The Role of Nonbonded Interactions in the Conformational Dynamics ofOrganophosphorous Hydrolase Adsorbed onto Functionalized Mesoporous Silica Surfaces

Diego E. B. Gomes,†,‡ Roberto D. Lins,†,§ Pedro G. Pascutti,‡ Chenghong Lei,† andThereza A. Soares*,†,§

Pacific Northwest National Laboratory, P.O. Box 999, MSIN K7-90, Richland, Washington 99352, Instituto deBiofısica Carlos Chagas Filho, UniVersidade Federal do Rio de Janeiro, Rio de Janeiro, RJ 21949-900, Brazil,and Departamento de Quımica Fundamental, UniVersidade Federal de Pernambuco, Recife, PE, 50590-470, Brazil

ReceiVed: August 30, 2009; ReVised Manuscript ReceiVed: October 12, 2009

The enzyme organophosphorous hydrolase (OPH) catalyzes the hydrolysis of a wide variety of organophos-phorous compounds with high catalytic efficiency and broad substrate specificity. The immobilization ofOPH in functionalized mesoporous silica (FMS) surfaces increases significantly its catalytic specific activity,as compared to the enzyme in solution, with important applications for the detection and decontamination ofinsecticides and chemical warfare agents. Experimental measurements of immobilization efficiency as a functionof the charge and coverage percentage of different functional groups have been interpreted as electrostaticforces being the predominant interactions underlying the adsorption of OPH onto FMS surfaces. Explicitsolvent molecular dynamics simulations have been performed for OPH in bulk solution and adsorbed ontotwo distinct interaction potential models of the FMS functional groups to investigate the relative contributionsof nonbonded interactions to the conformational dynamics and adsorption of the protein. Our results supportthe conclusion that electrostatic interactions are responsible for the binding of OPH to the FMS surface.However, these results also show that van der Waals forces are detrimental for interfacial adhesion. In addition,it is found that OPH adsorption onto the FMS models favors a protein conformation whose active site is fullyaccessible to the substrate, in contrast to the unconfined protein.

Introduction

The development of new materials has enabled the use ofporous solid support for protein immobilization via encapsula-tion, entrapment, and covalent linking with useful applicationsin biotechnology, biocatalysis, protein-delivery systems, andsensing.1-8 Immobilization on a solid support can enhanceenzyme stability while maintaining near-native activity andselectivity.9,10 A range of supports has been described forproteins, each having associated advantages and disadvantagesfor use in protein biocatalysis.11 Functionalized mesoporoussilica (FMS) support has attracted considerable interest as anenzyme-immobilization matrix due to the large, rigid, anduniform mesopore structure and favorable interaction environ-ment provided by tunable functional groups.11-14 These featuresallow much higher protein loading densities and enhancedenzyme activity and stability as compared with conventionalmesoporous silica supports.

It has been previously reported by Lei and co-workers thatthe immobilization of the metalloenzyme organophosphoroushydrolase (OPH) in FMS enhances its stability and increasesits catalytic specific activity by 200% as compared to the enzymein solution.15 Immobilization of other enzymes into FMSsupports corroborates these findings and further demonstratesthat FMS with appropriate pore sizes and functional groupscould result in faster kinetics, greatly enhanced enzymaticactivity, and stability for target enzymes as compared withconventional approaches of enzyme immobilization.16,17 The

enzyme organophosphorous hydrolase (OPH; EC 3.1.8.1) fromthe soil bacteria BreVundimonas diminuta (previously Pseudomo-nas diminuta) catalyzes the hydrolysis of a wide variety ofphosphotriester (P-O), phosphonothioate (P-S), phosphofluo-ridate (P-F), and phosphonocyanate (P-CN) compounds withhigh catalytic efficiency and broad substrate specificity.18-24

