Theoretical Studies on Catalysis Mechanisms of Serum ... - MDPI

2 downloads 0 Views 7MB Size Report
Jul 7, 2018 - the experimental research of Ben-David Moshe et al. ...... Thompson, J.D.; Higgins, D.G.; Gibson, T.J. CLUSTAL W: Improving the sensitivity of ...
molecules Article

Theoretical Studies on Catalysis Mechanisms of Serum Paraoxonase 1 and Phosphotriesterase Diisopropyl Fluorophosphatase Suggest the Alteration of Substrate Preference from Paraoxonase to DFP Hao Zhang 1 ID , Ling Yang 2, * Rong-Zhen Liao 5 1 2 3 4

5

*

ID

, Ying-Ying Ma 3 , Chaoyuan Zhu 4 , Shenghsien Lin 4 and

College of Life Science and Engineering, Northwest Minzu University, Lanzhou 730030, China; [email protected] MIIT Key Laboratory of Critical Materials Technology for New Energy Conversion and Storage, School of Chemistry and Chemical Engineering, Harbin Institute of Technology, Harbin 150001, China Inner Mongolia University of Technology, Hohhot 010051, China; [email protected] Department of Applied Chemistry, Institute of Molecular Science and Center for Interdisciplinary Molecular Science, National Chiao-Tung University, Hsinchu 30050, Taiwan; [email protected] (C.Z.); [email protected] (S.L.) Key Laboratory for Large-Format Battery Materials and System, Ministry of Education School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan 430074, China; [email protected] Correspondence: [email protected]; Tel.: +86-0451-8640-3305

Received: 2 June 2018; Accepted: 5 July 2018; Published: 7 July 2018

 

Abstract: The calcium-dependent β-propeller proteins mammalian serum paraoxonase 1 (PON1) and phosphotriesterase diisopropyl fluorophosphatase (DFPase) catalyze the hydrolysis of organophosphorus compounds and enhance hydrolysis of various nerve agents. In the present work, the phosphotriesterase activity development between PON1 and DFPase was investigated by using the hybrid density functional theory method B3LYP. Based on the active-site difference between PON1 and DFPase, both the wild type and the mutant (a water molecule replacing Asn270 in PON1) models were designed. The results indicated that the substitution of a water molecule for Asn270 in PON1 had little effect on the enzyme activity in kinetics, while being more efficient in thermodynamics, which is essential for DFP hydrolysis. Structure comparisons of evolutionarily related enzymes show that the mutation of Asn270 leads to the catalytic Ca2+ ion indirectly connecting the buried structural Ca2+ ion via hydrogen bonds in DFPase. It can reduce the plasticity of enzymatic structure, and possibly change the substrate preference from paraoxon to DFP, which implies an evolutionary transition from mono- to dinuclear catalytic centers. Our studies shed light on the investigation of enzyme catalysis mechanism from an evolutionary perspective. Keywords: enzyme evolution; hydrolysis; β-propeller protein; plasticity; reaction mechanism

1. Introduction Studies based on structural comparisons of evolutionally related enzymes can not only identify the role of the catalysis related residues, but also aid in understanding the way followed by the evolution of enzymes for new catalytic functions [1–7]. Recently, such studies have received more attention for the general insights into the enzymatic function. After reporting the evolution of

Molecules 2018, 23, 1660; doi:10.3390/molecules23071660

www.mdpi.com/journal/molecules

Molecules 2018, 23, 1660

2 of 12

the phosphotriesterase activity in the metallo-β-lactamase family [8], in the present work, we will investigate phosphotriesterase activity development in other evolutionally related enzymes, as well as the co-evolution of the active-site structure and catalytic function. The phosphotriesterase activity is highly focused for their detoxification ability toward agricultural pesticides, by degrading organophosphate (OP) compounds [9–11]. Molecules 2018, 23, x FOR PEER REVIEW 2 of 12 The mammalian serum paraoxonase 1 (PON1) is a calcium-dependent serum esterase which catalyzes the hydrolysis of a broad range of organic esters and OP compounds [12]. Although PON1 is phosphotriesterase activity in the metallo-β-lactamase family [8], in the present work, we will investigate phosphotriesterase activity other evolutionally enzymes, as well a lactonase with native substrates γanddevelopment δ-lactones inwhich have long related alkyl side chains [13–15], it is as the co-evolution of the active-site promiscuous structure and catalytic function. The phosphotriesterase activity more fascinating that PON1 possesses OP hydrolase activity, particularly on paraoxon, is highly focused for their detoxification ability toward agricultural pesticides, by degrading which is attributed to its considerable plasticity of catalytic structure [4,15]. The crystal structures organophosphate (OP) compounds [9–11]. show that PON1 a six-blade with calcium ionsserum in itsesterase centralwhich tunnel [15,16]. The has mammalian serum β-propeller paraoxonase 1fold (PON1) is atwo calcium-dependent catalyzes theishydrolysis of a broad range ofinside organicthe esters and OP and compounds [12]. Although PON1 The structural Ca2+ completely embedded protein, the catalytic Ca2+ is located at the a lactonase withcavity. native substrates γ- and δ-lactones which have long alkyl side chains [13–15], it is bottom of isthe active site more fascinating that PON1 possesses promiscuous OP hydrolase activity, particularly on paraoxon, The squid phosphotriesterase diisopropyl fluorophosphatase (DFPase) from Loligo vulgaris [17–19], which is attributed to its considerable plasticity of catalytic structure [4,15]. The crystal structures another structurally related β-propeller protein, relatively show that PON1 has acalcium six-bladecontaining β-propeller fold with two calcium ionsshows in its central tunnelspecific [15,16]. substrate 2+ 2+ structuralcatalyzing Ca is completely embedded inside the protein, fluorophosphate and the catalytic Ca (DFP) is located at G-type the preference,The efficiently the hydrolysis of diisopropyl and nerve bottom oftabun the active site cavity. agents, including (GA), sarin (GB), soman (GD), and cyclohexyl sarin (GF). The detoxification The squid phosphotriesterase diisopropyl fluorophosphatase (DFPase) from Loligo vulgaris of the OP [17–19], agent is achieved by the hydrolytic reaction producing a phosphate or phosphonate and another structurally related calcium containing β-propeller protein, shows relatively specific a fluoride substrate ion [19,20]. Sinceefficiently both DFPase and display distinctfluorophosphate detoxication(DFP) by catalyzing OP preference, catalyzing thePON1 hydrolysis of diisopropyl and including as tabun (GA), sarin (GB), soman (GD), and cyclohexyl sarin (GF). The hydrolysis,G-type theynerve wereagents, well studied catalytic bioscavengers to degrade OP compounds. of the OP agent achievedcompared by the hydrolytic reaction a phosphate or PON1detoxification and DFPase have been isalways together, forproducing their structural similarity and phosphonate and a fluoride ion [19,20]. Since both DFPase and PON1 display distinct detoxication functionalbyhomology [4,16]. DFPase, isolated from the central nervous system of the squid, catalyzing OP hydrolysis, they were well studied as catalytic bioscavengers to degrade OP shows strong preference for the hydrolysis of P–F or P–CN bonds, which are absent in natural compounds. PON1 DFPasenative have been alwaysiscompared for their which structural similarity and compounds [21–23].and PON1’s function likely totogether, be a lactonase hydrolyzes the lactones functional homology [4,16]. DFPase, isolated from the central nervous system of the from the oxidized lipids [24,25], but its promiscuous hydrolase activity onsquid, OP shows received more strong preference for the hydrolysis of P–F or P–CN bonds, which are absent in natural compounds attention [4,15,16]. Thus, the common detoxifying function of both enzyme systems via OP hydrolysis [21–23]. PON1’s native function is likely to be a lactonase which hydrolyzes the lactones from the was focused uponlipids here.[24,25], but its promiscuous hydrolase activity on OP received more attention oxidized [4,15,16]. Thus, the that common detoxifying of both enzyme systems OP hydrolysis was It was suggested PON1 andfunction DFPase employ similarviacatalytic mechanisms as focused upon here. phosphotriesterase, due to their structural similarities of active sites [15,26]. The attacking nucleophile It was suggested that PON1 and DFPase employ similar catalytic mechanisms as for phosphotriester hydrolysis was identified, by our recent work, to be an activated water molecule, phosphotriesterase, due to their structural similarities of active sites [15,26]. The attacking with the nucleophile attacking the phosphorus center (Figureby1)our [27]. This conclusion was supported by nucleophile for phosphotriester hydrolysis was identified, recent work, to be an activated water molecule, with the nucleophile attacking the phosphorus center (Figure 1) [27]. the experimental research of Ben-David Moshe et al.: that the E53Q and D269N mutants This in PON1 both was lactonase supported by theparaoxonase experimental research of Ben-David Moshemutation et al.: that studies the E53Q combined possessed conclusion measurable and activity, and sufficient and D269N mutants in PON1 both possessed measurable lactonase and paraoxonase activity, and with related molecular dynamics simulations suggested that the water activated by Glu53 and Asp269 sufficient mutation studies combined with related molecular dynamics simulations suggested that should bethe the most likelybyattacking nucleophile [4]. Since, in attacking our previous work, we have fully water activated Glu53 and Asp269 should be the most likely nucleophile [4]. Since, previous property work, we details have fully discussed the chemical property the catalytic discussed in theour chemical of the catalytic mechanism [27],details in theofpresent work, we will mechanism [27], in the present work, we will mainly focus on the rate-determining reaction step mainly focus on the rate-determining reaction step of the OP hydrolysis catalyzed both by of DFPase and the OP hydrolysis catalyzed both by DFPase and PON1, and investigate the biological significance of PON1, andactive-site investigate the biological significance of active-site residues. residues.

