion, which is hydrolyzed by PON1.1-2 PON1 has been intensively investigated because of its ability to inactivate various organophosphates (e.g., paraoxon, sarin, VX, and soman) as environmental risk and terrorist threat.3-5 Recently, it has been found that PON1 also catalyzes the formation of lactones such as γ-butyrolatone, ߜ-hexalactone, mevalonic lactone,6 and γ-valerolactone which is a promising platform molecule for the synthesis of chemicals and fuels.7 The promiscuity of PON1 has thus been an attractive target not only to address environmental concerns but also for novel chemical syntheses. A structure-reactivity study of PON1 involving 50 substrates of 3 different classes (i.e., lactones, esters and phosphotriesters) suggested that lactone is the native substrate of PON1 and that the native activity of PON1 is lactonase.3 Therefore, elucidation of the lactonase mechanism catalyzed by PON1 is one of the main objectives required to understanding the characteristics and functions of PON1. It has been demonstrated that PON1 is highly dependent on the calcium ion for its stability and activity toward the hydrolysis of phenyl acetate. One calcium ion in the center of PON1 was proposed as catalytic calcium and another one in the bottom of protein was proposed as structural calcium.2, 8 The structural information and extensive site-directed mutagenesis studies on residues close to the catalytic calcium proposed that residue His115 with a distorted dihedral angle, which is usually observed in catalytic residues of many other enzymes, functions as one of catalytic residues for lactonase and esterase reactions catalyzed by PON1. His134, which is in close proximity to His115, also showed a significant impact on lactonase and esterase activity, and it was proposed as another catalytic residue for the lactonase and esterase reactions 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

catalyzed by PON1. In a simple proposed mechanism, His115 functions as a general base to deprotonate a reactive water to generate hydroxide ion, which then attacks the substrate. Additionally, His134 functions as a proton shuttle to increase the basicity of His115.1-2 A more detailed mechanism for lactonase of PON1 was proposed as a result of molecular docking studies of the crystal structure of PON1 with various ligands (i.e., substrate, putative intermediate and product). In this mechanism, His115 and Glu53 both function as general bases to generate an attacking hydroxide ion. The protonated form of Asp269 functions as a proton donor to facilitate the protonation of the alkoxide leaving group. As seen from this mechanism, His115 and Glu53 retain a proton at the end of the reaction. Meanwhile, Asp269 remains deprotonated state in the enzyme-product complex.9 The state of the ionizable residues (i.e., His115, Glu53, and Asp269) in the enzyme-substrate complex is different from that of the enzyme-product complex. In other words, the state of the free enzyme does not regenerate after one round of reaction. The docking study of the enzyme with different substrate and putative intermediates is not sufficient to obtain a reliable and clear mechanism for the hydrolysis of lactones by PON1. Thus, the more advanced techniques should be done to gain a better understanding of the mechanism for this reaction. For the past few decades, hybrid Quantum Mechanics/Molecular Mechanics (QM/MM) calculations have been shown to be a valuable tool for elucidating mechanisms of many enzyme-catalyzed reactions.10-14 Mechanistic insights derived from QM/MM calculations provide detailed information that permits a better understanding of the roles of participating residues in the reaction. In this study, a kinetic study and an ab initio QM/MM calculation was used to elucidate the reaction mechanism for the hydrolysis of two 5-membered-ring lactones (GBL and (R)-γvalerolactone (GVL)) and to identify the catalytic role of the second residue in the histidine dyad, His134. Ab initio Hartree-Fock methods15 with 3-21G(d,p) and 6-31G(d,p) basis sets 4

ACS Paragon Plus Environment

Page 4 of 38

Page 5 of 38

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

have been used for QM region to identify the reaction geometry. A higher quantum theory (i.e., Møller–Plesset perturbation theory (MP2)15 with various basis sets) has also been used for single point energy calculations. The results showed a high correlation between the computational method and the experiment. Our calculations also revealed that His134 stabilizes the reaction transition state, which leads to a decrease in the activation energy of the lactonase reaction. 2.

Materials and Methods

2.1. Chemicals The G3C9 PON1 gene was synthesized from Genescript (New York, USA). The vector pET22B (+) and E. coli Origami B (DE3) were purchased from Novagen (Darmstadt, Germany). The restrictive enzymes NdeI and EcoRI were purchased from Takara (Shiga, Japan). The lactones (i.e., GBL and GVL) were purchased from Sigma Aldrich (St. Louis, MO, United States of America) 2.2. Methods 2.2.1. Computational Methods 2.2.1.1. Setting up the Enzyme-substrate system The lactonase reaction is described in Scheme 1. To set up enzyme-substrate (ES) complex, γ-lactones (i.e., GBL and GVL) were firstly optimized at HF/6-31G(d) level in Gaussian 03,16 and then were docked to PON1 (i.e., PDB code 3SRG with resolution 2.19 Å)9 using Autodock 4.2 program.17 The docking process for the γ-lactones was performed in a grid of 36x30x30 points centered at active site of PON1 with a grid spacing of 0.375Å. The Lamarckian Genetic Algorithm (LGA) was applied for docking process with 100 trials and the maximum number of energy evaluation was set to 2.5*107. The charge of calcium of PON1 was set to +2. The default values were used for other parameters during molecular docking. The docked structures were sorted into conformationally similar bins based on their 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

root mean square deviation (RMSD) tolerance (2 Å).17 The docked structures of the both γlactones, which don’t possess any torsional degree of freedom, oriented in only one configuration inside the active site of PON1 for each lactone. The binding energy of the docked structures and PON1 were almost identical in each lactone. In addition, the orientation of the lactones was very similar to that of the inhibitor quinolin-2(1H)-one (OCH) in PON1 crystal structure.9 The docked structure which has the lowest energy in the cluster was selected for setting up ES complex system. In the PON1 crystal structure, the distance between oxygen of crystallographic water (i.e., WAT 359) and the catalytic calcium ion was 2.4 Å. On the other hand, the distance between the oxygen of WAT359 and an inhibitor carboxyl carbon atom oriented as a substrate for lactonase was quite far (5.05 Å). We set up two ES complex systems to determine whether WAT359 stabilizes the catalytic calcium or participate as a reactant in the lactonase reaction. The first ES complex system (system 1) consisted only of enzyme and two substrates (i.e., lactone and WAT359). The second ES complex system (system 2) was constructed by adding one additional water molecule (WAT363) (where the distance between the oxygen of WAT363 and the lactone carboxyl carbon is approximately 3 Å) to the first ES complex system. The following steps for the preparation of the two ES complex systems were identical. Hydrogen atoms were added to the ES complex using the HBUILD module of CHARMM.18 All protein atoms in the MM region are described by the CHARMM force field (i.e., CHARMM force field 36 for protein atoms).18–19 The topology and parameter files for the lactones were generated by The CHARMM General Force Field using paramchem,20 which involved in CHARMM-GUI.21 The protonation state of titratable residues (i.e., Glu, Asp, Lys, and His residues) of PON1 was determined by proPKa3.0 program22 in the PDB2PQR server at pH=7.0. His115 is assumed to act as general base to deprotonate the reactive water; therefore, His115 was assigned as a HSD state in which Ne2 of H115 closely contacts one of the hydrogen atoms of 6

