MM study of the catalytic mechanism of nicotinamidase.

Organic & Biomolecular Chemistry View Article Online

Published on 13 January 2014. Downloaded by Aston University on 18/01/2014 08:07:28.

PAPER

Cite this: DOI: 10.1039/c3ob42182a

View Journal

A QM/MM study of the catalytic mechanism of nicotinamidase† Xiang Shenga and Yongjun Liu*a,b Nicotinamidase (Pnc1) is a member of Zn-dependent amidohydrolases that hydrolyzes nicotinamide (NAM) to nicotinic acid (NA), which is a key step in the salvage pathway of NAD+ biosynthesis. In this paper, the catalytic mechanism of Pnc1 has been investigated by using a combined quantum-mechanical/molecular-mechanical (QM/MM) approach based on the recently obtained crystal structure of Pnc1. The reaction pathway, the detail of each elementary step, the energetics of the whole catalytic cycle, and the roles of key residues and Zn-binding site are illuminated. Our calculation results indicate that the catalytic water molecule comes from the bulk solvent, which is then deprotonated by residue D8. D8 functions as a proton transfer station between C167 and NAM, while the activated C167 serves as the nucleophile. The residue K122 only plays a role in stabilizing intermediates and transition states. The oxyanion hole formed by the amide backbone nitrogen atoms of A163 and C167 has the function to stabilize the hydroxyl anion of nicotinamide. The Zn-binding site rather than a single Zn2+ ion acts as a Lewis acid to influence the reaction. Two elementary steps, the activation of C167 in the deamination process and

Received 4th November 2013, Accepted 15th December 2013

the decomposition of catalytic water in the hydrolysis process, correspond to the large energy barriers of 25.7 and 28.1 kcal mol−1, respectively, meaning that both of them contribute a lot to the overall reaction

DOI: 10.1039/c3ob42182a

barrier. Our results may provide useful information for the design of novel and efficient Pnc1 inhibitors

www.rsc.org/obc

and related biocatalytic applications.

1.

Introduction

Nicotinamide adenine dinucleotide (NAD+) is a ubiquitous and essential coenzyme found in all living cells1,2 and plays an important role in cellular metabolism and energy production by accepting and donating electrons. NAD+ can be biosynthesized through two major pathways in most organisms: the de novo pathway from simple components, such as tryptophan and aspartic acid, and the salvage pathway, in which NAD+ is regenerated from its degradation products. In the salvage pathway of most organisms such as bacteria,3,4 yeasts,5 protozoa,6 plants7 a School of Chemistry and Chemical Engineering, Shandong University, Jinan, Shandong 250100, China b Northwest Institute of Plateau Biology, Chinese Academy of Sciences, Xining, Qinghai 810001, China. E-mail: [email protected]; Fax: +86 531 885 644 64; Tel: +86 531 883 655 76 † Electronic supplementary information (ESI) available: The values of d1, d2, d3 and d4 in the optimized structure of all species in DA-p1 and DA-p2 for the deamination process and HL-p1 for the hydrolysis process. Time dependences of RMSDs from 5 ns MD simulations for deamination and hydrolysis processes. The energy profile along the reaction coordinate in HL-p3. The distances between Zn2+ and the ligated atoms of Zn-coordinated molecules, the ESP charges of the Zn-binding site without NAM, and the ESP charges of Zn2+ and Zn-coordinated residues and water molecules of species in DA-p2 for the deamination process and HL-p1 for the hydrolysis process. See DOI: 10.1039/ c3ob42182a

This journal is © The Royal Society of Chemistry 2014

and many metazoans,8 the precursor nicotinamide (NAM) is firstly converted to nicotinic acid (NA) catalyzed by nicotinamidase (EC 3.5.1.19, Pnc1), and then adenylated to nicotinamide mononucleotide (NAMN), which is the convergent point with the de novo pathway. But in mammals, Pnc1 cannot be encoded and NAM is directly converted to NAMN using nicotinamide phosphoribosyltransferase.9 Pnc1 is a member of Zn-dependent amidohydrolases that hydrolyzes NAM to NA and ammonia, as shown in Scheme 1. It can also catalyze the conversion of the NAM analogue pyrazinamide (PAM) into pyrazinoic acid (PA), and has another name pyrazinamidase (PncA). As Pnc1 is an essential enzyme for the pathogens in mammals and is absent in mammal NAD+ salvage pathways, it is suggested to be a potential antibiotic target. In the current short-course treatment of the highly contagious tuberculosis recommended by the World Health Organization,10,11 Pnc1 serves as the drug target to hydrolyze the prodrug PAM to the active form PA. But the emergence of strains

Scheme 1

The catalytic reaction of nicotinamidase (Pnc1).

Org. Biomol. Chem.

View Article Online


Paper

which are resistant to PAM brings a great challenge to public health.12,13 Design of novel inhibitors with high efficiency is extremely urgent, and a more comprehensive understanding of enzymatic details of Pnc1 may also be of pharmaceutical value. In the past few decades, many crystal structures of Pnc1 from different species, such as Pyrococcus horikoshii,14 yeast,15,16 Acinetobacter baumanii,17 Streptococcus pneumonia,18 Mycobacterium tuberculosis,19 Leishmania infantum,20 Streptococcus mutans21 and human caspase-6,21 have been solved. Firstly, based on the structure from Pyrococcus horikoshii, Kim’s group revealed the binding site of the Zn2+ ion and confirmed that the Zn2+ ion can increase the activity of the enzyme.14 In 2009, Hunter’s group reported two crystal structures of Pnc1 in complex with NA and PA, respectively.17 And the metal ion Zn2+ shows an octahedral coordination with D54, H56, H89, two water molecules and NA or PA. Subsequently, several crystal structures at high resolution from Streptococcus pneumonia in unligand- and ligand-bound forms were obtained by Ealick’s group.18 These structures provided details of substrate binding and revealed the structure of the proposed thioester reaction intermediate. Recently, Cleland’s group obtained the first crystal structure of Pnc1 in complex with a NAM analogue in the active site from eucaryon.16 Consistent with the structures of Pnc1 from Acinetobacter baumanii and Streptococcus pneumonia, the substrate binds to the octahedralcoordinated Zn2+ ion through the nitrogen atom of the pyridine ring, and a putative catalytic triad D8, K122 and C167 (in yeast, PDB code: 3V8E) found in all other known Pnc1 is also observed. Kinetic studies and mutant experiments have been carried out to investigate the catalytic reaction of Pnc1,22–25 which was proposed to follow a ping-pong mechanism, as shown in Fig. 1. The whole catalytic cycle can be divided into two halfreactions, namely the deamination process and the hydrolysis process separated by a thioester intermediate. In the deamination process, the residue C167 firstly donates a proton to D8 to form a thiolate anion, which then attacks the carbonyl carbon of the substrate to form an intermediate. Subsequently, a NH3 molecule is released and a thioester intermediate is formed. In the hydrolysis process, a water molecule is proposed to attack on the carbonyl carbon of thioester to form another tetrahedral intermediate, which then collapses to release the final product NA. During the reaction, as a member of the catalytic triad, the residue K122 is proposed to have the assisting role to orient C167 and D8 and stabilize the species along the reaction pathway.18 Mutant studies also revealed that the Zn-binding site plays an important role in the catalytic reaction.19 Although a rough picture of the catalytic mechanism has been obtained, open questions still remain. For example, the detailed description of each elementary step is absent. The identity of catalytic water in the hydrolysis process is still a controversial issue. Some research groups hypothesized that the catalytic water comes from the Zn2+-coordinated water molecules, and the coordinated water molecule may exist in its natural form or deprotonated form, which attacks the thiol ester directly.14,18 Some other groups suggested that a Zncoordinated water molecule is firstly deprotonated to a