Figure 1. The reaction mechanism for PON1 and DFPase.

Figure 1. The reaction mechanism for PON1 and DFPase.

Molecules 2018, 23, 1660

Molecules 2018, 23, x FOR PEER REVIEW

3 of 12

3 of 12

Thecrystal crystal structures structures of of several several structurally andand the the The structurally related relatedenzymes enzymesPON1, PON1,DFPase, DFPase, gluconolactonase from Xanthomonas campestris (XC5397) [28] adopt a classical six-blade β-propeller gluconolactonase from Xanthomonas campestris (XC5397) [28] adopt a classical six-blade β-propeller fold, while the coordinated ligand Asn270 of the metal center is observed in PON1, exclusively fold, while the coordinated ligand Asn270 of the metal center is observed in PON1, exclusively (Figure 2 A-C). Among all the coordinating residues in PON1, the role of Asn270 has always been (Figure 2A–C). Among all the coordinating residues in PON1, the role of Asn270 has always been ignored. The site-specific saturation mutagenesis indicated Asn270 seems to play the smallest role in ignored. The site-specific saturation mutagenesis indicated Asn270 seems to play the smallest role in catalysis for both paraoxon and thio-buthyl butyriclactone (TBBL) [4]. However, from an catalysis for both paraoxon and thio-buthyl butyriclactone (TBBL) [4]. However, from an evolutionary evolutionary perspective, the mutation rate at active sites, such a core functional region, should be perspective, the mutation rate at active sites, a core functional region, should be lead very to low across very low across a protein family. Once suchsuch mutations have occurred, they should new a protein family. Onceinsuch mutations have occurred, should lead to new functional branches functional branches a protein family [29–31]. Thusthey Asn270, preserved in PON1 exclusively, in among a protein family [29–31]. Thus Asn270, preserved in PON1 exclusively, among the family with the the family with the typical six-blade β-propeller fold, should have a certain evolutionary typical six-blade β-propeller fold, should have a certain evolutionary meaning for the core enzymatic meaning for the core enzymatic function. Thus, in this study, we investigated the role of Asn270 for function. Thus,function in this study, we and investigated the role of Asn270 for the catalytic function of PON1 the catalytic of PON1 the potential reason for it preserving in PON1, but being absent and theinpotential reason it preserving in PON1,theory but being absent in DFPase, by using hybridwith density DFPase, by usingfor hybrid density functional (DFT) with B3LYP functional combined the clustertheory models of thewith active site. The methodscombined were widely andthe successfully used on functional (DFT) B3LYP functional with cluster models ofthe theenzyme active site. reaction modeling by Ramos’s group of zincused enzymes Himo and Siegbahn’s group [36–39], The methods were widely and successfully on the[32–35], enzyme reaction modeling by Ramos’s group et al.[32–35], [40–43]. Himo and Siegbahn’s group [36–39], and Russo et al. [40–43]. of and zincRusso enzymes

Figure Activesite sitestructures structures of of PON1 PON1 (PDB DFPase (PDB code: 2GVW) [18] [18] Figure 2. 2.Active (PDBcode: code:3SRE) 3SRE)[15] [15](A), (A), DFPase (PDB code: 2GVW) (B) and XC5397 (PDB code: 3DR2) [28] (C), respectively. These figures are drawn by the PyMOL [44]. (B) and XC5397 (PDB code: 3DR2) [28] (C), respectively. These figures are drawn by the PyMOL [44].

Although the evolution of the enzymatic structures and mechanisms have been mentioned Although thefar, evolution of the mechanisms beenenzyme, mentioned early [45,46], so most studies stillenzymatic focused on structures the catalyticand mechanism of an have individual early [45,46], so far, mostinformation studies stillonfocused on the function. catalytic In mechanism individual which provides limited the enzymatic this study, of in an order to exploreenzyme, the which providesoflimited information onactivity the enzymatic function. this study, in order explore development the phosphotriesterase between PON1 and In DFPase, we have made atomore the catalytic mechanisms the evolutionally enzymes. thecomprehensive development study of theon phosphotriesterase activity of between PON1 and related DFPase, we haveThe made knowledge of how nature has evolved new catalytic reactions via the conserved structural features a more comprehensive study on the catalytic mechanisms of the evolutionally related enzymes. at knowledge active sites of can help understand the diversity of the reactions enzyme catalytic mechanisms and further The how nature has evolved new catalytic via the conserved structural features help guide enzyme design in the laboratory [5,6]. at active sites can help understand the diversity of the enzyme catalytic mechanisms and further help

guide enzyme design in the laboratory [5,6]. 2. Results and Discussion

2. Results and Discussion

2.1. Effects of Asn270 Mutation on Catalytic Reaction

2.1. Effects Asn270 Mutation on Catalytic The ofactive-site models of wild typeReaction PON1 (Active-site-W) were built on the basis of the high-resolution crystal structure (PDB code: Then, the models of the catalysis mechanism The active-site models of wild type 3SRE) PON1[15]. (Active-site-W) were built on the basis of the via the attacking of a water molecule (named W1 here) were designed by adding the substrates high-resolution crystal structure (PDB code: 3SRE) [15]. Then, the models of the catalysis mechanism and DFP into the Active-site-W model, called PON1W-Paraoxon model and PON1W-DFP viaparaoxon the attacking of a water molecule (named W1 here) were designed by adding the substrates model, respectively. These models contain one Ca2+ ion and the coordinating residues (Asp269, paraoxon and DFP into the Active-site-W model, called PON1W-Paraoxon model and PON1W-DFP Glu53, Asn168, Asn224, and Asn270) (Figure 3, A-React and C-React). These residue ligands were model, respectively. These models contain one Ca2+ ion and the coordinating residues (Asp269, Glu53, truncated, and the side chains were retained. In order to keep the optimized structures close to the Asn168, Asn224, and Asn270) (Figure 3, A-React and C-React). These residue ligands were truncated, crystal structures, the truncation atoms were frozen at their X-ray positions in the geometry and the side chains were atoms retained. In order to asterisks. keep the Hydrogen optimizedatoms structures close to the crystal optimization. The frozen are marked with were added manually,

structures, the truncation atoms were frozen at their X-ray positions in the geometry optimization.

Molecules 2018, 23, 1660

4 of 12

Molecules 2018,are 23, xmarked FOR PEER with REVIEW 4 of truncated 12 The frozen atoms asterisks. Hydrogen atoms were added manually, and the bonds were saturated by hydrogen atoms. This procedure has been widely applied in the studies on and the truncated bonds were saturated by hydrogen atoms. This procedure has been widely reactionapplied mechanisms [11,47–49]. Themechanisms total charge of models zero. in the studies on reaction [11,47–49]. Theis total charge of models is zero.