ACS Paragon Plus Environment

Page 6 of 38

Page 7 of 38

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

the reactive water. His134 was also assigned to be in the HSD state so that it can function as a proton shuttle for His115. The ES complex was solvated in a pre-equilibrated sphere of TIP3P waters with a 25 Å radius centered at the catalytic calcium atom. Any water molecule found within 2.8 Å of a heavy atom of the protein, substrate and specific water molecules (i.e., the reactive water and water molecules stabilizing for structural and catalytic calcium) was removed. The solvent was relaxed for 20 ps of molecular dynamics simulations with all protein residues, substrate, and the specific water molecules fixed. A pre-equilibrated TIP3 box was again added to MD-relaxed solvated system similarly with the above. In this time, any water molecule found within 2.8 Å of a heavy atom of the protein, substrate and specific water molecules and MD-relaxed solvent molecules in previous step were removed. The second 20 ps MD relaxation for newly added solvent molecules was performed with all protein residues, substrate, the specific water molecules, and MD-relaxed solvent molecules fixed. This process was repeated until the number of added water within 18 Å of the reactive center is unchanged. 23–24 O2 O1Hw2 C4 C3

PON1 Ow

O1 Hw1

C2

C3

C1 Hw2

R

C2

C1

Ow1Hw1 C4 O2

R

R= H: γ-butyrolactone = CH3: γ-valerolactone

γ-hydroxy acids

Scheme 1. The hydrolysis of 5-membered-ring lactones catalyzed by PON1 Next, the generalized solvent boundary potential (GSBP)25 was used for the molecular dynamics simulations. The system was partitioned into a 20-Å spherical inner region centered at the catalytic calcium and the rest of the system is outer region. In the inner region, the fully mobile region within 16 Å from the center was solved with molecular dynamics (MD) and the buffer region (i.e., 16-20 Å from the center) was solved with Langevin dynamics. Protein 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

atoms in the buffer region were harmonically constrained with force constants determined from its crystallographic B-factors. All water molecules in the inner region were restrained to keep them inside the inner region during the MD simulation. The static field of the outer region atoms is evaluated by GSBP with described parameters24 (i.e., cubic grid of 1113 centered on the Ca, coarse cubic grid of 1.2 Å spacing, fine grid of 0.4 Å spacing, protein dielectric constant of ߝp=1, water dielectric constant ߝw=80, and no salt was included). The reaction field matrix M is also evaluated using first 20th-order spherical harmonics with 400 basis functions.14, 24 The solvated ES complex in the 20-Å mobile region was minimized with the short steepest descent (SD)26 and adopted basis Newton Raphson (ABNR) methods27 to remove nonphysical interactions (i.e., residues of protein, substrate, and solvent). The minimized ES complex was gradually heated to 300 K and equilibrated at 300 K for 100 ps for each process to relax the solvated ES structure. The equilibrated ES complex was used as a starting structure for QM/MM calculations. 2.2.1.2. Verification of the enzyme-substrate complex system The simplest 5-membered-ring lactone for the lactonase reaction, GBL, was used to check the validity of the ES complex system. The important interactions during the molecular dynamics (i.e., heating and equilibration) of ES complex systems 1 and 2 were firstly used to confirm the roles of the crystallographic water (WAT359) and added water (WAT363) in each ES complex system. For better confidence, the equilibrated ES complexes were further minimized with QM/MM calculations. For the QM/MM calculations, the equilibrated enzyme-substrate complex was partitioned into a QM subsystem and a MM subsystem. The QM subsystem consisted of the substrate (GBL or GVL), the side chains of two putative catalytic residues (His115 and His134), the catalytic calcium ion (Ca357) and its ligands (i.e., the side chain of Glu53, Asn168, Asn224, Asp269, Asn270, and WAT359) for ES complex system 1. The QM subsystem also contained WAT363 in ES complex system 2. Link atoms 8

ACS Paragon Plus Environment

Page 8 of 38

Page 9 of 38

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

were added between the Cᆁ and Cࢼ atoms of each residue. QM/MM minimizations of the two ES complex systems at the Hartree-Fock level with the 3-21G(d,p) and 6-31G(d,p) basis sets15 for the QM subsystem was performed by CHARMM interfaced with GAMESS (US version) program.18, 28 To save time for QM/MM calculations, protein atoms beyond 20 Å away from the catalytic calcium ions were fixed. The criterion used to determine completion of the geometry optimization is that the root-mean-square (RMS) force is smaller than 0.01 (kcal.mol-1.Å-1). 2.2.1.3. Potential Energy Surface (PES) scan The QM/MM PES scan was used to generate a potential energy surface along the reaction path of the lactonase reaction. To verify the role of His134 in the lactonase reaction, we used two partitioned schemes in the PES scan using GBL as the substrate. The first scheme utilized the large QM subsystem used for ES complex systems 1 and 2 mentioned in section 2.2.1.2. The second scheme used the smaller QM subsystem that did not contain the side chain of His134. The ES complex optimized by the HF/3-21G(d,p) level was used as a starting structure. In the PES scan, the lactonase reaction was divided into two steps (Scheme 2). The first step began from the ES complex to a tetrahedral intermediate (TI) and the second step moved from the TI to the product (EP). In the first step, the PES was calculated as a function of two coordinates. The first coordinate describes the deprotonation of the reactive water by His115 was defined as the difference between two inter-atomic distances: one for the Ow-Hw1 bond to be cleaved, and another for the Hw1-Ne2(H115) bond to be formed, ξstep1=dOw-Hw1-dHw1-Ne2(H115). The second coordinate was also defined as the difference between two inter-atomic distances: one for the C4-O1 bond of the lactone to be cleaved and another for the Ow-C4 bond to be formed, ξstep2=dC4-O1-dOw-C4. In the second step of the lactonase reaction, the PES was also calculated as a function of two coordinates. The first reaction coordinate was defined as 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 38

ξstep2=dC4-O1-dOw-C4. The second coordinate, which describes the proton transfer from protonated His115 to the hydroxyl oxygen of TI, was defined as the difference between two inter-atomic distances: one for the Ne2(H115)-Hw1 bond to be cleaved and another for the O1Hw1 bond to be formed, ξstep3=dNe2(H115)-Hw1-dO1-Hw1. The label of the atoms in PES scan is indicated in Scheme 1. (a)

(b) Ca2+

2+

Ca

dC4-Ow

R

Hw2

C4

C2

Ow

C1 H

O2

O2

C3

O1

C3

dHw1-Ne2

dC4-O1

R

His115

H

N H

C4

Ow

dC4-O1

C2 C1

Ne2

ζstep1=dOw-Hw1-dHw1-Ne2

Hw2

dOw-Hw1

Hw1

dC4-Ow

O1

dO1-Hw1

ζstep2=dO1-C4-dC4-Ow

ζstep2=dO1-C4-dC4-Ow

dHw1-Ne2

Hw1 Ne2

His115

N H

ζstep3=dNe2-Hw1-dHw1-O1

Scheme 2. Illustration of reaction coordinates chosen for the hydrolysis of lactones during PES scan. (a): from enzyme-substrate complex (ES) to tetrahedral intermediate (TI); (b): from TI to enzyme-product complex (EP) The PES calculation was firstly performed at the HF/3-21G(d,p) level for the QM subsystem. The reaction coordinates (i.e., ξstep1, ξstep2 in the first step and ξstep2, ξstep3 in the second step) were increased in steps of 0.1 Å for each reaction coordinate with harmonic restraints of 5000 kcalmol-1Å-2. The criterion used to determine completion of the geometry optimization was that the RMS force is smaller than 0.01 (kcal.mol-1.Å-1).29 In addition, the reaction coordinates were increased in steps of 0.025 Å to in regions around transition states to obtain the better coordinates for transition states. Reaction minima (ES, TI, and EP) were further optimized without any restraint at the HF/3-21G(d,p) level to obtain a better geometry. Each reaction coordinate around transition states was further optimized at HF/631G(d,p) level while maintaining the restraint determined in the HF/3-21G(d,p) level. On the 10