Org. Biomol. Chem.

Organic & Biomolecular Chemistry

Fig. 1

Proposed catalytic mechanism of Pnc1.

hydroxyl group, which then abstracts a proton from the bulk solvent water, and the latter serves as a nucleophile in the hydrolysis.26 However, the Lewis acid strength of the Zn2+ ion is greatly reduced by its high coordinate number. This decreases the possibility of a Zn-coordinated water molecule to be catalytic water, which is common in Zn-dependent enzymes with four-coordinate Zn2+ ions.17 Furthermore, more evidence supports the proposal that the nucleophilic water comes from the bulk solvent.16,17 Secondly, the Zn2+ ion and Zn-coordinated residues have been proved to be crucial for the catalytic reaction, but there is no detailed explanation about their roles. In addition, the roles of some key residues and energetics of the whole catalytic cycle are also not fully understood. It is generally known that explicit description of the enzymatic mechanism is significant for exploring the biochemical characterization of enzymes and developing novel drugs that have high efficiency, and some useful information cannot be acquired by experimental studies alone. Therefore, theoretical studies at the atomistic level are necessary. By using docking and MD simulations, Zhang’s group has revealed an optimal binding/unbinding pathway of ligand to PncA27 and Sethumadhavan’s group has studied the PA resistance mechanism in three mutants of PncA.28 However, it is somewhat surprising that there is still no mechanistic research on the Pnc1 catalytic reaction using theoretical approaches. Here, we present a theoretical study on the catalytic mechanism of Pnc1 by using a combined quantum mechanics and molecular mechanics (QM/MM) method.29–32 The basic idea of the QM/MM method is to divide the entire system into two portions, the QM region and the MM region. The QM region involving the formation and cleavage of chemical bonds is described quantum mechanically whereas the MM region that represents the surrounding protein is treated by a MM force field. In this methodology, both the chemical reactions occurred in active sites and the effect of the protein and the


View Article Online



Paper

Fig. 2 Two possible pathways in the deamination process of the Pnc1 catalytic reaction. In pathway DA-p1, the formation of the C–S bond is prior to the proton transfer from D8 to NAM, while in DA-p2 the proton transfer from D8 to NAM is prior to the formation of the C–S bond.

solvent environment can be considered. In recent years, this method has been extensively applied in the studies of the enzymatic reaction mechanism.33–38 Herein, the catalytic mechanism of Pnc1 was systematically studied by using the QM/MM method. In the deamination process, two possible pathways (as shown in Fig. 2) were considered. In the first pathway (DAp1), the formation of the C–S bond is prior to the proton transfer from D8 to NAM, whereas in the second pathway (DA-p2),

the proton transfer from D8 to NAM is prior to the formation of the C–S bond. In the hydrolysis process, to explore the source of the hydroxyl group of the final product, three possible pathways (HL-p1, HL-p2 and HL-p3, as shown in Fig. 3) were considered. In HL-p1 and HL-p2, both of the catalytic water molecules are deprotonated by residue D8. But, in HL-p1 the catalytic water comes from the bulk solvent whereas in HLp2 it is from the Zn-coordinated water. In HL-p3, the catalytic

Fig. 3 Three possible pathways in the hydrolysis process of the Pnc1 catalytic reaction. In HL-p1 the catalytic water molecule comes from the bulk solvent, which is deprotonated by residue D8. In HL-p2, the catalytic water molecule is one of the Zn-coordinated water molecules, which is also deprotonated by residue D8. In HL-p3, the catalytic water molecule is also a Zn-coordinated water molecule, but it is firstly deprotonated by residue D51 to generate a Zn-binding hydroxyl group.


Org. Biomol. Chem.

View Article Online

Paper


water is derived from a Zn-coordinated water molecule, which is firstly deprotonated by residue D51. In this paper, the binding structure, detailed energetic profiles of the overall reaction and the structures of all species along the reaction pathway were presented, and the roles of some key residues and the Zn-binding site were illuminated as well.


2. Computational methods 2.1.

QM/MM models

Since our present study aims to elucidate the reaction mechanism of the Pnc1 catalytic cycle, the calculations should begin with the Michaelis complex. However, it is generally difficult to experimentally determine the structure of the Michaelis complex, because it instantly converts to the enzyme–product complex under the regular experimental condition. Therefore, our initial computational model was constructed from a yeast Pnc1 intermediate structure (PDB code: 3V8E), which is actually a complex of an inhibitor nicotinaldehyde complexed with Pnc1. The crystal structure and active pocket are shown in Fig. 4. This choice is mainly due to the high similarities between the inhibitor and the natural substrate NAM. To recover the enzyme in wild-type without a substrate, part of the nicotinaldehyde bound to residue C167 was deleted manually. The substrate NAM molecule was primarily optimized with the Gaussian 09 package39 at the level of B3LYP/6-31G(d,p), and then was docked into the active pocket in place of the inhibitor using the Autodock4.0 program.40 When docking, the Gasteiger charges41 were set for all atoms. During the calculation, the protein was kept rigid while all the torsional bonds of the substrate were kept free. By using the Grid module, the grid scale was set as 40 Å × 40 Å × 40 Å with a spacing of 0.375 Å between the grid points. 100 Independent docking runs were performed. The docking results were clustered based on a root-mean-square deviation (RMSD) criterion of 2.0 Å. Finally, only one cluster was obtained, which means the 100 poses have very similar conformation and protein–ligand interaction