Although the biological role of OPH remains unclear, it isthought to have evolved in the last few decades, since phos-photriesters are not naturally occurring compounds and werefirst synthesized during the Second World War.25,26 Hydrolysisof its most efficient substrate, paraoxon, is remarkably efficient,with kcat and kcat/KM values for paraoxon (104 s-1 and 107 s-1

M-1, respectively) that approach the limit of diffusion of thereactants in solution.18 Modifications of the active site wereshown to alter the substrate specificity and enhance the catalyticefficiency for slower substrates.19,20,27,28 Variants of OPH havealso been shown to be stereoselective for the hydrolysis of chiralorganophosphate triesters.29 This is a highly desirable catalystfeature, since major chemical warfare agents are racemicmixtures with substantial differences in toxicity of the individualenantiomers. Therefore, the immobilization of OPH in FMSsupport offers great potential for the development of biosensorsfor bioremediation and detoxification of toxic pesticides andnerve agents such as soman, sarin, and VX.30,31

The maximum amount of enzyme adsorbed on a mesoporousmaterial varies with pH value, ionic strength, pore diameter,pore volume, surface characteristics, isoelectric point (pI),morphology, and particle size.32-34 Protein adsorption can bemanipulated via different types of interactions: electrostaticforces, hydrogen bonding, van der Waals forces, conformationalentropy, and hydrophobic interactions.10,11 Moreover, surfaceadsorption can be increased by the gain of conformational

* To whom correspondence should be addressed. E-mail: tasoares@pnl.gov.† Pacific Northwest National Laboratory.‡ Universidade Federal do Rio de Janeiro.§ CCEN, UFPE.

J. Phys. Chem. B 2010, 114, 531–540 531

10.1021/jp9083635 ! 2010 American Chemical SocietyPublished on Web 11/25/2009

entropy upon unfolding at the surface, shift of pH close to theisoelectric point, and specific ion binding.35 In addition, thenature of the solid surface, including its hydrophobicity andcharge density, has a strong effect on the structure andconformation of the protein layer.1,10,11 Surface modification ofmesoporous silica with functional groups that are compatibleto those found in the surface of proteins can enhance theseinteractions and influence the final loading and activity of theimmobilized protein.36,37 Hydrophobic and ionic interactions arecommonly the most important forces driving protein adsorptionto mineral surfaces, but the relative significance of theseinteractions in a given system depends on the details of theprotein structure and the particular surface involved.12,13,34,38-42

However, it is difficult to assess the exact contribution of thedifferent interfacial interactions to the adsorption of proteinsonto the FMS surface at the microscopic level. This is in partbecause of the amorphous nature of the SiO2 walls, which makesit very difficult to extract structural information at atomicresolution from experimental data only,8,12,14 yet the molecularcharacterization of such interfacial interactions is an essentialstep for the rational design of FMS surface properties that willlead to the efficient immobilization of enzymes for environ-mental and industrial applications.

It has been shown that the adsorption of OPH (pI ) 8.3) toFMS functionalized with negative -CH2-CH2-COO- groupsat low coverage (2%) and pH 7.5 results in optimal affinity andimmobilization efficiency, whereas its adsorption to FMSfunctionalized with positive -CH2-CH2-NH3

+ groups at highcoverage (20%) leads to significantly lower immobilizationefficiency.15 OPH was also shown to adsorb to unfunctionalizedmesoporous silica whose silanol surface is negatively charged,although the entrapped enzyme exhibited low specific enzymaticactivity.15 From these results, electrostatic interactions have beenproposed to be the predominant forces governing the adsorptionof OPH onto FMS surfaces. However, such a hypothesis is notfully compatible with the observation that OPH adsorbs to FMSfunctionalized with a low coverage (2%) of positively charged-CH2-CH2-NH3