ACS Paragon Plus Environment

Page 11 of 38

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

while, the reaction minima (ES, TI, and EP) were further optimized without any restraint at HF/6-31G(d,p) level. The criterion used for geometry optimization is the identical to that used for the calculation at the HF/3-21G(d,p) level. The resulting geometries (ES, TS1, TI, TS2, and EP) obtained from HF/3-21G(d,p) and HF/6-31G(d,p) were used for further single point energy calculation with higher QM methods (i.e. MP2 with different basis sets) to calculate their energies more accurately. 2.2.2. Experimental Methods 2.2.2.1. Cloning, expression, and purification of G3C9 PON1 The G3C9 PON1 gene, which is the best clone from third round evolution of PON1 for Escherichia coli (E. coli) expression without a fusion protein (i.e., thioredoxin),30 was chosen for the kinetic assay in this study. The G3C9 PON1 gene from Genscript was cloned into pET22b (+) (Novagen) at two double restrictive sites NdeI/EcoRI. The G3C9 PON1pET22b(+) gene was transformed into E. coli Origami B (DE3). The positive transformants were inoculated to LB medium with 1 mM CaCl2 and two selective antibiotics (i.e., Kanamycin 15 µg/ml and Ampicilin 100 µg/ml) and grown at 25°C, 150 rpm for 20 h and induced by 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG). The cells were harvested by centrifugation and disrupted by a Novagen bug buster (Damstadt, Germany). The G3C9 PON1 in the cleared lysate was further purified by a Qiagen Ni-NTA column purification kit (Venlo, Netherlands). The Bradford micro assay using Sigma Aldrich reagent was used to determine the purified enzyme concentration. 2.2.2.2. Kinetics of G3C9 PON1 The kinetic assay for lactonase of two lactones by G3C9 PON1 (i.e., 2 µg/ml for GBL and 0.4 µg/ml for GVL) were performed in 90 mM Tris-HCl buffer with 4.5 mM CaCl2, pH 7.0 with increasing concentration of lactones (i.e., from 2 mM to 16 mM for GBL and from 0.5 mM to 4 mM for GVL) at 30°C, 300 rpm for 20 minutes. The residual lactones in 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 38

reaction samples were extracted and analyzed by Gas Chromatography (GC).6 The residual lactones in the reaction samples were extracted immediately with equal amount of ethyl acetate by 1 minute shaking and brief centrifugation. The lactones in the ethyl acetate were analyzed by Gas Chromatography Mass Spectroscopy (GC-MS) (i.e., Agilent 6890 GC with 5793 N Mass spectrum detector and with DB-5MS column). Lineweaver-Burk plots were plotted to determine Km and Vmax of G3C9 PON1.31 3.

Results and Discussion

3.1. Kinetics of G3C9 PON1 The kinetic parameters for the hydrolysis of two lactones (i.e., GBL and GVL) catalyzed by G3C9 PON1 at 30°C and pH=7.0 is seen in Table 1. The turnover number (kcat) for the hydrolysis of GBL is approximately 4.0-fold higher than that for the hydrolysis of GVL (i.e., 147.209 and 36.621 s-1, respectively). The turnover number for hydrolysis of GBL in this study is comparable with that catalyzed by G2E6, which is the second round evolution for PON1 at pH=8.0 (i.e., 147.209 and 111 s-1, respectively).25 [30] Table 1. Kinetic parameters and Gibbs free energy for lactonase of some lactones catalyzed by G3C9 PON1 wild type at pH 7.0 substrate

structure

GBL

O

GVL

O

O

O

H

kcat

Km

kcat/Km (mM-1.s-1)

+

∆G +

(s-1)

(mM)

147.209

36.800

4.00

14.73

36.621

1.316

27.82

15.57

(kcal/mol)a

CH3

a

The Gibbs free energy calculated based on the transition state theory.32–33 The Gibbs free energy of activation for the hydrolysis of GBL and GVL calculated based

on transition state theory32–33 are 14.73 and 15.57 kcal/mol, respectively (Table 1). The free energy of activation for the hydrolysis of GVL is larger than that for the hydrolysis of the 12

ACS Paragon Plus Environment

Page 13 of 38

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

smaller lactone, GBL (0.8 kcal/mol). This free energy of activation will be compared with that obtained from QM/MM calculations. 3.2. Identification of possible ES complex system To determine whether the water resolved in the X-ray structure acts as a reactant for the lactonase reaction or only functions as a stabilizing ligand for the catalytic calcium ion, two enzyme-substrate (ES) complex systems were set up and used for QM/MM calculation. The substrate GBL was used for this work. 3.2.1. Enzyme-substrate complex system 1 The ES complex system 1 consists of the enzyme with two substrates (i.e., GBL and WAT359). The ES complex for system 1 is seen in Figure S1 (a) of the Supporting Information. The overall structure of the ES complex was stable during the heating and equilibration processes using classical molecular dynamics (MD) simulations (i.e., the rootmean-square-deviation (RMSD) of overall structure and the system temperature are approximately 0.9 Å and 300 K, respectively in Figure S2 of the Supporting Information). The structure of ES complex system 1 after MD simulations is seen in Figure S1 (b) of the Supporting Information. Some important interactions were carefully checked to ensure stability. The distance of the GBL carboxyl carbon and the oxygen of WAT359 is approximately 4.8 Å for the entire MD simulations (Figure S3). The distance between the hydrogen atom of WAT359 and Ne2 of His115 is approximately 3.5 Å (Figure S4 of the Supporting Information). From this MD simulation, it is observed that it would be impossible for WA359 to attack the lactone carboxyl carbon. This crystallographic water seems to be used for stabilizing the catalytic calcium ion. For further verification of this hypothesis, the ES complex obtained from the MD simulation was further optimized using QM/MM simulation. The QM region contains the catalytic calcium ion, 2 substrates (GBL and WAT359), and the side chains of the ligands coordinating catalytic calcium (i.e., Glu 53, Asn 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

168, Asn224, Asp269, and Asn270). The QM/MM optimized structure can be seen in Figure 1. The structures obtained from QM/MM simulation at Hartree-Fock method with different basis sets (i.e., HF/3-21G(d,p) and HF/6-31G(d,p)) were similar. Some important structural parameters of the ES complex can be seen in Table S1 of the Supporting Information. The distances from the catalytic calcium ion to its stabilizing ligands obtained from the QM/MM simulations with HF/3-21G(d,p) and HF/6-31G(d,p) levels were highly consistent with those obtained from the crystal structure. The distance between the oxygen of WAT359 and the GBL carboxyl carbon was relatively far (i.e., 5.15 Å and 5.21 Å at HF/3-21G(d,p) and HF/631G(d,p) levels, respectively). The distance between the oxygen of WAT359 and the GBL carboxyl carbon obtained from QM/MM simulations was very similar to the distance of WAT359 and the carboxyl carbon of the inhibitor OCH in crystal structure (i.e., 5.05 Å). The result of the MD and QM/MM simulations demonstrated that WAT359 seems to act as a stabilizing ligand for the catalytic calcium ion rather than to attack the carboxyl carbon of GBL. Hence, another water molecule, which would function as a reactant, is required for the lactonase reaction of GBL. (b)

(a)

Figure 1. The QM/MM optimized structure of PON1, GBL, and the crystallographic water. (a): HF/3-21 (d,p) level; (b): HF/6-31G(d,p) level. GBL and the crystallographic water (licorice); His115 and His134 (CPK); catalytic calcium (CPK, pink); the ligands coordinating catalytic calcium: Glu53, Asn168, Asn224, Asp269, Asn270 (lines) 14