Fig. 4

energies. The typical conformation was chosen for the following MD simulations. The protonation states of ionizable residues were determined according to the experimental condition and the pKa values were predicted by the PROPKA 3.1 program.42 Two Zn-coordinated histidine residues (H53 and H94) were determined as singly protonated on δ site, and the Zn-coordinated aspartic acid residue (D51) was set to its deprotonated state. Two other ionizable residues D8 and K122 involved in the catalytic triad were modeled in their deprotonated and protonated states, respectively. The protonation states of all residues were checked carefully by the VMD program.43 Based on heavy-atom positions and standard bond parameters, hydrogen atoms of residues and crystal water molecules were added by using the HBUILD facility in the CHARMM package.44 Then, the obtained Michaelis complex model was solvated into a water sphere of 34 Å radius centered on the carbonyl carbon atom of substrate NAM, where the crystallographic water molecules in the crystal structure were kept in their original positions. The solvent was relaxed by a short MD simulation while all the other atoms were fixed. Finally, the system was neutralized by randomly adding 6 Na+ ions. The resulting neutral system consists of 19 048 atoms. To equilibrate the prepared system, a series of minimizations and a final 5 ns MD simulation were performed with a stochastic boundary condition, temperature at 298 K, and pressure at 1 atm. The time step is 1 fs. During the MD simulation, the system was divided into two regions: an inner reaction region (r < 31 Å) where the simulation was performed by Newton’s equations of motion, and an outer buffer region (31 Å < r < 34 Å) where the atoms were described by Langevin dynamics with friction and random force. The system can be kept at thermal equilibrium by using this hybrid method that couples the water molecules in the buffer region to a heat bath. The root-mean-squared deviations (RMSDs) of the protein in deamination and hydrolysis processes were derived, and are shown in Fig. S1,† which reveals that the backbone of the protein only changes slightly after 3 ns. As the dynamics trajectory was stable and the active pocket is almost unchanged,

The crystal structure and active pocket of Pnc1 (PDB code: 3V8E).

Org. Biomol. Chem.


View Article Online



a typical snapshot after MD simulations was chosen for the following QM/MM studies. The CHARMM22/CMAP all-atom force field45 for the protein and the TIP3P model46 for water molecules were employed. As it is difficult to properly describe the zinc coordination shell by a MM force field,47–49 the Zn2+ ion was modeled with Stote’s scheme50 and the Zn-coordinated molecules were fixed to retain the zinc coordination structure during the MM minimizations and dynamics simulations.51 The initial model for the hydrolysis process was derived from the reaction intermediate that the first product ammonia was released. For the HL-p1 pathway, the catalytic water from the bulk solvent takes the place of ammonia. Thus, the ammonia molecule was removed and four water molecules were added manually to the catalytic site. To avoid unreasonable movement of the system caused by the addition of water, a 2 ns MD simulation was firstly performed, where the Zncoordinated molecules and the surrounding residues of the four added water molecules were fixed. Then, a final 5 ns MD simulation was performed, in which only the Zn-coordinated molecules were fixed. The resulting snapshot was used for the following QM/MM calculations. For HL-p2, the initial model was also constructed from the intermediate that the ammonia molecule was removed, but a Zn-coordinated water molecule was moved manually to the reaction center. No further treatment was taken for HL-p3. 2.2.

QM/MM calculations

The QM/MM model was prepared from the snapshot of the classical MD trajectory. The entire system was partitioned into QM and MM subsystems. For the deamination process, the QM subsystem contains 77 atoms in the active site, including the side chains of D8, K122, E129, C167 and three Zn-coordinate residues (D51, H53, H94), two Zn-coordinated water molecules (W1 and W2), the substrate NAM and the Zn2+ ion. For the hydrolysis process, in HL-p1, the formed ammonia was removed and two water molecules (W3 and W4) were added to the QM subsystem, while in HL-p2 and HL-p3 only ammonia

Paper

was removed. The QM subsystems for the deamination process and HL-p1 for the hydrolysis process are illustrated in Fig. 5. To describe the zinc coordination shell, the strategy with the Stuttgart ECP/basis set52 (SDD) for the zinc atom and a 6-31G(d) basis set for all other atoms has been tested previously and employed widely.51,53–55 As some proton transfer processes are included in the catalytic reaction, in this paper, a higher level of B3LYP(SDD, 6-31G(d,p)) with polarization function on all atoms was used to gain more accurate results. All other residues and surrounding water molecules were set as an MM part and described by the CHARMM22 force field. The QM/MM boundaries were described by the link atom approach with a charge shift scheme.56,57 The prepared QM/MM system was firstly minimized and then employed to map out the optimal energy paths with the reaction coordinate. The MM atoms beyond a distance of 10 Å from the substrate NAM were kept frozen, whereas the remaining atoms, including all the atoms in the active pocket, were completely relaxed and fully optimized. The electronic embedding scheme was used to incorporate the MM point charge into the one-electron Hamiltonian of the QM calculation to avoid hyperpolarization of the QM wave function.58 The QM/MM calculations were performed by the ChemShell package59 incorporating the Turbomole module60 and DL-POLY program61 for the QM and MM regions, respectively. Geometry optimizations were implemented by the hybrid delocalized internal coordinates (HDLC) optimiser62 in Chemshell. During the optimizations, the limited memory Broyden–Fletcher–Goldfarb–Shanno (L-BFGS) algorithm63,64 was used to search for minima and the partitioned rational function optimization (P-RFO) algorithm65 was used for a transition state search. The L-BFGS algorithm is one of the Quasi-Newton methods and approximates the inverse Hessian matrix using a limited memory variation of the BFGS update. It is an excellent method for large scale optimization and is well suited for optimization problems with many variables. The P-RFO algorithm is Hessian eigenmodefollowing, and the found transition states are characterized by

Fig. 5 The selected quantum mechanics (QM) subsystem in QM/MM calculations for the deamination process (a) and pathway HL-p1 of the hydrolysis process (b).