Figure 3. Optimized geometries of reactants along the reaction paths of (A) PON1W-Paraoxon, (B)

Figure 3. Optimized geometries of reactants along the reaction paths of (A) PON1W-Paraoxon, PON1M-Paraoxon, (C) PON1W-DFP, and (D) PON1M-DFP models. The frozen atoms are marked (B) PON1M-Paraoxon, (C) PON1W-DFP, and (D) PON1M-DFP models. The frozen atoms are marked with asterisks. with asterisks.

Furthermore, in order to investigate the effects of Asn270 on the phosphotriesterase activity, the mutant models, Active-site-M, PON1M-Paraoxon, and PON1M-DFP models, were derived from the Furthermore, in order to investigate the effects of Asn270 on the phosphotriesterase activity, the Active-site-W, PON1W-Paraoxon, and PON1W-DFP models, respectively, by replacing Asn270 into mutant amodels, Active-site-M, PON1M-Paraoxon, andand PON1M-DFP models, derived water molecule (named W2 here) (Figure 3, B-React D-React). These mutantwere models have thefrom the Active-site-W, PON1W-Paraoxon, PON1W-DFP by replacing Asn270 into same coordination as in DFPase and (Figure 2), which alsomodels, represent respectively, the active-site models of DFPase. In the(named main-chain of Glu53, makesand a stable hydrogen bond mutant with the models side-chainhave the a wateraddition, molecule W2oxygen here) atom (Figure 3, B-React D-React). These nitrogen atomas of in Asn270. Thus(Figure the main-chain oxygen atom of Glu 53the andactive-site the hydrogen bond with same coordination DFPase 2), which also represent models of DFPase. the side-chain nitrogen atom of Asn270 were considered in the wild type models In addition, the main-chain oxygen atom of Glu53, makes a stable hydrogen bond with the side-chain (PON1W-Paraoxon and PON1W-DFP), while absent in the mutant models (PON1M-Paraoxon and nitrogenPON1M-DFP) atom of Asn270. Thus theofmain-chain oxygen atom of Glu 53 and the hydrogen bond with for the mutation Asn270. the side-chain of Asn270 were in the wild type models (PON1W-Paraoxon Thenitrogen reaction isatom supposed to proceed byconsidered a two-step addition–elimination (An + Dn) mechanism [50], which goes through a common pentavalent intermediate observed in and phosphoryl transfer for the and PON1W-DFP), while absent in the mutant models (PON1M-Paraoxon PON1M-DFP) reactions. Similar to the mechanism of DFPase [27], the first step is the activation of H 2O by Asp269 mutation of Asn270. and Glu53 to form a nucleophile hydroxide and a metastable pentavalent complex. Here, we mainly The reaction is supposed to proceed by a two-step addition–elimination (An + Dn) mechanism [50], focused on the Dn step. Optimized geometries of the models active-site-W, active-site-M, and the which goes through a common pentavalent intermediate observed in phosphoryl transfer reactions. four Michaelis complexes (A–D-React in Figure 3) PON1W-Paraoxon, PON1M-Paraoxon, Similar PON1W-DFP, to the mechanism of DFPase [27], the infirst step theTable activation of H by either Asp269 and 2 Othat and PON1M-DFP were listed Table 1. is From 1, we can see Glu53 to form aAsn270 nucleophile hydroxide a metastable pentavalent complex. Here, we mainly replacing or adding substratesand results in almost consistent changes of the coordination lengths of Ca–OGlu53 and Ca–Oof Asn168 while the Ca–Oactive-site-M, Asp269 and Ca–Oand Asn224 the four focusedbonds. on theBond Dn step. Optimized geometries thedecreased, models active-site-W, bonds increased. For the mutant models, the substitute water molecule W2 connects the side chain of Michaelis complexes (A–D-React in Figure 3) PON1W-Paraoxon, PON1M-Paraoxon, PON1W-DFP, and Asp269 via a strong hydrogen bond, and the hydrogen bond distances of OW2bOAsp269 are ca. 2.6–2.7

PON1M-DFP were listed in Table 1. From Table 1, we can see that either replacing Asn270 or adding substrates results in almost consistent changes of the coordination bonds. Bond lengths of Ca–OGlu53 and Ca–OAsn168 decreased, while the Ca–OAsp269 and Ca–OAsn224 bonds increased. For the mutant models, the substitute water molecule W2 connects the side chain of Asp269 via a strong hydrogen

Molecules 2018, 23, 1660

5 of 12

bond, and the hydrogen bond distances of OW2 bOAsp269 are ca. 2.6–2.7 Å along the reaction paths (see Figures 3 and 4), which is consistent with the crystal structures of DFPase. The optimized geometries of transition states and products were shown in Figure 4 Molecules 2018, 23, x FOR REVIEW of 12 activated (A: PON1W-Paraoxon, B:PEER PON1M-Paraoxon, C: PON1W-DFP, D: PON1M-DFP). 5The nucleophile W1 attacks the phosphorus center and leads the P–O or P–F bond, broken to form the Å along the reaction paths (see Figures 3 and 4), which is consistent with the crystal structures of hydrolysisDFPase. product. Glu53 helps orientate the W1 by accepting a hydrogen bond for the wild model, The optimized geometries transition and from products shown in Figure 4 (A: following but acts as a general base to abstractofthe secondstates proton W1were for the mutant models PON1W-Paraoxon, B: PON1M-Paraoxon, C: PON1W-DFP, D: PON1M-DFP). The activated the corresponding transition state. Both the residues Asn168 and Asn224 help the leaving group nucleophile W1 attacks the phosphorus center and leads the P–O or P–F bond, broken to form the leave through hydrogen bonds for models PON1M-Paraoxon, PON1W-DFP, and hydrolysis product. Glu53 helps orientatePON1W-Paraoxon, the W1 by accepting a hydrogen bond for the wild model, PON1M-DFP, which is similar to the DFPase [27], while for the PON1W-Paraoxon model, Asn168 but acts as a general base to abstract the second proton from W1 for the mutant models following the corresponding transition state. Both the residues and Asn224 help the leaving group leave the whole and Asn224 keep donating hydrogen bonds to theAsn168 oxygen atom of phosphate group along through hydrogen bonds for models PON1W-Paraoxon, PON1M-Paraoxon, PON1W-DFP, and reaction path. PON1M-DFP, which is similar to the DFPase [27], while for the PON1W-Paraoxon model, Asn168 and Asn224 keep donating hydrogen bonds to the oxygen atom of phosphate group along the whole Table reaction 1. Coordination bond lengths of the optimized geometries for the Active-site-W, Active-site-M, path.

PON1W-Paraoxon, PON1M-Paraoxon, PON1W-DFP, and PON1M-DFP models. “∆” is the difference 1. Coordination bond lengths of the optimized for the Active-site-W, Active-site-M, relative toTable the Active-site-W model. Distances are in geometries angstroms.

PON1W-Paraoxon, PON1M-Paraoxon, PON1W-DFP, and PON1M-DFP models. “Δ” is the difference relative to the Active-site-W model. Distances are in angstroms. Coordination Active-Site-W Active-Site-M/∆ PON1W-Paraoxon/∆ PON1M-Paraoxon/∆’ PON1W-DFP/∆ PON1M-DFP/∆’ Bonds Coordinatio Active-Site- Active-Site-M/ PON1W-Paraoxon PON1M-Paraox PON1W-DFP/ PON1M-DFP/ Ca-OGlu53 n Bonds 2.51 Ca-OAsn168Ca-OGlu53 2.73 Ca-OAsn224Ca-OAsn1682.40 Ca-OAsp269Ca-OAsn2242.49 Ca-OAsn270Ca-OAsp2692.41 Ca-OAsn270

W 2.51 2.73 2.40 2.49 2.41

2.47/−0.04Δ 2.67/−2.47/−0.04 0.06 2.38/−2.67/−0.06 0.02 2.38/−0.02 2.51/0.02 / 2.51/0.02 /

2.45/−/Δ 0.06 2.45/−0.06 2.61/ −0.12 2.61/−0.12 2.42/0.02 2.42/0.02 2.52/0.03 2.52/0.03 2.42/0.01 2.42/0.01

on /Δ’ 2.37/ −0.14 2.37/−0.14 2.52/−0.21 2.52/−0.21 2.46/0.06 2.46/0.06 2.64/0.15 2.64/0.15 / /

Δ2.46/−0.05 Δ’ 2.39/−0.12 2.46/−0.05 2.64/−0.09 2.39/−0.122.56/−0.17 2.64/−0.09 2.43/0.03 2.56/−0.17 2.48/0.08 2.43/0.03 2.51/0.02 2.48/0.08 2.62/0.13 2.51/0.02 2.43/0.02 2.62/0.13 / 2.43/0.02 /

Figure 4. Optimized geometries of transition states and products along the reaction paths of (A) PON1W-Paraoxon, (B) PON1M-Paraoxon, (C) PON1W-DFP, and (D) PON1M-DFP models. The frozen atoms are marked with asterisks.