ACS Paragon Plus Environment

Page 14 of 38

Page 15 of 38

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

3.2.2. Enzyme-substrate (ES) complex system 2 The construction of ES complex system 2 was mentioned in section 2.2.1.1. The distance between the oxygen of the added water (WAT363) and the carboxyl carbon of GBL was initially set at approximately 3 Å. The ES complex system 2 can be seen in Figure S5 (a) of the Supporting Information. The overall structure of ES complex system 2 is stable during the gradual heating and equilibration steps of the MD simulations (i.e., with the RMSD and the system temperature are approximately 0.9 Å and 300 K, respectively in Figure S6 of the Supporting Information). The structure of ES complex system 2 after the MD simulations can be seen in Figure S5 (b) of the Supporting Information. During the MD simulations without any restraint, the added water found a stable position around the lactone (i.e., the distance between the oxygen of WAT363 and the carboxyl carbon of GBL was approximately 3 Å). One of the hydrogen atoms of WAT363 interacted closely with Ne2 of His115 (i.e., distance between this hydrogen and Ne2 of His115 is approximately 2 Å). ES complex system 2 obtained from the MD simulations was further optimized with QM/MM simulations. The QM region includes the catalytic calcium ion, GBL, WAT363, and the side chains of the ligands coordinating the catalytic calcium (i.e., Glu 53, Asn 168, Asn224, Asp269, Asn270, and WAT359). The QM/MM HF/3-21G(d,p) optimized structure can be seen in Figure 2. The important structural parameters for ES complex system 2 can be seen in Table S2 of the Supporting Information. The interactions between the catalytic calcium ion and its stabilizing ligands (i.e., Glu53, Asn168, Asn224, Asp269, Asn270, and WAT359) are consistent between the QM/MM optimized structure and the crystal structure. WAT359 has played the important role of stabilizing the catalytic calcium ion. WAT363 closely interacted with the lactone (i.e., the distance between oxygen of WAT363 and the carboxyl carbon of GBL is 2.76 Å). The close proximity of these species suggests that the added water seems to react with GBL with that distance as a reactant. In addition, the added water also had close contact with His115 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 38

(i.e., the distance between the hydrogen of WAT363 and Ne2 of His115 is 1.93 Å). It appears as though His115 has ability to deprotonate the added water to produce a hydroxide ion, which may then attack the substrate. The added water is a reasonable second reactant for lactonase of GBL. The PES scan using QM/MM methods was performed to determine the reaction geometry and relative energy along the reaction coordinate in ES complex system 2. (a)

(b)

Added water

Added water

Figure 2. The QM/MM optimized structures of PON1, GBL, and the added water. (a): HF/321G(d,p) level; (b): HF/6-31G(d,p) level. GBL and the added water (licorice); the crystallographic water (CPK, orange); His115 and His134 (CPK), catalytic calcium (CPK, pink); the ligands coordinating catalytic calcium: Glu53, Asn168, Asn224, Asp269, Asn270 (lines). 3.3. Potential energy surface (PES) scan The potential energy surface (PES) scan was first performed for lactonase of the simplest 5-membered-ring lactone, GBL. As mentioned in section 2.2.1.3, a small QM subsystem and a large QM subsystem were constructed to verify the role of His134. The calculated potential energy surface along the reaction path for lactonase with GBL as substrate can be seen in Figures 3 and 4. The potential energy difference of some important stationary points can be seen in Table 2. The calculations showed that the reaction paths with both the small and the large QM subsystems are consistent with each other. The reaction proceeds from the reactant (ES) to the first transition state 1 (TS1) and then to the tetrahedral intermediate (TI) in the 16

ACS Paragon Plus Environment

Page 17 of 38

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

first step of the reaction (Figure a of Figures 3 and 4). The reaction continues from TI to a second transition state (TS2) and finally to the product (EP) in the second step of the reaction (Figure b of Figures 3 and 4). The PES scan showed that the first and second steps follow a concerted reaction mechanism. That is the deprotonation of the reactive water and nucleophilic attack to the carboxyl carbon of GBL occurs simultaneously in the first step, and the proton transfer from protonated His115 and the cleaving of the scissile bond between C4 and O1 of GBL also occurs simultaneously in second step (Figures 3 and 4). In TS1, the distance between the oxygen of the reactive water and carboxyl carbon of lactone was 1.54 Å, and the distances between the hydrogen to the oxygen of the reactive water and the hydrogen to Ne2(H115) were very similar (i.e., 1.21 and 1.26 Å, respectively, in Figures 3 and 4). In TS2, the scissile bond dC4-O1 was 1.71, and the distances between the hydrogen to the hydroxyl oxygen of TI and the hydrogen to Ne2 (H115) were equidistant (i.e., 1.23 Å). The potential energy of TS2 in both the small and the large QM subsystem was higher than that of TS1 at the HF/3-21G(d,p) level and MP2 level with all investigated basis sets. The formation of the product from the tetrahedral intermediate is the rate-limiting step for the overall lactonase reaction with GBL. The incorporation of His134 into the QM region, compared with that without His134 in the QM region, decreased the reaction energy barrier by 1.3 kcal/mol at HF/3-21G(d,p) level and 2.6–2.9 kcal/mol at the MP2 level (Table 2). The Mulliken charge of His115 residue of the putative His dyad obtained from MP2/6-31G(d) can be seen in Table S3 of the Supporting Information. The Mulliken charges of His115 in all stationary points (i.e., ES, TS1, TI, TS2 and EP) of the large QM subsystem were slightly smaller than those of the smaller QM subsystem. Residue His134 not only contributes to the stabilization of the transition state to reduce the activation energy of lactonase reaction, but also facilitates the movement of the proton by increasing the basicity of His115. Hereafter, the energy and structural profile for lactonase with GBL are based on the large QM 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 38

subsystem. (a)

(b) EP TI TS2

TS1

ES

TI

Figure 3. PES scan of the 3SRG-GBL complex with added water using HF/3-21G(d,p) with the small QM subsystem. (a): ES complex to TI complex; (b): TI complex to EP complex. The reaction coordinates ξstep1, ξstep2, and ξstep3 are indicated in Scheme 2. (a)

(b) EP TI

TS2

TS1

ES

TI

Figure 4. PES scan of the 3SRG-GBL complex with the added water using HF/3-21G(d,p) with the large QM subsystem. (a): From enzyme-substrate complex (ES) to Tetrahedral intermediate (TI); (b): Form TI to enzyme-product complex (EP). The reaction coordinates ξstep1, ξstep2, and ξstep3 are indicated in Scheme 2.

18

ACS Paragon Plus Environment

Page 19 of 38

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 2. The QM/MM potential energy difference for the lactonase reaction of GBL with various QM theories potential energy difference (kcal/mol) HF/3HF/6MP2/3MP2/6MP2/621G(d,p) 31G(d,p) 21G(d,p) 31G(d) 31G(d,p) stationary points at Small QM subsystem HF/3-21G(d,p) ES 0.00 0.00 0.00 0.00 level TS1 15.76 10.77 21.14 22.32 TI 9.52 7.11 19.49 21.77 TS2 17.24 15.74 25.39 25.80 EP -5.37 -2.31 0.79 1.02 Large QM subsystem ES 0.00 0.00 0.00 0.00 TS1 15.03 9.37 19.28 20.46 TI 8.87 7.00 17.51 20.03 TS2 15.98 13.11 22.51 22.94 EP -4.93 -3.61 -1.68 -1.624 stationary points at Large QM subsystem HF/6-31G(d,p) ES 0.0 0.00 0.00 0.00 level TS1 36.22 7.57 20.24 21.30 TI 30.68 7.81 18.47 21.17 TS2 37.85 11.45 24.29 25.65 EP -0.64 -2.8 0.59 0.64 The calculated potential energy barriers obtained from the large QM subsystem at HF/321G(d,p) level (i.e., 16.0 kcal/mol) and MP2/3-21G(d,p) level (i.e., 13.1 kcal/mol) are highly consistent with the activation energy of 14.73 kcal/mol estimated from the experimental kcat value of the G3C9 PON1-catalyzed lactonase reaction with GBL at pH=7.0 using simple transition state theory (Table 1). The calculated potential energy barriers obtained from MP2/6-31G(d) (i.e., 22.5 kcal/mol) and MP2/6-31G(d,p) (i.e., 22.9 kcal/mol) are in reasonable agreement with experimental reaction energy (i.e., 14.73 kcal/mol). The structural parameters of some stationary points along the reaction path can be seen in Table 3 and Figure 5. The transition state 2 is stabilized by Glu53 and Asn168 through hydrogen bonds, and the catalytic calcium ion is an oxy-anion hole. The distance between OE2 of Glu53 and 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 38