Org. Biomol. Chem.

View Article Online

Paper


a single negative eigenvalue.59 There was no cutoff for the nonbonding MM and QM/MM interactions. Finally, high-level single point energy calculations were performed at a larger basis set of B3LYP/(SDD, 6-31++G(d,p)) to obtain accurate energies. The energies reported in the paper are all energies at this level.

bonds with residues D8, D51 and C167, implying the important role of K122 in the catalysis. In addition, two hydrogen bonds with lengths of 1.80 and 2.64 Å are formed between NAM and two residues A163 and D8. These hydrogen bonds form a large hydrogen bonding network in the active site. 3.2.


3. Results and discussion 3.1.

Structure of the Michaelis reactant complex

A reasonable structure of the Michaelis complex is critical for the description of the reaction mechanism. In our work, the initial model was constructed from a reaction intermediate, which was followed by a series of MD simulations and QM/ MM optimization. The Michaelis complex is quite stable during the MD simulation and QM/MM calculation, and thus the used QM/MM model is reasonable. The QM/MM optimized solvated model and the structure taken from the active site are shown in Fig. 6. Compared with the crystal structure of the reaction intermediate, there are some slight deviations on the positions of the residues in the two structures. These differences are reasonable considering the fact that the crystal structure is a reaction intermediate rather than a reactant complex. During the QM/MM optimization, the coordination shell of the Zn2+ ion is very stable. Similar to the crystal structure, the Zn2+ ion is octahedron-coordinated with H94, D51, H53, NAM, W1 and W2, and the distances between the Zn2+ ion and the binding atoms are 2.23, 2.06, 2.13, 2.67, 2.06 and 2.06 Å, respectively, indicating that these coordinated molecules interact strongly with the Zn2+ ion. The sulfhydryl group of C167 interacts with the carboxyl group of D8 in a favorable orientation for its activation and the distance between the hydrogen atom of the sulfhydryl group and the oxygen atom of the NAM carboxyl group is 2.93 Å, which indicates that these residues are nicely positioned for the proton transfer from C167 to D8. Several other hydrogen bonds are also found in the active site. The N–H bonds of the residue K122 form three hydrogen

Fig. 6

Reaction pathways

Based on the structure of the Michaelis complex, the reaction pathways were systematically studied by using QM/MM calculations. The whole catalytic reaction can be divided into two main parts namely the deamination process and the hydrolysis process. To clearly elucidate the catalytic cycle, the two processes are respectively discussed in the following sections. 3.2.1. Deamination process. From our calculations, we found that the deamination process contains four elementary steps, which can be described as three parts, as shown in Fig. 2: the activation of C167, the formation of the C–S bond coupled with the proton transfer from D8 to NAM, and the release of ammonia. Two possible pathways are considered for this process. In DA-p1, the formation of the C–S bond is prior to the proton transfer from D8 to NAM, whereas in DA-p2, the proton transfer is prior to the formation of the C–S bond. The optimized structures and key parameters of the reactant (R), transition states (TS1, TS2, TS3 and TS4) and intermediates (IM1, IM2, IM3 and IM4) are shown in Fig. 7. For a clear view, only the Zn2+ ion, NAM, D8, K122 and C167 are present in the figure. The residue C167 is firstly activated by donating a proton from its sulfhydryl group to the D8 carboxyl group. In this elementary step, the distance (r2) between the hydrogen atom of C167 and the oxygen atom of D8 decreases from 2.93 Å in R to 1.20 Å in TS1. After TS1, r2 further decreases to 0.96 Å, generating the sulfur anion intermediate to IM1. In the following reaction, two possible pathways were considered. In DA-p1, the sulfur anion of C167 is firstly ligated to the carbonyl carbon of NAM by a nucleophilic attack, with the distance (r3) of the two bonded atoms changing from 3.42 Å in IM1 to 1.96 Å in IM2-1. After forming a C–S bond, the carboxyl hydrogen atom of D8 transfers to the amino

The QM/MM optimized solvated model and the structure taken from the active site.

Org. Biomol. Chem.


View Article Online

Paper



Fig. 7

Optimized geometries for various species in the deamination process. The key bond distances are shown in angstrom.

group of NAM. From IM2-1 to TS3-1, the distance (r4) between the transferred hydrogen atom and nitrogen atom of the NAM amino group decreases from 1.68 Å to 1.37 Å. Downhill from TS3-1 to IM3, r4 further shortens to 1.02 Å, indicating the completion of the proton transfer. In DA-p2, the proton transfer firstly occurs with the distance (r4′) between the hydrogen atom and nitrogen atom changing from 2.05 Å to 1.07 Å to generate the intermediate IM2-2. In IM2-2, the strong electrostatic attraction between the ammonium group and sulfur anion induces a small conformational change of the substrate. The angle defined by the C–O bond of NAM and sulfur atom of C167 changes from 90.32° in IM1 to 134.69° in IM2-2 (the data are not shown in Fig. 7). Then, the distance (r3′) between the sulfur atom of C167 and carbonyl carbon atom of NAM decreases from 3.73 Å to 1.88 Å, indicating the formation of the C–S bond. The resulting tetrahedral intermediates (IM3) of the two


pathways are identical. In the following step, the sp3 hybrid carbon atom connected to the pyridine ring becomes sp2 hybridized as the release of ammonia. The length of the C–N bond (r5) changes to 2.94 Å (IM4) from 1.60 Å (IM3) via 1.86 Å (TS4). The energy profiles of the deamination process are shown in Fig. 8. It can be seen that the activation of C167 is calculated to be endothermic by 18.2 kcal mol−1 with an energy barrier of 25.7 kcal mol−1. The following two competitive pathways, the formation of the C–S bond in DA-p1 and proton transfer in DA-p2, correspond to very similar energy barriers (15.1 and 14.8 kcal mol−1). However, the relative energy of the intermediate IM2-1 is 10.9 kcal mol−1 higher than that of IM22, meaning IM2-2 is more stable than IM2-1. This can be understood by comparing the structures of IM2-1 and IM2-2. In IM2-2, the proton of the D8 carboxyl group has been transferred to NAM to form an ammonium group, which greatly

Org. Biomol. Chem.