Molecules 2018, 23, x FOR PEER REVIEW

6 of 12

Optimized geometries of transition states and products along the reaction paths of (A) MoleculesFigure 2018, 23,4.1660

6 of 12

PON1W-Paraoxon, (B) PON1M-Paraoxon, (C) PON1W-DFP, and (D) PON1M-DFP models. The frozen atoms are marked with asterisks.

The transition states for the nucleophilic attack by the activated nucleophile W1 on the phosphorus The transition for the nucleophilic attack bysaddle the activated nucleophile W1 on the center were optimizedstates and confirmed to be the first-order point with an imaginary frequency − 1 − 1 phosphorus center were optimized and confirmed to be the first-order saddle point with ankey (99i cm for the PON1W-Paraoxon model and 108i cm for the PON1M-Paraoxon model). The −1 for the PON1W-Paraoxon model and 108i cm−1 for the imaginary frequency (99i cm bond distances of the transition states were listed in Table 2. In the mutant model PON1M-Paraoxon PON1M-Paraoxon The distances of the transition states were listed in Table 2. while In (B-TS), the distance ofmodel). Ha –OW1 is key 1.55bond Å, 0.05 Å shorter than that in the wild type model A–TS, the mutant model PON1M-Paraoxon (B-TS), the distance of H a–OW1 is 1.55 Å, 0.05 Å shorter than the distance of Hb –OW1 is 1.02 Å, 0.01 Å longer than that in A–TS. The key distance of OW1 –P is 1.66 Å, that in the wild type model A–TS, while the distance of Hb–OW1 is 1.02 Å, 0.01 Å longer than that in same as that in A-TS; while the distance of P–Ophosphate is 2.21 Å, which is 0.02 Å shorter than that A–TS. The key distance of OW1–P is 1.66 Å, same as that in A-TS; while the distance of P–Ophosphate is in A-TS. 2.21 Å, which is 0.02 Å shorter than that in A-TS.

Table 2. Important bond distances of TS in the PON1W-Paraoxon, PON1M-Paraoxon, PON1W-DFP, Table 2. Important bond distances of TS in the PON1W-Paraoxon, PON1M-Paraoxon, PON1W-DFP, and PON1M-DFP models. “∆” is the difference relative to the wild type models. Distances are and PON1M-DFP models. “Δ” is the difference relative to the wild type models. Distances are in in angstroms. angstroms.

TSTS Bonds Bonds H a -OW1 Ha -OW1 H b -OW1 Hb -OW1 OW1 -P OW1-P P-O/F P-O/F

PON1W-Paraoxon PON1M-Paraoxon/Δ PON1M-Paraoxon/∆ PON1W-DFP PON1W-DFP PON1M-DFP/Δ PON1M-DFP/∆ PON1W-Paraoxon 1.60 1.55/−0.05 1.70 1.64/−0.06 1.60 1.55/−0.05 1.70 1.64/−0.06 1.01 1.02/0.01 1.00 1.01/0.01 1.01 1.02/0.01 1.00 1.01/0.01 1.66 1.66/0.00 1.65 1.65/0.00 1.66 1.66/0.00 1.65 1.65/0.00 2.23 2.21/ − 0.02 2.24 2.19/ 2.23 2.21/−0.02 2.24 2.19/−0.05 −0.05

The energyprofiles profilesfor forthe therate-determining rate-determining step, elimination (P–O or P–F The energy step,the theleaving-group leaving-group elimination (P–O or P–F bond broken) step, were summarized shown in Figure 5. For the PON1W-Paraoxon model, the the bond broken) step, were summarized shown in Figure 5. For the PON1W-Paraoxon model, −1 (0.8 kcal mol−1 without the solvation correction). Replacing Asn270 energy barrier is 0.4 kcal mol energy barrier is 0.4 kcal mol−1 (0.8 kcal mol−1 without the solvation correction). Replacing Asn270 into W2, the energy barrier faintly increases by 0.4 kcal mol−1 both with and without the solvation into W2, the energy barrier faintly increases by 0.4 kcal mol−1 both with and without the solvation correction. Particularly, compared to the PON1W-Paraoxon product (A-Prod), the stabilities of the correction. Particularly, compared to the PON1W-Paraoxon product (A-Prod), the stabilities of the PON1M-Paraoxon product (B-Prod) increased ca. 7.0 kcal -1mol-1 relative to the corresponding PON1M-Paraoxon product (B-Prod) increased ca. 7.0 kcal mol relative to the corresponding reactant. reactant. Thus, the mutation of Asn270 has little effect on the forward reaction rate of PON Thus, the mutation of Asn270 has little effectmuch on theharder forward reaction hydrolysis, while makes the reverse reaction to take place.rate of PON hydrolysis, while

makes the reverse reaction much harder to take place. 9

-1

Relative Energies (kcal mol )

6 3 0

React a

-3

0.0(0.0)

-6

0.0(0.0)

-9

0.0(0.0)

-12

0.0(0.0)

b c d

TS a 0.4(0.8) b 0.8(1.2) c 8.7(7.9) d 8.5(7.1)

-15 -18 -21 -24 -27

Prod -17.2(-15.3)

a

b -24.1(-21.9) c 3.6(2.8) d -1.3(-2.5)

a: PON1W-Paraoxon b: PON1M-Paraoxon c: PON1W-DFP d: PON1M-DFP Reaction Coordinates

Figure 5. Potential energy profiles of the catalytic reactions for (a) PON1W-Paraoxon, (b) Figure 5. Potential energy profiles of the catalytic reactions for (a) PON1W-Paraoxon, (b) PON1M-Paraoxon, (c) PON1W-DFP, and (d) PON1M-DFP models with and without (in parenthesis) PON1M-Paraoxon, (c) PON1W-DFP, and (d) PON1M-DFP models with and without (in parenthesis) the solvation correction. the solvation correction.

Similar in the the mutant mutantmodel modelPON1M-DFP PON1M-DFP (Table 2 and Similargeometry geometrychanges changes were were observed observed in (Table 2 and Figure 4). The transition states of PON1W-DFP (C-TS) and PON1M-DFP (D-TS) models were Figure 4). The transition states of PON1W-DFP (C-TS) and PON1M-DFP (D-TS) models were optimized optimized andtoconfirmed to be the saddle first-order saddle with an imaginary frequency and confirmed be the first-order point withpoint an imaginary frequency 139i and139i 159iand cm−1 , respectively. In the PON1M-DFP model (D-TS), the distance Ha –OW1 is 1.64 Å, 0.06 Å shorter than that in the wild type model C-TS. The distance Hb –OW1 is 1.01 Å, 0.01 Å longer than that in C-TS. The key distance of OW1 –P is 1.65 Å same as that in C-TS; while the distance of P–F is 2.19 Å, which is 0.05 Å shorter than that in C-TS. In the PON1W-DFP model, the P–F bond breaking reaction has a