the hydrogen of the carboxyl group of TS2 is 1.71 Å, and the distance between HD21 of Asn168 and O2 of TS2 is 1.79 Å. These interactions might play an important role in lowering the activation energy compared to that of un-catalyzed reactions. Table 3. Selected geometries for stationary points along the reaction path of the PON1– catalyzed lactonase of GBL at the HF/3-21G(d,p) and HF/6-31G(d,p) levels. interactionsa

bond distances (Å) HF/3-21G(d,p) level ES

a

TS1 TI

TS2 EP

HF/6-31G(d,p) level ES

TS1

TI

TS2 EP

Ow(WAT)…C4(GBL)

2.69 1.54 1.48 1.37 1.32

2.91 1.52 1.44 1.34 1.30

C4(GBL)…O1(GBL)

1.35 1.44 1.47 1.72 2.59

1.33 1.41 1.43 1.79 2.75

Ow(WAT)…Hw1(WAT)

0.95 1.21 1.68 3.18 2.81

0.95 1.22 1.87 3.25 2.83

Hw1(WAT)…NE2(H115) 1.94 1.26 1.02 1.23 1.97

2.02 1.24 1.02 1.09 2.23

O1(GBL)…Hw1(WAT)

3.27 2.71 2.41 1.23 0.95

3.37 2.60 2.43 1.44 0.95

Ca357…O2(GBL)

2.44 2.33 2.35 2.33 2.39

2.56 2.34 2.36 2.36 2.45

HD21(N168)…O2(GBL)

2.18 1.79 1.72 1.79 2.53

2.32 2.01 1.94 2.02 2.56

Hw2(WAT)…OE2(E53)

1.87 1.70 1.81 1.71 1.60

2.04 1.81 1.95 1.81 1.67

The name of GBL and water indicated in Scheme 1; Ca357 indicates the catalytic calcium.

WAT indicate the crystallographic water interacting with the catalytic calcium and added water.

20

ACS Paragon Plus Environment

Page 21 of 38

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

(a)

(b)

(c)

ES (d)

TS1

TI

(e)

TS2

EP

Figure 5. Snapshots of the QM/MM stationary points along the reaction path for the PON1catalyzed hydrolysis of GBL at the HF/3-21G(d,p) level: (a) Enzyme-substrate complex, (b) transition state 1, (c) tetrahedral intermediate, (d) transition state 2, and (e) enzyme-product complex. GBL, the added water, His115, and His134 (as dynamic bonds); catalytic calcium (as CPK, pink); the ligands coordinating the catalytic calcium: Glu53, Asn168, Asn224, Asp269, Asn270, and the crystallographic water (as line). The additional HF/6-31G(d,p) geometry optimizations for the stationary points (i.e., ES, TS1, TI, TS2, and EP) along the reaction path of GBL, which were obtained from PES scans at HF/3-21G(d,p) level, showed that the geometries of the reaction maxima and minima obtained at QM/MM HF/6-31G(d,p) level are very similar to those obtained at the HF/321G(d,p) level (Table 3 and Figure 6). In addition, the calculated potential energy barriers for the hydrolysis of GBL at the MP2 level based on QM/MM HF/6-31G(d,p) geometries are only slightly different from those based on QM/MM HF/3-21G(d,p) geometries. Specially, the differences in activation energy values based on HF/3-21G(d,p) and HF/6-31G(d,p) 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 38

geometries at MP2 level with all basis sets are 1.7–2.8 kcal/mol which are shown in Table 2 and Figure 7. (a)

(b)

(c)

ES (d)

TS1

TI

(e)

EP

TS2

Figure 6. Snapshots of the QM/MM stationary points along the reaction path for the PON1catalyzed hydrolysis of GBL at the HF/6-31G(d,p) level: (a) Enzyme-substrate complex, (b) transition state 1, (c) tetrahedral intermediate, (d) transition state 2, and (e) enzyme-product complex. GBL, the added water, His115, and His134 (as dynamic bonds); catalytic calcium (as CPK, pink); the ligands coordinating the catalytic calcium: Glu53, Asn168, Asn224, Asp269, Asn270, and the crystallographic water (as line).

22

ACS Paragon Plus Environment

Page 23 of 38

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

(a)

(b)

TS1 TS2

Figure 7. Energy profile obtained from further optimization at the HF/6-31G(d,p) level around (a) the transition state1 and (b) the transition state 2 of the 3SRG-GBL complex at the HF/3-21G(d,p) level. The reaction coordinates ξstep1, ξstep2, and ξstep3 are indicated in Scheme 2. The PES scans were also done for the lactonase reaction of GVL, which contains one additional methyl group (Scheme 1). The enzyme-substrate complex was prepared similarly to that of enzyme-substrate system 2 with GBL. The ES complex before and after the MD simulations can be seen in Figure S9 of the Supporting Information. The equilibrated structure of the ES complex system for GVL was equilibrated for an additional 100ps to obtain a better stationary structure (Figure S10 of the Supporting Information).

The overall

structure of the ES complex is stable during the second equilibration step (i.e., The RMSD and the system temperature are approximately 0.9 Å, and 300 K, respectively, in Figure S10 of the Supporting Information). The important interactions during the MD simulations can be seen in Figures S11 and S12. The QM/MM optimized structure at HF/3-21G(d,p) level can be seen in Figure 8. The important geometric parameters of the ES complex can be seen in Table S4 of the Supporting Information. The interactions between the catalytic calcium ion and its stabilizing ligands (i.e., Glu53, Asn168, Asn224, Asp269, Asn270, and crystallographic water) are consistent between the optimized structure and the crystal structure. The optimized enzyme-substrate complex with GVL at the HF/3-21G(d,p) level was used as starting 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

structure for QM/MM PES scans.

Added water

Figure 8. The QM/MM HF/3-21G(d,p) optimized structure of PON, GVL, and the added water. GVL and the added water (licorice); crystallographic water (CPK, orange); His115 and His134 (CPK), catalytic calcium (CPK, pink); the ligands coordinating the catalytic calcium: Glu53, Asn168, Asn224, Asp269, and Asn270 (lines). The PON1-catalyzed lactonase reaction for GVL proceeded similarly to that for GBL (Figure 9). The potential energy difference of some important stationary points can be seen in Table 4. The calculated potential energy barriers obtained from the large QM subsystem at MP2/3-21G(d,p) level (i.e., 16.2 kcal/mol) are highly consistent with the activation energy of 15.57 kcal/mol estimated from the experimental kcat value of G3C9 PON1-catalyzed lactonase with GVL at pH=7.0 using simple transition state theory (Table 1). The calculated potential energy barriers obtained from HF/3-21G(d,p), MP2/6-31G(d), and MP2/6-31G(d,p) (i.e., 20.5, 25.0, and 25.6 kcal/mol, respectively) were also in a reasonable agreement with the experimental reaction energy (i.e., 15.57 kcal/mol). The geometry of some stationary points along the reaction path can be seen in Table 5 and Figure 10. The transition state 2 is also stabilized by Glu53 and Asn168 through hydrogen bonds and the catalytic calcium ion as an oxy-anion hole. The distance between OE2 of Glu53 and the hydrogen of the carboxyl group of TS2 is 1.67 Å, and the distance between HD21 of Asn168 and O2 of the TS2 is 1.88 Å. The effect of the catalytic calcium, Glu53, and Asn168 on GVL is similar to that on GBL. 24