View Article Online


Paper


Fig. 8 The energy profiles of the entire Pnc1 catalytic reaction. For the deamination process, both the pathways DA-p1 and DA-p2 are shown. But for the hydrolysis process, only HL-p1 is shown here.

stabilizes the sulfur anion of C167. The calculated energy barrier for the formation of IM3 is 1.4 kcal mol−1 in DA-p1, while it is 10.3 kcal mol−1 in DA-p2. The relative energy of IM3 is comparable with that of IM2-2. In the final step, the release of the first product ammonia is calculated to be facile with an energy barrier of only 3.1 kcal mol−1. In general, the overall deamination process is calculated to be endothermic by 8.6 kcal mol−1, and the activation of C167 contributes a lot to the overall energy barrier. 3.2.2. Hydrolysis process. Our calculation results suggest HL-p1 to be the most possible pathway for the hydrolysis of the thioester intermediate. In this pathway, the catalytic water molecule comes from the bulk solvent. The optimized structure of the active site by MD simulations and subsequent QM/ MM calculations is shown in Fig. 9. Among the four added water molecules, only two water molecules (W3 and W4) are located in the vicinity of the thioester intermediate. In particular, W3 situated in a favorable position for its nucleophilic attack on the carboxyl carbon atom of NAM. W3 forms a hydrogen bond with D8 with a length of 2.03 Å, and the oxygen atom of W3 is 2.89 Å, far away from the carbonyl carbon of NAM. In addition, the ESP charge of the W3 oxygen atom is −1.01, indicating that the water molecule here is a strong nucleophile. The Zn2+ ion is still octahedron-coordinated and the residue K122 is still hydrogen bonded with D8 and D51 with lengths of 1.71 Å and 1.86 Å, respectively. Compared with the initial structure in the deamination process, there are only minor changes in the coordination of Zn2+ and lengths of hydrogen bonds. Based on our calculations, the hydrolysis process in the HLp1 pathway proceeds in three elementary steps, including the water decomposition, the cleavage of the C–S bond and the proton transfer from D8 to C167. The optimized structures and key parameters of the intermediates (IM5, IM6 and IM7), transition states (TS5, TS6 and TS7) and product (P) are shown in Fig. 10. In the first step, the water W3 is calculated to be decomposed in a concerted manner. In TS5, the distance (r6) between

Org. Biomol. Chem.

Fig. 9 The optimized structure of an active site for the hydrolysis process.

the oxygen atom of W3 and carbonyl carbon of NAM decreases to 1.58 Å from 2.89 Å, and the hydrogen bond length (r8) between D8 and W3 shortens to 1.26 Å from 2.03 Å. Downhill from TS5, these distances further shorten to form intermediate IM6, and r6 and r8 become 1.44 Å and 0.97 Å, respectively. In the cleavage process of the C–S bond from IM6 to IM7, the bond length (r3) increases from 1.92 Å to 3.11 Å via 2.28 Å in TS6, indicating the regeneration of the sulfur anion intermediate. The last step of the entire cycle is the hydrogen transfer from D8 to C167 to generate the product NA. The catalytic residues return back to their initial states and the enzyme activity is restored. The distance (r9) between the sulfur atom of C167 and carboxyl hydrogen of D8 decreases from 2.37 Å to 1.33 Å via 1.62 Å. The energy profiles of the hydrolysis process of the pathway HL-p1 are shown in Fig. 8. It can be seen that the water decomposition, from IM5 to IM6, is the most energy-demanding step with an energy barrier of 28.1 kcal mol−1 and is


View Article Online



Fig. 10

Paper

Optimized geometries of various species in pathway HL-p1 for hydrolysis process. The key bond distances are shown in angstrom.

endothermic by 18.9 kcal mol−1. Thus, water decomposition is the rate limiting step of the hydrolysis process. In general, the whole hydrolysis process is exothermic by 8.9 kcal mol−1. The optimized structures and energy profiles of the water decomposition in HL-p2 are shown in Fig. 11. In this pathway, the Zn-coordinated water molecule W2 moves to the vicinity of the thioester intermediate from the coordination shell, resulting in a small conformational change of the zinc coordination shell. The most notable change is the formation of a hydrogen bond between W1 and NAM with a length of 2.40 Å, which leads to the increase of the distance between the Zn2+ ion and nitrogen atom of NAM from 2.81 Å in IM5 to 3.22 Å in IM5′. Thus, the interaction between the Zn2+ ion and NAM in IM5′ is weaker than that in IM5. The calculated energy barrier of this process is 33.2 kcal mol−1, which is about 5.0 kcal mol−1 higher

Fig. 11 The optimized structures and energy profiles of the water decomposition in HL-p2. The key bond distances are shown in angstrom.


than that of the rate-limiting step of the pathway HL-p1, indicating HL-p2 to be an unfavorable pathway. For HL-p3, the first step is the deprotonation of the Zn-coordinated water molecule W1 by D51 to form a Zn-binding hydroxyl group. We scanned the reaction coordinate (defined as the distance between the hydrogen atom of W1 and oxygen atom of the D51 carboxyl group) in a step size of 0.02 Å starting from the initial optimized structure of the hydrolysis process. Along the reaction coordinate from 1.82 Å to 1.00 Å (Fig. S2 in ESI†), the energy was continuously increased and no saddle point was found. Thus, these two pathways HL-p2 and HL-p3 can be ruled out. 3.3.

Role of some key residues and the Zn-binding site

Experimental data suggested that residues D8, K122 and C167 form a catalytic triad,14,16,17 which is similar to that of the nitrilase superfamily.66 Our calculations revealed that residues D8 and C167 are directly involved in the bond formation and cleavage processes. Specifically, D8 acts as the proton transfer station between C167 and NAM, whereas C167 serves as a nucleophile to attack the carbonyl carbon of the substrate. With regard to K122, the optimized structures of all species show that it always forms hydrogen bonds with D8 and C167 (as shown in Table S1†). Therefore, the role of K122 is suggested to be mainly in stabilizing the intermediates and transition states. An oxyanion hole formed by the amide backbone nitrogen atoms of A163 and C167 has been proposed in the ref. 16 and 17. The structure of the oxyanion hole in the reactant R is shown in Fig. 12. The amide backbone of A163 is hydrogen bonded with the carboxyl oxygen of NAM with a bond length (d3) of 1.80 Å, while the amide backbone of A167 locates at a farther place of NAM with a distance (d4) of 3.31 Å. The values of d3 and d4 in the optimized structures of all species are shown in Table S1.† During the reaction, the hydrogen bond between A163 and NAM is rather stable; only slight changes

Org. Biomol. Chem.