Molecules 2018, 23, x FOR PEER REVIEW

7 of 12

159i cm−1, respectively. In the PON1M-DFP model (D-TS), the distance Ha–OW1 is 1.64 Å, 0.06 Å Molecules 2018, 23, 1660 7 of 12 shorter than that in the wild type model C-TS. The distance Hb–OW1 is 1.01 Å, 0.01 Å longer than that

in C-TS. The key distance of OW1–P is 1.65 Å same as that in C-TS; while the distance of P–F is 2.19 Å, which is 0.05 Å shorter than that in C-TS. In the PON1W-DFP model, the P–F bond breaking reaction relatively higher energy barrier 8.7 kcal mol−1 (7.9 kcal mol−1 without the solvation correction), which −1 has a relatively higher energy barrier 8.7 kcal mol (7.9 kcal mol−1 without the solvation correction), is consistent with the larger model and basis sets results 6.0 kcal mol−1 [27]. Replacing Asn270 into which is consistent with the larger model and basis sets results 6.0 kcal mol−1 [27]. Replacing Asn270 − 1 − 1 H2 O, the energy barrier decreases 0.2 kcal mol (0.8 kcal mol without the solvation correction). into H2O, the energy barrier decreases 0.2 kcal mol−1(0.8 kcal mol−1 without the solvation correction). −1 However, thethe mutation endothermal3.6 3.6kcal kcalmol mol −1 to to However, mutationmakes makesthe theforward forwardreaction reaction change change from from endothermal −1 exothermal 1.31.3 kcal mol replacingAsn270 Asn270into intoHH essential DFP hydrolysis, 2O exothermal kcal mol−1(Figure (Figure5). 5). Thus, Thus, replacing 2O is is essential forfor DFP hydrolysis, andand provides much more driving provides much more drivingforce forcefor forthe theforward forward reaction. reaction. From the geometry results (Table 2 and Figure 4), mutantmodel modelPON1M-Paraoxon, PON1M-Paraoxon, From the geometry results (Table 2 and Figure 4), in the the mutant thethe decreased Ha H –O TSsuggest suggesta aless lessnucleophilic nucleophilic hydroxide decreased a–O distanceand andthe theincreased increased H Hbb—O –-OW1 hydroxide W1W1distance W1atatTS generated relative the wildtype typemodel. model. This This should should be increased energy generated relative to to the wild be the thereason reasonfor forthe thefaintly faintly increased energy barrier of the paraoxon hydrolysis. The two newhydrogen hydrogenbonds bondsbetween betweenAsn168, Asn168, Asn224, Asn224, and and O OLL of barrier of the paraoxon hydrolysis. The two new of the leaving group help stabilize the product of the mutant model. In the mutant model the leaving group help stabilize the product of the mutant model. In the mutant model PON1M-DFP, thedistance decreased P–Fsuggests distance a atmore TS suggests a more easily broken P–F bond relative thePON1M-DFP, decreased P–F at TS easily broken P–F bond relative to the wild to type 2− the wild model. The much stronger O makes nucleophile makes the reaction easier to take place, due model. Thetype much stronger O2− nucleophile the reaction easier to take place, due to another to another protonto HbGlu53, transferwhich to Glu53, which should also be thefor reason the increased exothermic proton Hb transfer should also be the reason the for increased exothermic heat of heat of DFP hydrolysis. DFP hydrolysis. In a word, the energy barrier of P–F bond broken in the PON1W-DFP model is higher by about In a word, the energy barrier of P–F bond broken in the PON1W-DFP model is higher by about −1 than that of the P–O bond broken in the PON1W-Paraoxon model. It suggests that the 8 kcal mol 8 kcal mol−1 than that of the P–O bond broken in the PON1W-Paraoxon model. It suggests that the nucleophile water molecule prefers the substrate Paraoxon to DFP in wild type, which agrees with nucleophile water molecule prefers the substrate Paraoxon to DFP in wild type, which agrees with the catalysis characteristics of PON1 [51,52]. Substituting water molecule for Asn270 can weakly thereduce catalysis of PON1 water and molecule for Asn270 weakly the characteristics paraoxonase activity and [51,52]. enhance Substituting the DFPase activity, especially, providecan essential reduce the paraoxonase activity and enhance the DFPase activity, and especially, provide essential thermostability for the DFP hydrolysis reaction to occur. However, such enzymatic activity changes thermostability for mutation the DFP of hydrolysis reaction totooccur. However, such activity changes mediated by the Asn270 are too faint be a decisive factor forenzymatic the enzymatic properties mediated by the are toowith faint the to beexperimental a decisive factor for the enzymatic properties of of PON1 andmutation DFPase. ofIt Asn270 is consistent conclusion that the site-specific PON1 and DFPase. It is consistent the(among experimental conclusion that the site-specific saturation mutagenesis indicated with Asn270 all active-site residues) plays a relativelysaturation minor mutagenesis indicated Asn270 (among active-site residues) plays a relatively minor role in catalysis role in catalysis for both paraoxon andallTBBL [4].

for both paraoxon and TBBL [4].

2.2. Effects of Asn270 Mutation on Enzymatic Structure

2.2. Effects of Asn270 Mutation on Enzymatic Structure

Furthermore, taking into account the whole amino acid sequences, a phylogenetic tree of PON1/2/3, DFPase, several bimetalloenzymes possessingtree Furthermore, takingand intoXC5397, account together the wholewith amino acid sequences, a phylogenetic activities, constructed as shown in Figurebimetalloenzymes 6. Among these enzymes, of phosphotriesterase PON1/2/3, DFPase, and was XC5397, together with several possessing PON1/2/3, DFPase,activities, and XC5397 adopt mononuclear active sites; hyperthermophilic phosphotriesterase was constructed as metal shown in Figure 6. Among thesearchaeon enzymes, Sulfolobus solfataricus (SsoPox) [53], phosphotriesterase (PTE) [54], organophosphorus acid PON1/2/3, DFPase, and XC5397 adopt mononuclear metal active sites; hyperthermophilic archaeon anhydrolase (OPAA) [55] and methyl parathion hydrolase (MPH) [56] all adopt binuclear metal Sulfolobus solfataricus (SsoPox) [53], phosphotriesterase (PTE) [54], organophosphorus acid anhydrolase active sites. The evolutionary relationship by the phylogenetic tree shows that XC5397 adopts a (OPAA) [55] and methyl parathion hydrolase (MPH) [56] all adopt binuclear metal active sites. mononuclear metal active site, getting close to SsoPox and PTE, which adopt binuclear metal active The evolutionary relationship by the phylogenetic tree shows that XC5397 adopts a mononuclear metal site.

active site, getting close to SsoPox and PTE, which adopt binuclear metal active site.

Figure 6. Evolutional relationship of PON1/2/3, DFPase, XC5397, SsoPox, PTE, OPAA, and MPH. The residue sequences are from PDB data or sequence date as shown in sequence names.

Structure comparisons of evolutionarily related enzymes PON1, DFPase, and XC5397 show that the residue ligand Asn270 in the active site of PON1 is mutated into Gly230 in DFPase (Gly243 in

Molecules 2018, 23, x FOR PEER REVIEW

8 of 12

Figure 6. Evolutional relationship of PON1/2/3, DFPase, XC5397, SsoPox, PTE, OPAA, and MPH. The residue sequences are from PDB data or sequence date as shown in sequence names. Molecules 2018, 23, 1660

8 of 12

Structure comparisons of evolutionarily related enzymes PON1, DFPase, and XC5397 show that the residue ligand Asn270 in the active site of PON1 is mutated into Gly230 in DFPase (Gly243 in XC5397), of which the coordination of catalytic Ca2+ ion was completed by a water molecule (W2) XC5397), of which the coordination of catalytic Ca2+ ion was completed by a water molecule (W2) 2+ ion indirectly connects the buried structural Ca2+ ion via (Figure 7). It outout that thethe catalytic CaCa 2+ ion indirectly connects the buried structural Ca2+ ion via (Figure 7).turns It turns that catalytic a series of hydrogen bondsbonds involving Asp121 in DFPase (Asp135 in XC5397) and three water a series of hydrogen involving Asp121 in DFPase (Asp135 in XC5397) and threemolecules water 2+ ions (including two coordinated water molecules). Significantly, such connections between both Ca molecules (including two coordinated water molecules). Significantly, such connections between 2+ ions are highly are both highly in DFPase and XC5397. Caconsistent consistent in DFPase and XC5397.