ACS Paragon Plus Environment

Page 24 of 38

Page 25 of 38

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

(a)

(b) EP TI

TS2

TS1

ES

TI

Figure 9. PES scan of the 3SRG-GVL complex with the added water at HF/3-21G(d,p) with the large QM subsystem. (a): From enzyme-substrate complex (ES) to Tetrahedral intermediate (TI); (b): Form TI to enzyme-product complex (EP). The reaction coordinates ξstep1, ξstep2, and ξstep3 are indicated in Scheme 2. (a)

(b)

(c)

TS1

ES (d)

TI

(e)

TS2

EP

Figure 10. Snapshots of the QM/MM stationary points along the reaction path for the PON1catalyzed hydrolysis of GVL at the HF/3-21G(d,p) level: (a) Enzyme-substrate complex, (b)

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

Page 26 of 38

transition state 1, (c) tetrahedral intermediate, (d) transition state 2, and (e) enzyme-product complex. GVL, the added water, His115, and His134 (as dynamic bonds); catalytic calcium (as CPK, pink); the ligands coordinating the catalytic calcium: Glu53, Asn168, Asn224, Asp269, Asn270, and the crystallographic water (as line). Table 4. The QM/MM potential energy difference for the lactonase reaction of GVL with the large QM subsystem using various QM theories. potential energy difference (kcal/mol)

stationary points

ES

at TS1

HF/3-

HF/6-

MP2/3-

MP2/6-

MP2/6-

21G(d,p)

31G(d,p)

21G(d,p)

31G(d)

31G(d,p)

0.00

-

0.00

0.00

0.00

16.93

-

9.86

23.67

23.34

7.83

-

6.08

17.59

20.05

HF/3-

TI

21G(d,p)

TS2

20.45

-

16.20

25.01

25.61

EP

-4.17

-

-1.48

0.50

0.80

ES

-

0.00

0.00

0.00

0.00

-

38.41

9.06

21.34

22.75

stationary points

at TS1

HF/6-

TI

-

30.21

6.57

17.13

19.78

31G(d,p)

TS2

-

41.05

14.32

25.37

26.56

EP

-

-2.11

-1.62

-0.87

-0.78

26

ACS Paragon Plus Environment

Page 27 of 38

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 5. Selected geometries for stationary points along the reaction path of the PON1 – catalyzed lactonase of GVL at the HF/3-21G(d,p) and HF/6-31G(d,p) levels interactionsa

bond distances (Å) HF/3-21G(d,p) level ES

TS1 TI

TS2 EP

HF/6-31G(d,p) level ES

TS1 TI

TS2 EP

Ow(WAT)…C4(GVL)

3.15 1.54 1.47 1.36 1.31

3.31 1.51 1.43 1.34 1.29

C4(GVL)…O1(GVL)

1.35 1.44 1.47 1.71 2.58

1.32 1.41 1.42 1.74 2.86

Ow(WAT)…Hw1(WAT)

0.95 1.20 1.62 3.12 3.26

0.95 1.14 1.80 3.24 3.32

Hw1(WAT)…NE2(H115) 1.96 1.25 1.03 1.21 2.14

2.02 1.34 1.02 1.11 2.35

O1(GVL)…Hw1(WAT)

4.18 2.47 2.52 1.26 0.95

4.26 2.64 2.76 1.41 0.95

Ca357…O2(GVL)

2.46 2.33 2.34 2.32 2.43

2.57 2.35 2.34 2.33 2.45

HD21(N168)…O2(GVL) 2.12 1.79 1.75 1.88 2.21

2.27 1.97 1.96 2.14 2.38

Hw2(WAT)…OE2(E53)

2.09 1.72 1.93 1.79 1.69

a

1.89 1.69 1.78 1.67 1.61

The name of GVL and water indicated in Scheme 1; Ca357 indicates the catalytic calcium.

WAT indicates the crystallographic water interacting with the catalytic calcium and added water. The structures of stationary points (i.e., ES, TS1, TI, TS2, and EP) along the reaction coordinate the hydrolysis of GVL obtained at QM/MM HF/6-31G(d,p) level are similar to those obtained at the HF/3-21G(d,p) level (Table 5 and Figure 11). The calculated potential energy barriers for the hydrolysis of GVL, which are based on geometries calculated QM/MM HF/6-31G(d,p), are also slightly different from those based on geometries calculated at QM/MM HF/3-21G(d,p). In particular, the differences in activation energies based on the HF/3-21G(d,p) and HF/6-31G(d,p) geometries at the MP2 level with all basis sets are 0.4–1.9 kcal/mol (Table 4 and Figure 12).

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

(a)

(b)

(c)

TS1

ES (d)

Page 28 of 38

TI

(e)

EP

TS2

Figure 11. Snapshots of the QM/MM stationary points along the reaction path for the PON1catalyzed hydrolysis of GVL at the HF/6-31G(d,p) level: (a) Enzyme-substrate complex, (b) transition state 1, (c) tetrahedral intermediate, (d) transition state 2, and (e) enzyme-product complex. GVL, the added water, His115, and His134 (as dynamic bonds); catalytic calcium (as CPK, pink); the ligands coordinating the catalytic calcium: Glu53, Asn168, Asn224, Asp269, Asn270, and the crystallographic water (as line). (a)

(b)

TS1 TS2

Figure 12. Energy profile obtained from further optimization at HF/6-31G(d,p) around (a) 28

ACS Paragon Plus Environment

Page 29 of 38

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

transition state 1 and (b) transition state 2 of HF/3-21G(d,p) PES scanned 3SRG-GVL complex. The reaction coordinates ξstep1, ξstep2, and ξstep3 are indicated in Scheme 2. The activation energy for the lactonase of GVL is larger than that of GBL in all investigated QM/MM methods based on both HF/3-21G(d,p) and HF/6-31G(d,p) geometries (Table 6). The difference in activation energy for the PON1-catalyzed lactonase reactions of GVL and GBL obtained from MP2/6-31G(d) and MP2/6-31G(d,p), which are both based on HF/3-21G(d,p) geometries, are 2.5 and 2.7 kcal/mol, respectively. On the other hand, the difference in activation energy for PON1-catalyzed lactonase reactions of GVL and GBL obtained from MP2/6-31G(d) and MP2/6-31G(d,p), which are both based on HF/6-31G(d,p) geometries, are 1.1 and 0.9 kcal/mol, respectively. These values agree well with the analogous energy difference obtained from experiment (i.e., 0.8 kcal/mol) (Table 6). The superimposition of the reaction states for lactonase of GBL and GVL in the ES complex and the TS2 complex can be seen in Figure 13. Two residues (Tyr71 and Ile74) interact closely with atom C1 of GBL in the ES complex (3.85 Å and 4.12 Å for Tyr71 and Ile74, respectively). GVL, which possesses one additional methyl group at C1, was pushed further from the reactive water and His115 by two residues Tyr71 and Ile74 in the ES complex. In the TS2 complex, two other residues (Leu69 and Val346) that closely interact with C1 of GBL (4.06 Å for both residues) pushed GVL further from His115. The steric hindrance of Tyr71, Ile74, Leu69 and Val346 for GVL might cause the increase in activation energy for the lactonase reaction of GVL compared to that of GBL.

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

Table 6.