View Article Online


Paper

Fig. 12 The structure of the oxyanion hole in the reactant R and intermediate IM2-1. The distances are shown in angstrom.

can be observed in d3 except in IM2-2, which is possibly due to the strong electrostatic attraction between the ammonium group and sulfur anion. In R, the amide backbone of A167 interacts weakly with the substrate. Nevertheless, d4 is greatly shortened in the structures of intermediates and transition states when NAM is in its hydroxyl anion form. For example, in intermediate IM2-1 (also shown in Fig. 12), d4 greatly shortens to 2.13 Å. Thus, we can conclude that the oxyanion hole plays an important role in stabilizing the hydroxyl anion of NAM in the reaction, which can facilitate the nucleophilic attack by C167 and catalytic water. The metal dependence of Pnc1 kinetics and steady-state kinetics for Pnc1 mutants showed that the Zn-binding site is necessary for the catalytic reaction.16,18,67 After removing the metal cations, the wild type of Pnc1 was completely inactive. With the addition of the Zn2+ ion, the activity recovered greatly with a 90% rate relative to that of wild type Pnc1.18 The kcat/Km value was reduced by 200-fold when the pyridyl nitrogen atom of NAM was substituted by a carbon atom, indicating the important role of the Zn2+ ion for the binding substrate. Mutants of the Zn-binding residues D51, H53 and H94 individually to Ala reduce the kcat values by 10–50-fold lower than that of the wild type Pnc1.16 This means that these binding residues also contribute to the catalysis greatly. However, it is still unclear how the Zn-binding site regulates the reaction. In this study, detailed changes of the role of the Zn-binding site were further explored. The distances between the Zn2+ ion and its coordinated atoms in the deamination (DA-p1) and hydrolysis processes (HL-p1) are compared, which are shown in Fig. 13(a). One can see that the distances between Zn2+ and the nitrogen atom of NAM dZn–N(NAM) show a clear fluctuation along the reaction coordinate. In R, the distance dZn–N(NAM) is 2.67 Å, while in IM1 it decreases to 2.43 Å. It is mainly caused by the formation of the sulfur anion which directly interacts with NAM. In the following process, the deprotonated C167 nucleophilically attacks the carbonyl carbon atom to generate a hydroxyl anion. The electron attractivity of NAM is increased and the interaction between NAM and the Zn2+ ion is enhanced accordingly. Thus, in IM2-1, dZn–N(NAM) decreases to a minimum of 2.31 Å in the deamination process. After IM2-1, the proton transfer from D8 to the amino group of NAM can partly neutralize the positive charge of NAM, and dZn–N(NAM) begins to

Org. Biomol. Chem.


increase. In IM3 dZn–N(NAM) elongates to 2.5 Å and further increases to 2.95 Å in IM4 after the release of ammonia. Among the other distances of Zn-coordinated molecules, the distances between the Zn2+ ion and the ligated atoms of D51, H94, W1 and W2 are relatively stable. But the distances between the Zn2+ ion and D53 dZn–N(D53) show a contrary tendency, which may be caused by the trans effect. The same cases are found in the hydrolysis process of DA-p1 and the two processes of DA-p2 (Fig. S3†). For each geometry of the deamination process (DA-p1) and the hydrolysis process (HL-p1), the ESP charges of the Zn2+ ion, Zn-coordinated residues and water molecules, and Zn-binding site in the absence of NAM were calculated, which are shown in Fig. 13b and 13c. We can see that the positive charges of the Znbinding site are proportional to dZn–N(NAM), indicating that the Zn2+ ion influences the reaction possibly by means of the distance from NAM. In addition, we note that the positive charges of the Zn2+ ion are disproportional to the change of dZn–N(NAM), but are inversely proportional to the electric charge changes of D51, H53, H93, W1 and W2. Therefore, we can conclude that all these binding residues and water molecules participate in regulating the electric charge of the Zn2+ ion. The same situation is found in DA-p2, as shown in Fig. S3.† Thus, it can be concluded that the Zn-binding site rather than the single Zn2+ ion acts as a Lewis acid in the enzymatic reaction.

4.

Conclusions

In this paper, the catalytic mechanism of the Pnc1 has been investigated by using a combined quantum-mechanical/molecular-mechanical (QM/MM) approach. Our calculation results indicate that the deamination process contains four elementary steps and the subsequent hydrolysis process includes three steps. In the deamination process, a sulfur anion intermediate is firstly formed, but in the following two elementary reactions, both the proton transfer from D8 carboxyl to NAM and formation of the C–S bond may occur firstly, i.e., the proton transfer may occur either prior to or later than the formation of the C–S bond. In the hydrolysis process, the catalytic water molecule comes from the bulk solvent, which is then deprotonated by D8. D8 acts as a proton transfer station between C167 and NAM, while C167 serves as the nucleophile to attack the carbonyl carbon of NAM. The residue K122 always interacts with C167 and D8, playing a role in stabilizing intermediates and transition states. The oxyanion hole formed by the amide backbone nitrogen atoms of A163 and C167 plays an important role in stabilizing the hydroxyl anion of the substrate. The Zn-binding site rather than the single Zn2+ ion acts as a Lewis acid to influence the reaction, in which all binding residues and water molecules participate in regulating the electric charge of the Zn2+ ion. At the QM/MM B3LYP/6-31++G(d,p)// CHARMM22 level of theory, two elementary steps, the activation of C167 in the deamination process and the decomposition of catalytic water in the hydrolysis process, are calculated to correspond to the large energy barriers of 25.7


View Article Online

Paper



Fig. 13 The distances between Zn2+ and the ligated atoms of Zn-coordinated molecules (a), the ESP charges of the Zn-binding site without NAM (b), and the ESP charges of Zn2+ and Zn-coordinated residues and water molecules (c) of all species in DA-p1 for the deamination process and HL-p1 for the hydrolysis process.

and 28.1 kcal mol−1, respectively. It means that these two residues contribute a lot to the overall energy barrier. Our results may provide useful information for the development of novel and efficient inhibitors that cannot be resisted by strains, and for the enzymatic redesign of biocatalytic applications.

Abbreviations Pnc1 NAD+ NAM NA NAMN

Nicotinamidase Nicotinamide adenine dinucleotide Nicotinamide Nicotinic acid Nicotinamide mononucleotide


PncA PAM PA

Pyrazinamidase Pyrazinamide Pyrazinoic acid.