Figure 7. Structures of the catalytic Ca2+ ion and the buried structural Ca2+ ion in PON1 (PDB code:

Figure 7. Structures of the catalytic Ca2+ ion and the buried structural Ca2+ ion in PON1 (PDB code: 3SRE) [15] (A), DFPase (PDB code: 2GVW) [18] (B), and XC5397 (PDB code: 3DR2) [28] (C), 3SRE) [15] (A), DFPase (PDB code: 2GVW) [18] (B), and XC5397 (PDB code: 3DR2) [28] (C), respectively. respectively. These figures are drawn by the PyMOL [44]. These figures are drawn by the PyMOL [44].

Co-evolving with the mutation of Asn270, the catalytic Ca2+ ion indirectly connects the buried structural Ca2+ ion viathe a series of hydrogen bonds and water molecules DFPase Co-evolving with mutation of Asn270, theinvolving catalytic Asp121 Ca2+ ion indirectly connectsinthe buried 2+ (FigureCa 7). Together the of evolutionary relationship shown by the and phylogenetic tree (Figure 6), it structural ion via with a series hydrogen bonds involving Asp121 water molecules in DFPase implies an evolutionary transition from monoto dinuclear catalytic centers, which is also suggested (Figure 7). Together with the evolutionary relationship shown by the phylogenetic tree (Figure 6), it by Moshe Ben-Davidtransition et al. [4,15,57]. Such connections Ca2+ ions can reduce plasticity of implies an evolutionary from monoto dinuclearofcatalytic centers, whichthe is also suggested catalytic structure, and weaken the promiscuous binding of ligands and the catalytic promiscuity. It by Moshe Ben-David et al. [4,15,57]. Such connections of Ca2+ ions can reduce the plasticity of catalytic can be interpreted that PON1, with higher plasticity of catalytic structure, can accommodate structure, and weaken the promiscuous binding of ligands and the catalytic promiscuity. It can be promiscuous binding of ligands, such as relatively large substrate paraoxon, while DFPase, with interpreted that PON1, with higher plasticity of catalytic structure, can accommodate promiscuous lower plasticity of catalytic structure, prefers relatively small substrate DFP instead of binding of ligands, such as relatively large substrate paraoxon, while DFPase, with lower plasticity accommodating promiscuous substrates. It also can be concluded that Asn270 in PON1 plays a of catalytic structure, prefers relatively small substrate DFP instead of accommodating promiscuous minor role on enzyme catalytic reaction, but plays a key role on the plasticity of catalytic structure. substrates. It also can be concluded Asn270 in PON1 a minor rolesuch on enzyme catalytic This conclusion is supported by ourthat previous studies that aplays charged residue, as aspartate or reaction, but plays a key role on the plasticity of catalytic structure. This conclusion is supported glutamate in the active site, plays a much more considerable role on catalytic reaction than a neutralby our residue previous [8].studies that a charged residue, such as aspartate or glutamate in the active site, plays a

much more considerable role on catalytic reaction than a neutral residue [8]. 3. Methods

3. Methods

All calculations in this study were performed using DFT method with the B3LYP function

All calculations in this studyset were using DFThydrogen, method with the nitrogen, B3LYP function [58–60]. [58–60]. The 6-31G(d, p) basis wasperformed used for the carbon, oxygen, and calcium Theatoms, 6-31G(d, basis set was for the carbon, hydrogen, oxygen, calcium atoms, the thep)6-311+G(d) basisused set was used for phosphorous atom. Atnitrogen, the sameand level, the frequency calculations were to obtain zero-point energies to confirm the nature of the 6-311+G(d) basis set performed was used for phosphorous atom. At the(ZPE) sameand level, the frequency calculations stationary points transition statesenergies along the(ZPE) reaction profiles. Intrinsic (IRC) were performed to and obtain zero-point and to confirm thereaction naturecoordinate of the stationary calculations were also explored to make sure that the transition state connects the corresponding points and transition states along the reaction profiles. Intrinsic reaction coordinate (IRC) calculations reactant and the product. were also explored to make sure that the transition state connects the corresponding reactant and The polarization effects of the enzyme environment were evaluated by performing single-point the product. calculations on the effects optimized structures the same theory level as thebygeometry optimizations The polarization of the enzyme at environment were evaluated performing single-point using the conductor-like polarizable continuum model (CPCM) method [61–64]. The dielectric calculations on the optimized structures at the same theory level as the geometry optimizations using constants were set to four and eighty, which are usually used in modeling protein surroundings and the conductor-like polarizable continuum model (CPCM) method [61–64]. The dielectric constants were set to four and eighty, which are usually used in modeling protein surroundings and the aqueous solution, respectively. The barrier discrepancies are only 0.1 kcal mol−1 with the constants 4 and 80; here, we just present the results with ε = 4. All calculations were performed with the Gaussian 03 program package [65].

Molecules 2018, 23, 1660

9 of 12

The topological tree was constructed using Neighbour-Joining in MEGA 6.06 [66]. Sequence alignment was performed with Clustal W [67]. 4. Conclusions In the present work, we have studied the role of Asn270 played in PON1 on hydrolyzing paraoxon and DFP using the popular density functional method B3LYP. The wild type and mutant models were designed, based on the difference of the active sites between PON1 and DFPase (the first shell residue Asn270 replaced by a water molecule). The results suggests the mutation of Asn270 has little effect on the enzyme activity in kinetics, while provides more efficient in thermodynamics, which is essential for DFP hydrolysis. Furthermore, the structures of evolutionarily related enzymes PON1, DFPase, and XC5397 were compared, which proposes the mutation of Asn270 leads to that the catalytic Ca2+ ion indirectly connects the buried structural Ca2+ ion via hydrogen bonds in DFPase. Such connections between Ca2+ ions can reduce the plasticity of enzymatic structure, and possibly alter the substrate preference from paraoxon to DFP, which also implys an evolutionary transition from mono- to dinuclear catalytic centers. Author Contributions: H.Z. and L.Y. conceived and designed the computational models; L.Y. performed most of the calculations; Y.-Y.M. and C.Z. contributed to conceptualization and methodology; R.-Z.L. and S.L. performed the validation; H.Z. and L.Y. wrote the paper, which was improved by the rest of authors. Funding: This research was supported by grants from the National Natural Science Foundation of China (grant nos. 21203042, 21303009, 21503083), the Natural Science Foundation of Gansu Province (grant no.145RJYA285), and the Fundamental Research Funds for the Central Universities (grant no. HIT. NSRIF. 2013057 and grant no. 31920150027). Acknowledgments: We appreciate Sven de Marothy (from Stockholm University) for providing xyzviewer to create all the figures of the molecule models and the dispersion correction calculations. Conflicts of Interest: The authors declare no conflict of interest.

References 1.

2.

3.

4.

5. 6.

7.

Zalatan, J.G.; Fenn, T.D.; Herschlag, D. Comparative enzymology in the alkaline phosphatase superfamily to determine the catalytic role of an active-site metal ion. J. Mol. Biol. 2008, 384, 1174–1189. [CrossRef] [PubMed] Huang, H.; Patskovsky, Y.; Toro, R.; Farelli, J.D.; Pandya, C.; Almo, S.C.; Allen, K.N.; Dunaway-Mariano, D. Divergence of structure and function in the haloacid dehalogenase enzyme superfamily: Bacteroides thetaiotaomicron BT2127 is an inorganic pyrophosphatase. Biochemistry 2011, 50, 8937–8949. [CrossRef] [PubMed] López-Canut, V.; Roca, M.; Bertrán, J.; Moliner, V.; Tuñón, I. Promiscuity in alkaline phosphatase superfamily. Unraveling evolution through molecular simulations. J. Am. Chem. Soc. 2011, 133, 12050–12062. [CrossRef] [PubMed] Ben-David, M.; Wieczorek, G.; Elias, M.; Silman, I.; Sussman, J.L.; Tawfik, D.S. Catalytic metal ion rearrangements underline promiscuity and evolvability of a metalloenzyme. J. Mol. Biol. 2013, 425, 1028–1038. [CrossRef] [PubMed] Brown, S.D.; Babbitt, P.C. New insights about enzyme evolution from large scale studies of sequence and structure relationships. J. Biol. Chem. 2014, 289, 30221–30228. [CrossRef] [PubMed] Bora, R.P.; Mills, M.J.; Frushicheva, M.P.; Warshel, A. On the challenge of exploring the evolutionary trajectory from phosphotriesterase to arylesterase using computer simulations. J. Phys. Chem. B 2015, 119, 3434–3445. [CrossRef] [PubMed] Sunden, F.; AlSadhan, I.; Lyubimov, A.Y.; Ressl, S.; Wiersma-Koch, H.; Borland, J.; Brown, C.L.; Johnson, T.A.; Singh, Z.; Herschlag, D. Mechanistic and evolutionary insights from comparative enzymology of phosphomonoesterases and phosphodiesterases across the alkaline phosphatase superfamily. J. Am. Chem. Soc. 2016, 138, 14273–14287. [CrossRef] [PubMed]

Molecules 2018, 23, 1660

8. 9. 10. 11. 12. 13. 14. 15.