Page 30 of 38

The activation energies for the lactonase of two lactones (GBL and GVL)

catalyzed by PON1 obtained from QM/MM calculation and experiment substrates

activation energy (kcal/mol) QM/MM simulation

experiment

HF/3-21G(d,p) geometries

HF/6-31G(d,p) geometries

MP2/

MP2/

MP2/

MP2/

3-21G

6-31G(d) 6-31G

3-21G

6-31G(d) 6-31G

(d,p)

MP2/

(d,p)

(d,p)

MP2/

(d,p)

GBL

13.11

22.51

22.94

11.45

24.29

25.65

14.73

GVL

16.20

25.01

25.61

14.32

25.37

26.56

15.57

3.1

2.5

2.9

1.1

0.9

0.8

+

2.7

a

∆∆E +

a

∆∆E + is difference activation energy of GVL for GBL; ∆∆E + = ∆E(+GVL) − ∆E(+GBL)

+

+

+

+

It should also be mentioned that we observed a sharp decrease in potential energy for both GBL and GVL around a point (at ξstep1=1.0 and ξstep2=0.0) of PES scans in the first half of the lactonase reaction. The structures at that point were unphysical and do not belong to our reaction path (Figure S13 of the Supporting Information). The mechanism for PON1-catalyzed lactonase was also proposed from this study (Scheme 3). In this reaction, His115 functions to deprotonate the reactive water to generate an attacking hydroxide ion in the first half of reaction to generate a TI complex. Protonated His115 transfers a proton to the oxygen of the hydroxyl in TI to form the product. The catalytic calcium ion functions as an oxy-anion to stabilize both the substrate and the putative reaction state. Glu53 functions to stabilize the hydrogen of the reactive water and the hydrogen atoms of putative states. Asn168 stabilizes the transition states and the tetrahedralintermediate for the lactonase reactions. 30

ACS Paragon Plus Environment

Page 31 of 38

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

Asn168 Asn168

O

O Glu53

O O

Glu53

O

Ca2+

O

Ca2+

NH2

H

His115

NH2

O

H

His115

δO

δ+

H

O O

N

N

H

WAT363

O

O

NH

NH N

N R R

His134 H

His134

H N H

N H

Transition state 1 (TS1)

ES complex

Asn168

O Glu53

O O

Ca2+ H

NH2 O

His115

O H

N+

O

NH N

R H

His134 N H

Tetrahedral intermediate (TI)

O Asn168

O Asn168

Glu53 O

Glu53

O O

O

Ca2+ Ca2+

NH2 O

H

NH2

H

O

His115

O

O His115

δ+ H

N

O H

R H

N

O NH R

NH

N H

N His134

His134 N H

N H

Transition state 2 (TS2)

EP complex

Scheme 3. Proposed mechanism for PON1-catalyzed lactonase of 5-membered-ring lactones

31

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

(a)

Page 32 of 38

(b)

Tyr71 Val346 Leu69 Ile74

Figure 13. Superimposition of HF/3-21G(d,p) QM/MM reaction states for lactonase of GBL and GVL at (a): ES complex and (b) TS2 complex. Enzyme-GBL system (yellow); EnzymeGVL system (blue); GBL, GVL, the reactive water, and His115 (dynamic bond); Leu69, Tyr71, Ile74, and Val346 (CPK) 4.

Conclusion In this study, we performed QM/MM simulations of the PON1-catalyzed lactonase

reaction with various levels of ab initio QM methods. We have identified reaction paths along the hydrolysis of 5-membered-ring lactones using PES scans at Hartree-Fock level with 321G(d,p) and 6-31G(d,p) basis sets. Second, the activation energy barriers obtained from the QM/MM study were in good agreements with the experiments. Interestingly, the low level methods (MP2/3-21G(d,p)) gave excellent agreements with the experiments. The activation energies obtained from MP2/3-21G(d,p) single point energy calculations based on HF/321G(d,p) and HF/6-31G(d,p) geometries for lactonase of GVL and GBL were 14.32–16.2 and 11.45–13.1 kcal/mol, respectively. These values are highly consistent with the experiment (i.e., 15.57 and 14.73kcal/mol for lactonase of GVL and GBL, respectively). On the other hand, the activation energies obtained from MP2/6-31G(d) and MP2/6-31G(d,p) single point energy calculations based on both QM/MM HF/3-21G(d,p) and HF/6-31G(d,p) geometries for lactonase of GBL and GVL were also in relatively good agreement with the experiments. Importantly, the difference in the QM/MM energy barrier for the lactonase reaction of GVL 32

ACS Paragon Plus Environment

Page 33 of 38

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

and GBL is the same tendency with experimental values. Specially, the difference in the QM/MM energy barrier for PON1-catalyzed lactonase reactions with (R)-γ-valerolactone and γ-butyrolactone obtained from MP2/6-31G(d) and MP2/6-31G(d,p) single point energy

calculations based on HF/6-31G(d,p) geometries were 1.1 and 0.9 kcal/mol, respectively. These values are highly consistent with the difference value obtained from experiment (i.e., 0.8 kcal/mol). Four residues (Leu69, Tyr71, Ile74 and Val346) might be as selective pocket for lactonase of 5-membered-ring lactones. In addition, QM/MM with different QM subsystems demonstrated the essential role of the second residue of the His dyad (i.e., His134). This residue stabilizes of the transition state of the reaction (i.e., His134 contributed approximately 3 kcal/mol to the activation energy) and on the proton shuttle for His115. Based on the QM/MM calculations, the detailed mechanism for lactonase was proposed. In this mechanism, the catalytic calcium ion and residue Asn168 function as an oxy-anion hole to stabilize the reaction states (i.e., ES, TS1, TI, TS2, and EP). Residue Glu53 stabilizes not only the catalytic calcium ion, but also the reaction states through hydrogen bonds with the reactive water and other reactant states (i.e., TS1, TI, TS2, and EP). Supporting information Selected geometries of ES complex system 1 (Table S1) and system 2 (Tables S3 and S4) for lactonase of GBL and GVL after QM/MM optimization; MP2/6-31G(d) Mulliken charges of residues His115 in the stationary points (Table S2); snapshots of the ES complex system 1 (Figure S1) and system 2 (Figures S5 and S9) for lactonase of GBL and GVL before and after heating and equilibration period; some important parameters (i.e., temperature, RMSD) of the ES complex system 1 (Figure S2) and system 2 (Figures S6 and S10) during heating and equilibration period; the distances of some important interactions (i.e., C4(GBL or GVL) with Ow and Hw with Ne2(His115)) of ES complex system 1 (Figures S3 and S4) and system 2 (Figures S7, S8, S11, and S12) during equilibration period; the structures at ξstep1=1.0 and 33

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

ξstep2=0.0 of the PES scan. This material is available free of charge via the Internet at http://pubs.acs.org. Acknowledgements We thank the Korea CCS Research & Development Center (KCRC) (2014M1A8A1049296), KETEP (20133030000300, 20143030091040), and National Research Foundation Grant funded by the Korean government (MEST) (2013R1A1A1A05009866) for supporting this research. References (1) Khersonsky, O.; Tawfik, D. S. The Histidine 115-Histidine 134 Dyad Mediates the Lactonase Activity of Mammalian Serum Paraoxonases. J. Biol. Chem. 2006, 281, 7649– 7656. (2) Harel, M.; Aharoni, A.; Gaidukov, L.; Brumshtein, B.; Khersonsky, O.; Meged, R.; Dvir, H.; Ravelli, R. B. G.; McCarthy, A.; Toker, L.; Silman, I.; Sussman, J. L.; Tawfik, D. S. Structure and Evolution of the Serum Paraoxonase Family of Detoxifying and Antiatherosclerotic Enzymes. Nat. Struct. Mol. Biol. 2004, 11, 412–419. (3) Khersonsky, O.; Tawfik, D. S. Structure-Reactivity Studies of Serum Paraoxonase PON1 Suggest that Its Native Activity Is Lactonase. Biochemistry 2005, 44, 6371–6382. (4) Fairchild, S. Z.; Peterson, M. W.; Hamza, A.; Zhan, C.-G.; Cerasoli, D. M.; Chang W. E. Computational Characterization of How the VX Nerve Agent Binds Human Serum Paraoxonase 1. J. Mol. Model. 2011, 17, 97–109. (5) Peterson, M. W.; Fairchild, S. Z.; Otto, T. C.; Mohtashemi, M.; Cerasoli, D. M.; Chang, W. E. VX Hydrolysis by Human Serum Paraoxonase 1: A Comparison of Experimental and Computational Results. Plos One 2011, 6, 1–7. (6) Teiber, J. F.; Draganov, D. I.; La Du, B. N. Lactonase and Lactonizing Activities of Human Serum Paraoxonase (PON1) and Rabbit Serum PON3. Biochem. Pharmacol. 34