Acknowledgements This work was supported by the National Natural Science Foundation of China (21173129, 21373125).

References 1 S. J. Lin and L. Guarente, Curr. Opin. Cell Biol., 2003, 15, 241–246.

Org. Biomol. Chem.

View Article Online


Paper

2 H. Lin, Org. Biomol. Chem., 2007, 5, 2541–2554. 3 A. B. Pardee, E. J. Benz Jr., D. A. St Peter, J. N. Krieger, M. Meuth and H. W. Trieshmann Jr., J. Biol. Chem., 1971, 246, 6792–6796. 4 G. Sánchez-Carrón, M. I. García-García, R. Zapata-Pérez, H. Takami, F. García-Carmona and A. Sánchez-Ferrer, PLoS One, 2013, 8, e56727. 5 M. Ghislain, E. Talla and J. M. Francois, Yeast, 2002, 19, 215–224. 6 C. Yan and D. L. Sloan, J. Biol. Chem., 1987, 262, 9082–9087. 7 G. D. Wang and E. Pichersky, Plant J., 2007, 49, 1020–1029. 8 V. Balan, G. S. Miller, L. Kaplun, K. Balan, Z. Z. Chong, F. Li, A. Kaplun, M. F. VanBerkum, R. Arking, D. C. Freeman, K. Maiese and G. Tzivion, J. Biol. Chem., 2008, 283, 27810–27819. 9 R. Takahashi, S. Nakamura, T. Nakazawa, K. Minoura, T. Yoshida, Y. Nishi, Y. Kobayashi and T. Ohkubo, J. Biochem., 2010, 147, 95–107. 10 World Health Organization. WHO Report on the Tuberculosis Epidemic. Stop TB at the Source, Vol. 270, Tuberculosis Programme, Geneva, Switzerland, 2008. 11 C. C. Leung, H. L. Rieder, C. Lange and W. W. Yew, Eur. Respir. J., 2011, 37, 690–711. 12 J. González-Martín, J. M. García-García, L. Anibarro, R. Vidal, J. Esteban, R. Blanquer, S. Moreno and J. RuizManzano, Arch Bronconeumol., 2010, 46, 255–274. 13 Y. Zhang and D. Mitchison, Int. J. Tuberc. Lung Dis., 2003, 7, 6–21. 14 X. Du, W. Wang, R. Kim, H. Yakota, H. Nguyen and S. Kim, Biochemistry, 2001, 40, 14166–14172. 15 G. Hu, A. B. Taylor, L. McAlister-Henn and P. J. Hart, Arch. Biochem. Biophys., 2007, 461, 66–75. 16 B. C. Smith, M. A. Anderson, K. A. Hoadley, J. L. Keck, W. W. Cleland and J. M. Denu, Biochemistry, 2012, 51, 243–256. 17 P. K. Fyfe, V. A. Rao, A. Zemla, S. Cameron and W. N. Hunter, Angew. Chem., 2009, 121, 9340–9343. 18 J. B. French, Y. Cen, A. A. Sauve and S. E. Ealick, Biochemistry, 2010, 49, 8803–8812. 19 S. Petrella, N. Gelus-Ziental, A. Maudry, C. Laurans, R. Boudjelloul and W. Sougakoff, PLoS One, 2011, 6, e15785. 20 E. Gazanion, D. Garcia, R. Silvestre, C. Gérard, J. F. Guichou, G. Labesse, M. Seveno, A. Cordeiro-Da-Silva, A. Ouaissi, D. Sereno and B. Vergnes, Mol. Microbiol., 2011, 82, 21–38. 21 X. Liu, H. Zhang, X. J. Wang, L. F. Li and X. D. Su, PLoS One, 2011, 6, e24227. 22 A. Scorpio, P. Lindholm-Levy, L. Heifets, R. Gilman, S. Siddiqi, M. Cynamon and Y. Zhang, Antimicrob. Agents Chemother., 1997, 41, 540–543. 23 H. Zhang, J. Y. Deng, L. J. Bi, Y. F. Zhou, Z. P. Zhang, C. G. Zhang, Y. Zhang and X. E. Zhang, FEBS J., 2008, 275, 753–762. 24 N. Lemaitre, I. Callebaut, F. Frenois, V. Jarlier and W. Sougakoff, Biochem. J., 2001, 353, 453–458.

Org. Biomol. Chem.


25 D. R. Seiner, S. S. Hegde and J. S. Blanchard, Biochemistry, 2010, 49, 9613–9616. 26 J. B. French, Y. Cen, T. L. Vrablik, P. Xu, E. Allen, W. Hanna-Rose and A. A. Sauve, Biochemistry, 2010, 49, 10421–10439. 27 J. L. Zhang, Q. C. Zheng, Z. Q. Li and H. X. Zhang, PLoS One, 2012, 7, e39546. 28 V. Rajendran and R. Sethumadhavan, J. Biomol. Struct. Dyn., 2014, 32, 209–221. 29 A. Warshel and M. Levitt, J. Mol. Biol., 1976, 103, 227–249. 30 M. J. Field, P. A. Bash and M. Karplus, J. Comput. Chem., 1990, 11, 700–733. 31 H. M. Senn and W. Thiel, Angew. Chem., Int. Ed., 2009, 48, 1198–1229. 32 J. L. Gao, S. H. Ma, D. T. Major, K. Nam, J. Z. Pu and D. G. Truhlar, Chem. Rev., 2006, 106, 3188–3209. 33 D. M. Li, X. Q. Huang, K. L. Han and C. G. Zhan, J. Am. Chem. Soc., 2011, 133, 7416–7427. 34 K. Doitomi, T. Kamachi, T. Toraya and K. Yoshizawa, Biochemistry, 2012, 51, 9202–9210. 35 H. Gómez, I. Polyak, W. Thiel, J. M. Lluch and M. Masgrau, J. Am. Chem. Soc., 2012, 134, 4743–4752. 36 J. I. Mujika, X. Lopeza and A. J. Mulhollandb, Org. Biomol. Chem., 2012, 10, 1207–1218. 37 L. K. Du, J. Gao, Y. J. Liu, D. J. Zhang and C. B. Liu, Org. Biomol. Chem., 2012, 10, 1014–1024. 38 D. M. Li, Y. Wang and K. L. Han, Coord. Chem. Rev., 2012, 256, 1137–1150. 39 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, GAUSSIAN 09 (Revision B.01), Gaussian, Inc., Wallingford CT, 2009. 40 G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner, R. K. Belew, D. S. Goodsell and A. J. Olson, J. Comput. Chem., 2009, 16, 2785–2791. 41 J. Gasteiger and M. Marsili, Tetrahedron, 1980, 36, 3219– 3228. 42 M. H. M. Olsson, C. R. Søndergard, M. Rostkowski and J. H. Jensen, J. Chem. Theory Comput., 2011, 7, 525–537. 43 W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graphics, 1996, 14, 33–38.