16.

17. 18.

19.

20.

21.

22.

23. 24. 25. 26.

27.

28. 29.

10 of 12

Zhang, H.; Yang, L.; Yan, L.-F.; Liao, R.-Z.; Tian, W.-Q. Evolution of phosphotriesterase activities of the metallo-β-lactamase family: A theoretical study. J. Inorg. Biochem. 2018, 184, 8–14. [CrossRef] [PubMed] Aubert, S.D.; Li, Y.; Raushel, F.M. Mechanism for the hydrolysis of organophosphates by the bacterial phosphotriesterase. Biochemistry 2004, 43, 5707–5715. [CrossRef] [PubMed] Wong, K.-Y.; Gao, J. The reaction mechanism of paraoxon hydrolysis by phosphotriesterase from combined QM/MM simulations. Biochemistry 2007, 46, 13352–13369. [CrossRef] [PubMed] Chen, S.; Fang, W.; Himo, F. Theoretical study of the phosphotriesterase reaction mechanism. J. Phys. Chem. B 2007, 111, 1253–1255. [CrossRef] [PubMed] Gonzalvo, M.C.; Gil, F.; Hernandez, A.F.; Rodrigo, L.; Villanueva, E.; Pla, A. Human liver paraoxonase (PON1): Subcellular distribution and characterization. J. Biochem. Mol. Toxicol. 1998, 12, 61–69. [CrossRef] Gaidukov, L.; Tawfik, D.S. High affinity, stability, and lactonase activity of serum paraoxonase PON1 Anchored on HDL with ApoA-I. Biochemistry 2005, 44, 11843–11854. [CrossRef] [PubMed] Aharoni, A.; Gaidukov, L.; Khersonsky, O.; Gould, S.M.; Roodveldt, C.; Tawfik, D.S. The ‘evolvability’ of promiscuous protein functions. Nat. Genet. 2005, 37, 73–76. [CrossRef] [PubMed] Ben-David, M.; Elias, M.; Filippi, J.-J.; Duñach, E.; Silman, I.; Sussman, J.L.; Tawfik, D.S. Catalytic versatility and backups in enzyme active sites: The case of serum paraoxonase 1. J. Mol. Biol. 2012, 418, 181–196. [CrossRef] [PubMed] Harel, M.; Aharoni, A.; Gaidukov, L.; Brumshtein, B.; Khersonsky, O.; Meged, R.; Dvir, H.; Ravelli, R.B.G.; McCarthy, A.; Toker, L.; et al. Structure and evolution of the serum paraoxonase family of detoxifying and anti-atherosclerotic enzymes. Nat. Struct. Mol. Biol. 2004, 11, 412–419. [CrossRef] [PubMed] Scharff, E.I.; Koepke, J.; Fritzsch, G.; Lücke, C.; Rüterjans, H. Crystal structure of diisopropylfluorophosphatase from Loligo vulgaris. Structure 2001, 9, 493–502. [CrossRef] Blum, M.-M.; Löhr, F.; Richardt, A.; Rüterjans, H.; Chen, J.C.H. Binding of a designed substrate analogue to diisopropyl fluorophosphatase: Implications for the phosphotriesterase mechanism. J. Am. Chem. Soc. 2006, 128, 12750–12757. [CrossRef] [PubMed] Blum, M.-M.; Chen, J.C.H. Structural characterization of the catalytic calcium-binding site in diisopropyl fluorophosphatase (DFPase)—Comparison with related β-propeller enzymes. Chem. Biol. Interact. 2010, 187, 373–379. [CrossRef] [PubMed] Chen, J.C.-H.; Mustyakimov, M.; Schoenborn, B.P.; Langan, P.; Blum, M.-M. Neutron structure and mechanistic studies of diisopropyl fluorophosphatase (DFPase). Acta Crystallogr. D 2010, 66, 1131–1138. [CrossRef] [PubMed] Blum, M.-M.; Timperley, C.M.; Williams, G.R.; Thiermann, H.; Worek, F. Inhibitory potency against human acetylcholinesterase and enzymatic hydrolysis of fluorogenic nerve agent mimics by human paraoxonase 1 and squid diisopropyl fluorophosphatase. Biochemistry 2008, 47, 5216–5224. [CrossRef] [PubMed] Hartleib, J.; Ruterjans, H. High-yield expression, purification, and characterization of the recombinant diisopropylfluorophosphatase from Loligo vulgaris. Protein Expr. Purif. 2001, 21, 210–219. [CrossRef] [PubMed] Bigley, A.N.; Raushel, F.M. Catalytic mechanisms for phosphotriesterases. BBA Proteins Proteom. 2013, 1834, 443–453. [CrossRef] [PubMed] Khersonsky, O.; Tawfik, D.S. Structure–Reactivity studies of serum paraoxonase PON1 Suggest that Its native activity is Lactonase. Biochemistry 2005, 44, 6371–6382. [CrossRef] [PubMed] 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. [CrossRef] [PubMed] Blum, M.-M.; Mustyakimov, M.; Rüterjans, H.; Kehe, K.; Schoenborn, B.P.; Langan, P.; Chen, J.C.-H. Rapid determination of hydrogen positions and protonation states of diisopropyl fluorophosphatase by joint neutron and X-ray diffraction refinement. Proc. Natl. Acad. Sci. USA 2009, 106, 713–718. [CrossRef] [PubMed] Xu, C.; Yang, L.; Yu, J.-G.; Liao, R.-Z. What roles do the residue Asp229 and the coordination variation of calcium play of the reaction mechanism of the diisopropyl-fluorophosphatase? A DFT investigation. Theor. Chem. Acc. 2016, 135, 138. [CrossRef] Chen, C.-N.; Chin, K.-H.; Wang, A.H.J.; Chou, S.-H. The first crystal structure of gluconolactonase important in the glucose secondary metabolic pathways. J. Mol. Biol. 2008, 384, 604–614. [CrossRef] [PubMed] Lichtarge, O.; Bourne, H.R.; Cohen, F.E. An evolutionary trace method defines binding surfaces common to protein families. J. Mol. Biol. 1996, 257, 342–358. [CrossRef] [PubMed]

Molecules 2018, 23, 1660

30. 31. 32. 33.

34. 35.

36. 37. 38. 39. 40.

41.

42. 43. 44. 45. 46. 47. 48. 49. 50.

51. 52.