ACS Paragon Plus Environment

Page 34 of 38

Page 35 of 38

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

2003, 66, 887–896. (7) Martin, C. H.; Wu, D.; Jones Prather, K. L. Integrated Bioprocessing for the pHDependent Production of 4-Valerolactone from Levulinate in Pseudomonas Putida KT2440. Appl. Environ. Microbiol. 2010, 76, 417–424. (8) Kuo, C.; La Du B. N. Calcium Binding by Human and Rabbit Serum Paraoxonases Structural Stability and Enzymatic Activity. Drug Metab. Dispos. 1998, 26, 653–660. (9) Ben-David, M.; Elias, M., Filippi, J.; Duñach, E.; Silman, I.; Sussman, J. L.; Tawfik, D. S. Catalytic Versatility and Backups in Enzyme Active Sites: The Case of Serum Paraoxonase. J. Mol. Biol. 2012, 418, 181–196. (10) Senn, H. M.; Thiel, W. QM/MM Methods for Biomolecular Systems. Angew. Chem. Int. Ed. 2009, 48, 1198–1229. (11) Hermann, J. C.; Ridder, L.; Mulholland, A. J.; Holtje, H.-D. Identification of Glu166 as the General Base in the Acylation Reaction of Class A β-Lactamases through QM/MM Modeling. J. Am. Chem. Soc. 2003, 125, 9590–9591. (12) Xu, D.; Guo, H.; Cui, Q. Antibiotic Deactivation by a Dizinc β-Lactamase: Mechanistic Insights from QM/MM and DFT Studies. J. Am. Chem. Soc. 2007, 129, 10814–10822. (13) Zhang, Y.; Kua, J.; McCammon, J. A. Role of the Catalytic Triad and Oxyanion Hole in Acetylcholinase Catalysis: An Ab Initio QM/MM Study. J. Am. Chem. Soc. 2002, 124, 10572–10577. (14) Hou, G.; Cui, Q. QM/MM Analysis Suggests that Alkaline Phosphate (AP) and Nucleotide Pyrophosphase/Phosphodiesterase Slightly Tighten the Transition State for Phosphate Diester Hydrolysis Relative to Solution: Implication for Catalytic Promiscuity in the AP Superfamily. J. Am. Chem. Soc. 2012, 134, 229–246. (15) McQuarrie, D. A. Quantum Chemistry; University Science Books: Sausalito, California, U.S.A., 2008. 35

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

(16) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; et al. Gaussian 03, Revision C.02; Gaussian, Inc.: Wallingford, CT, 2004. (17) Morris, G. M.; Huey, R.; Lindstrom, W.; Sanner, M. F.; Belew, R. K.; Goodsell, D. S.; Olson, A. J. Autodock4 and Autodocktools4: Automated Docking with Selective Receptor Flexibility. J. Comput. Chem. 2009, 30, 2785–2791. (18) Brooks, B. R.; Brooks, C. L.; Mackerell, A. D.; Nilsson, L.; Petrella, R. J.; Roux, B.; Won, Y.; Archontis, G.; Bartels, C.; Boresch, S.; et al. CHARMM: The Biomolecular Simulation Program. J. Comput. Chem. 2009, 30, 1545–1614. (19) Best, R. B.; Zhu, X.; Shim, J.; Lopes, P. E. M.; Mittal, J.; Feig, M.; Mackerrel, A. D. Jr. Optimization of the Additive CHARMM All-Atom Protein Force Field Targeting Improved Sampling of the Backbone φ, ψ and Side-Chain χ1 and χ2 Dihedral Angles. J. Chem. Theo. Comptut. 2012, 8, 3257–3273. (20) Vanommeslaeghe, K.; MacKerell, A. D. Jr. Automation of the CHARMM General Force Field (CGenFF) I: Bond Perception and Atom Typing. J. Chem. Inf. Model. 2012, 52, 3144–3154. (21) Jo, S.; Kim, T.; Iyer, V. G.; Im, W. CHARMM-GUI: A Web-Based Graphical User Interface for CHARMM. J. Comput. Chem. 2008, 29, 1859–1865. (22) Olsson, H. M. M.; Søndergaard, R. C.; Michal, R.; Jensen, H. J. PROPKA3: Consistent Treatment of Internal and Surface Residues in Empirical pKa Predictions. J. Chem. Theory Comput. 2011, 7, 525–537. (23) Xu, D.; Guo, H.; Cui, Q. Antibiotic Binding to Dizinc β-Lactamase L1 from Stenotrophonomas Maltophilia: SCC-DFTB/CHARMM and DFT Studies. J. Phys. Chem. A 2007, 111, 5630–5636. (24) Schaefer, P.; Riccardi, D.; Cui, Q. Reliable Treatment of Electrostatics in Combined 36

ACS Paragon Plus Environment

Page 36 of 38

Page 37 of 38

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

QM/MM Simulation of Macromolecules. J. Chem. Phys. 2005, 123, 014905. (25) Im, W.; Bernèche, S.; Roux, B. Generalized Solvent Boundary Potential for Computer Simulations. J. Chem. Phys. 2001, 114, 2924–2937. (26) Levitt, M.; Lifson, S. Refinement of Protein Conformation Using a Macromolecular Energy Minimization Procedure. J. Mol. Biol. 1969, 46, 269–279. (27) Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J.; Swaminathan, S.; Karplus, M. CHARMM: A Program for Macromolecular Energy, Minimization, and Dynamics Calculations. J. Comput. Chem. 1983, 4, 187–217. (28) Schmidt, M.W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N., Nguyen, K. A.; Su, S.; et al. General Atomic and Molecular Electronic Structure System. J. Comput. Chem. 1993, 14, 1347–1363. (29) Lodola, A.; Mor, M.; Hermann, J. C.; Tarzia, G.; Piomelli, D.; Mulholland, A. J. QM/MM Modeling of Oleamide Hydrolysis in Fatty Acid Amide Hydrolase (FAAH) Reveals a New Mechanism of Nucleophile Activation. Chem. Comm. 2005, 35, 4399– 4401. (30) Aharoni, A.; Gaidukov, L.; Yagur, S.; Toker, L.; Silman, I.; Tawfik, D. S. Directed Evolution of Mammalian Paraoxonases PON1 and PON3 for Bacterial Expression and Catalytic Specification. Proc. Natl. Acad. Sci.U.S.A. 2004, 101, 482–487. (31) Shuler, M. L.; Kargi, F. Bioprocessing Engineering: Basic Concept; Prentice Hall PTR: Englewood Cliffs, New Jersey, U.S.A., 1992. (32) Eyring, H. The Activated Complex in Chemical Reactions. J. Chem. Phys. 1935, 3, 107–115. (33) Garcia-Viloca, M.; Gao, J.; Karplus, M.; Truhlar, D. G. How Enzymes Work: Analysis by Modern Rate Theory and Computer Simulations. Science 2004, 303, 186–195.

37

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

Table of Contents (TOC) Image

38

ACS Paragon Plus Environment

Page 38 of 38