View Article Online



44 B. R. Brooks, R. E. Bruccoleri, B. D. Olafson, D. J. States, S. Swaminathan and M. Karplus, J. Comput. Chem., 1983, 4, 187–217. 45 A. D. MacKerell Jr., D. Bashford, M. Bellott, R. L. Dunbrack Jr., J. D. Evanseck, M. J. Field, S. Fischer, J. Gao, H. Guo, S. Ha, D. Joseph-McCarthy, L. Kuchnir, K. Kuczera, F. T. K. Lau, C. Mattos, S. Michnick, T. Ngo, D. T. Nguyen, B. Prodhom, W. E. Reiher III, B. Roux, M. Schlenkrich, J. C. Smith, R. Stote, J. Straub, M. Watanabe, J. Wiórkiewicz-Kuczera, D. Yin and M. Karplus, J. Phys. Chem. B, 1998, 102, 3586–3616. 46 W. L. Jorgensen, J. Chandrasekhar, J. D. Madura, R. W. Impey and M. L. Klein, J. Chem. Phys., 1983, 79, 926–935. 47 J. Y. Liang and W. N. Lipscomb, Proc. Natl. Acad. Sci. U. S. A., 1990, 87, 3675–3679. 48 Y. P. Pang, Proteins, 2001, 45, 183–189. 49 R. A. Karjiban, M. B. A. Rahman, M. Basri, A. B. Salleh, D. Jacobs and H. A. Wahab, Protein J., 2009, 28, 14–23. 50 R. H. Stote and M. Karplus, Proteins: Struct., Funct., Genet., 1995, 23, 12–31. 51 R. B. Wu, S. L. Wang, N. J. Zhou, Z. X. Cao and Y. K. Zhang, J. Am. Chem. Soc., 2010, 132, 9471–9479. 52 M. Dolg, U. Wedig, H. Stoll and H. Preuss, J. Chem. Phys., 1987, 86, 866–872. 53 C. Corminboeuf, P. Hu, M. E. Tuckerman and Y. K. Zhang, J. Am. Chem. Soc., 2006, 128, 4530–4531. 54 R. B. Wu, P. Hu, S. L. Wang, Z. X. Cao and Y. K. Zhang, J. Chem. Theory Comput., 2010, 6, 337–343.


Paper

55 S. F. Sousa, P. A. Fernandes and M. J. Ramos, J. Am. Chem. Soc., 2007, 129, 1378–1385. 56 M. J. Field, P. A. Bash and M. Karplus, J. Comput. Chem., 1990, 11, 700–733. 57 A. H. de Vries, P. Sherwood, S. J. Collins, A. M. Rigby, M. Rigutto and G. J. Kramer, J. Phys. Chem. B, 1999, 103, 6133–6141. 58 D. Bakowies and W. Thiel, J. Phys. Chem., 1996, 100, 10580– 10594. 59 M. F. Ruiz-Lopez, J. Mol. Struct. (THEOCHEM), 2003, 632, ix. 60 R. Ahlrichs, M. Bär, M. Häser, H. Horn and C. Kölmel, Chem. Phys. Lett., 1989, 162, 165–169. 61 W. Smith and T. R. Forester, J. Mol. Graph. Model., 1996, 14, 136–141. 62 S. R. Billeter, A. J. Turner and W. Thiel, Phys. Chem. Chem. Phys., 2000, 2, 2177–2186. 63 J. Nocedal, Math. Comput., 1980, 35, 773–782. 64 D. C. Liu and J. Nocedal, Math. Prog., 1989, 45, 503–528. 65 A. Banerjee, N. Adams, J. Simons and R. Shepard, J. Phys. Chem., 1985, 89, 52–57. 66 R. N. Thuku, D. Brady, M. J. Benedik and B. T. Sewell, J. Appl. Microbiol., 2009, 106, 703–727. 67 P. Sheen, P. Ferrer, R. H. Gilman, G. Christiansen, P. Moreno-Román, A. H. Gutiérrez, J. Sotelo, W. Evangelista, P. Fuentes, D. Rueda, M. Flores, P. Olivera, J. Solis, A. Pesaresi, D. Lamba and M. Zimic, Am. J. Trop. Med. Hyg., 2012, 87, 153–161.

Org. Biomol. Chem.

MM free energy calculations.

A multi-scale computational study on the mechanism of Streptococcus pneumoniae Nicotinamidase (SpNic).

The catalytic mechanism of carboxylesterases: a computational study.

MM) methods.

MM and QM cluster models.

Catalytic mechanism of glycogen phosphorylase.

MM study.

MM study on the fast thiol oxidation step.

MM study on the reaction pathway leading to 2-aceto-2-hydroxybutyrate in the catalytic cycle of AHAS.

MM MERP and PES calculations on retaining glycosyltransferase ppGalNAcT2.

MM study of the reaction mechanism of the carboxyl transferase domain of pyruvate carboxylase from Staphylococcus aureus.

MM study.

MM MD study.

MM calculations.

The Catalytic Mechanism of the Marine-Derived Macrocyclase PatGmac.

Reassessment of the catalytic mechanism of glycogen debranching enzyme.

Participation of buffer in the catalytic mechanism of carbonic anhydrase.

Structural insights into the catalytic mechanism of aldehyde-deformylating oxygenases.

Some insights into the binding mechanism of the GABAA receptor: a combined docking and MM-GBSA study.

MM Study of the Molecular Mechanism of Cellulose Polymerization in Bacterial CESA.

Structural basis for the catalytic mechanism of homoserine dehydrogenase.

Data on the catalytic mechanism of thiol peroxidase mimics.

The active site structure and catalytic mechanism of arsenite oxidase.

Is histidine involved in the catalytic mechanism of unspecific carboxylesterases?