11 of 12

Gerlt, J.A.; Babbitt, P.C. Divergent evolution of enzymatic function: Mechanistically diverse superfamilies and functionally distinct suprafamilies. Annu. Rev. Biochem. 2001, 70, 209–246. [CrossRef] [PubMed] O’Brien, P.J.; Herschlag, D. Catalytic promiscuity and the evolution of new enzymatic activities. Chem. Biol. 1999, 6, R91–R105. [CrossRef] Sousa, S.F.; Fernandes, P.A.; Ramos, M.J. Unraveling the mechanism of the farnesyltransferase enzyme. J. Biol. Inorg. Chem. 2005, 10, 3–10. [CrossRef] [PubMed] Sousa, S.F.; Fernandes, P.A.; Ramos, M.J. Farnesyltransferase—New insights into the zinc-coordination sphere paradigm: Evidence for a carboxylate-shift mechanism. Biophys. J. 2005, 88, 483–494. [CrossRef] [PubMed] Sousa, S.F.; Fernandes, P.A.; Ramos, M.J. Farnesyltransferase: Theoretical studies on peptide substrate entrance—Thiol or thiolate coordination? J. Mol. Struct.-THEOCHEM 2005, 729, 125–129. [CrossRef] Sousa, S.F.; Fernandes, P.A.; Ramos, M.J. Theoretical studies on farnesyltransferase: Evidence for thioether product coordination to the active-site zinc sphere. J. Comput. Chem. 2007, 28, 1160–1168. [CrossRef] [PubMed] Himo, F.; Siegbahn, P.E.M. Quantum chemical studies of radical-containing enzymes. Chem. Rev. 2003, 103, 2421–2456. [CrossRef] [PubMed] Siegbahn, P.E.M.; Borowski, T. Modeling enzymatic reactions involving transition metals. Acc. Chem. Res. 2006, 39, 729–738. [CrossRef] [PubMed] Siegbahn, P.E.M. The effect of backbone constraints: The case of water oxidation by the oxygen-evolving complex in PSII. ChemPhysChem 2011, 12, 3274–3280. [CrossRef] [PubMed] Siegbahn, P.E.M.; Himo, F. The quantum chemical cluster approach for modeling enzyme reactions. WIREs Comput. Mol. Sci. 2011, 1, 323–336. [CrossRef] Leopoldini, M.; Russo, N.; Toscano, M. Which one among Zn(II), Co.(II), Mn(II), and Fe(II) is the most efficient ion for the methionine aminopeptidase catalyzed reaction? J. Am. Chem. Soc. 2007, 129, 7776–7784. [CrossRef] [PubMed] Abashkin, Y.G.; Burt, S.K.; Collins, J.R.; Cachau, R.E.; Russo, N.; Erickson, J.W. Metal-Ligand Interactions: Structure and Reactivity; Russo, N., Salahub, D.R., Eds.; Nato Science Series; Kluwer: Dordrecht, The Netherlands, 1996; Volume 474, pp. 1–22. Marino, T.; Russo, N.; Toscano, M. A comparative study of the catalytic mechanisms of the zinc and cadmium containing carbonic anhydrase. J. Am. Chem. Soc. 2005, 127, 4242–4253. [CrossRef] [PubMed] Leopoldini, M.; Russo, N.; Toscano, M. Role of the metal ion in formyl-peptide bond hydrolysis by a peptide deformylase active site model. J. Phys. Chem. B 2006, 110, 1063–1072. [CrossRef] [PubMed] DeLano, W.L. The PyMOL Molecular Graphics System; DeLano Scientific: Palo Alto, CA, USA, 2002. Allen, K.N.; Dunaway-Mariano, D. Phosphoryl group transfer: Evolution of a catalytic scaffold. Trends Biochem. Sci. 2004, 29, 495–503. [CrossRef] [PubMed] Valdez, C.E.; Smith, Q.A.; Nechay, M.R.; Alexandrova, A.N. Mysteries of metals in metalloenzymes. Acc. Chem. Res. 2014, 47, 3110–3117. [CrossRef] [PubMed] Blomberg, M.R.A.; Borowski, T.; Himo, F.; Liao, R.-Z.; Siegbahn, P.E.M. Quantum chemical studies of mechanisms for metalloenzymes. Chem. Rev. 2014, 114, 3601–3658. [CrossRef] [PubMed] Yang, L.; Liao, R.-Z.; Ding, W.-J.; Liu, K.; Yu, J.-G.; Liu, R.-Z. Why calcium inhibits magnesium-dependent enzyme phosphoserine phosphatase? A theoretical study. Theor. Chem. Acc. 2012, 131, 1275. [CrossRef] Salter, E.A.; Honkanen, R.E.; Wierzbicki, A. Modeling the antiferromagnetic MnIIMnII system within the protein phosphatase-5 catalytic site. J. Mol. Model. 2015, 21, 14. [CrossRef] [PubMed] Wymore, T.; Field, M.J.; Langan, P.; Smith, J.C.; Parks, J.M. Hydrolysis of DFP and the Nerve Agent (S)-Sarin by DFPase proceeds along two different reaction pathways: Implications for engineering bioscavengers. J. Phys. Chem. B 2014, 118, 4479–4489. [CrossRef] [PubMed] Rochu, D.; Chabrière, E.; Masson, P. Human paraoxonase: A promising approach for pre-treatment and therapy of organophosphorus poisoning. Toxicology 2007, 233, 47–59. [CrossRef] [PubMed] Stevens, R.C.; Suzuki, S.M.; Cole, T.B.; Park, S.S.; Richter, R.J.; Furlong, C.E. Engineered recombinant human paraoxonase 1 (rHuPON1) purified from Escherichia coli protects against organophosphate poisoning. Proc. Natl. Acad. Sci. USA 2008, 105, 12780–12784. [CrossRef] [PubMed]

Molecules 2018, 23, 1660

53.

54.

55.

56.

57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67.

12 of 12

Elias, M.; Dupuy, J.; Merone, L.; Mandrich, L.; Porzio, E.; Moniot, S.; Rochu, D.; Lecomte, C.; Rossi, M.; Masson, P.; et al. Structural basis for natural lactonase and promiscuous phosphotriesterase activities. J. Mol. Biol. 2008, 379, 1017–1028. [CrossRef] [PubMed] Vanhooke, J.L.; Benning, M.M.; Raushel, F.M.; Holden, H.M. Three-Dimensional Structure of the Zinc-Containing Phosphotriesterase with the Bound Substrate Analog Diethyl 4-Methylbenzylphosphonate. Biochemistry 1996, 35, 6020–6025. [CrossRef] [PubMed] Vyas, N.K.; Nickitenko, A.; Rastogi, V.K.; Shah, S.S.; Quiocho, F.A. Structural insights into the dual activities of the nerve agent degrading organophosphate Anhydrolase/Prolidase. Biochemistry 2009, 49, 547–559. [CrossRef] [PubMed] Dong, Y.-J.; Bartlam, M.; Sun, L.; Zhou, Y.-F.; Zhang, Z.-P.; Zhang, C.-G.; Rao, Z.; Zhang, X.-E. Crystal structure of methyl parathion hydrolase from Pseudomonas sp. WBC-3. J. Mol. Biol. 2005, 353, 655–663. [CrossRef] [PubMed] Heinz, U.; Adolph, H.W. Metallo-β-lactamases: Two binding sites for one catalytic metal ion? Cell. Mol. Life Sci. 2004, 61, 2827–2839. [CrossRef] [PubMed] Lee, C.; Yang, W.; Parr, R.G. Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B 1988, 37, 785–789. [CrossRef] Becke, A.D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98, 5648–5652. [CrossRef] Becke, A.D. A new mixing of Hartree–Fock and local density-functional theories. J. Chem. Phys. 1993, 98, 1372–1377. [CrossRef] Barone, V.; Cossi, M. Quantum calculation of molecular energies and energy gradients in solution by a conductor solvent model. J. Phys. Chem. A 1998, 102, 1995–2001. [CrossRef] Cammi, R.; Mennucci, B.; Tomasi, J. Second-order Møller–Plesset analytical derivatives for the polarizable continuum model using the relaxed density approach. J. Phys. Chem. A 1999, 103, 9100–9108. [CrossRef] Klamt, A.; Schuurmann, G. COSMO: A new approach to dielectric screening in solvents with explicit expressions for the screening energy and its gradient. J. Chem. Soc. Perkin Trans. 1993, 2, 799–805. [CrossRef] Tomasi, J.; Mennucci, B.; Cammi, R. Quantum mechanical continuum solvation models. Chem. Rev. 2005, 105, 2999–3094. [CrossRef] [PubMed] Frisch, M.; Trucks, G.; Schlegel, H.; Scuseria, G.; Robb, M.; Cheeseman, J.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. Gaussian 09, Revision A. 02; Gaussian, Inc.: Wallingford, CT, USA, 2009. Tamura, K.; Stecher, G.; Peterson, D.; Filipski, A.; Kumar, S. MEGA6: Molecular evolutionary genetics analysis version 6.0. Mol. Biol. Evol. 2013, 30, 2725–2729. [CrossRef] [PubMed] Thompson, J.D.; Higgins, D.G.; Gibson, T.J. CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994, 22, 4673–4680. [CrossRef] [PubMed]

Sample Availability: Samples of the compounds that the optimized geometries along the reaction paths are available from the authors. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).