+ groups, whereas an increase in the coverageof negatively charged CH2-CH2-COO- groups from 2 to 20%leads a sharp decrease in the immobilization efficiency.15 Theseresults appear to indicate that other interactions also play animportant role in modulating OPH adsorption to FMS surfaces.Indeed, previous immobilization studies for different enzymeshave suggested that hydrophobic interactions play a moreimportant role in protein-FMS surface adsorption than it hasbeen assumed.13,36,43 For instance, it was demonstrated that theimmobilization of penicillin acylase (pI ) 7.0) onto mesoporoussilica surfaces (SBA-15) functionalized with the hydrophobicgroups phenyltrimethoxysilane and vinyltriethoxysilane resultedin enhanced adsorption rates and adsorption capacity ascompared to the pure SBA-15, which is negatively charged.36

In this context, molecular dynamics (MD) simulations canprovide a microscopic level characterization of the role ofdifferent interactions in the adsorption of proteins to FMSsurfaces.44-46 As a basis for the understanding of the nature ofthese interactions in the adsorption and confinement of OPHinto FMS support, simulations have been performed for theenzyme OPH free in solution as well as adsorbed to two coarse-grained models of the FMS functional carboxyl groups. In onemodel, van der Waals interactions are made negligible byrepresenting the functional group as an oxyanion atom. In thesecond model, the van der Waals parameters for the ethyl acetatefunctional groups were used to fully account for these contribu-tions. To reproduce the spatial arrangement of functional groups

on the silicate wall, the FMS surface is represented as a cylinderwhose curvature and functional group coverage was estimatedfrom experimental measurements.47 Each sphere representinga functional group is spaced within 1 nm from each other andratio protein/functional groups of 1:57. The effect of the differentrepresentations of the FMS surface on the structural dynamicsand interfacial diffusion properties of OPH is investigated. Theseproperties are characterized through time-dependent analysis ofatom-positional root-mean-square deviation (rmsd) and root-mean-square atom-positional fluctuations (RMSF) from theX-ray structure of OPH, mean-square displacement (MSD) andinteratomic distances of OPH with respect to the FMS surfaces,solvent-accessible surface, and secondary structure content.Principal component analysis of the MD simulations have beenundertaken to investigate the mechanisms by which FMSconfinement steers local and global motions of OPH towardconformations of enhanced stability and catalytic competence.

Computational Methodology

OPH Model. High-resolution crystallographic structures ofOPH from BreVundimonas diminuta were used as initialcoordinates in the MD simulations (PDB ID 1EZ2 and1HZY).48,49 The substrate analogue diisopropylmethyl phos-phonate was removed from the X-ray structure 1EZ2 removedprior to simulations. Atom positional rmsd between the twocrystallographic structures was 0.02 nm and demonstrates thatthe two conformations are nearly indistinguishable. Missingatoms in the crystallographic structures were verified and addedif necessary with the WHAT IF web server.50 Atom additionswere necessary only to the terminal residues. Protonation stateswere assigned accordingly to pKa calculations with the programpropKa 2.0.51 All simulations were carried out using theGROMOS force field force parameter set 53A6.52,53 The activesite of OPH contains two zinc ions bridged by a hydroxide anionand a carbamylated lysine. The two zinc and hydroxide ionswere treated as a nonbonded model. The entire subsystempresents a formal charge of +3e. Charges for the hydroxideion and carbamylated lysine were assigned via a restrainedhyperbolic fit of the electrostatic potential54 on the nucleipositions of each atom after geometry optimizations at the HF/6-31G* level using the NWCHEM software55 as described inref 56. This procedure has been shown to be fully compatiblewith the parameter set 53A6 of the GROMOS force field.57-59

FMS Model. The interactions between OPH and the FMSsurface have two components: steric due to confinement natureof the mesoporous surface and nonbonded due to the function-alization of the mesopore. To study the confinement, we assumethat these interactions can be approximated by a coarserepresentation modeled as a cylindrical, uniform array ofparticles, each corresponding to a given functional group andspaced apart to represent the percentage of functional groupcoverage.15,47 In this coarse-grained model, each particle incor-porates the atomic attributes (van der Waals and Coulombparameters) characteristic of the given functional group derivedfrom a molecular mechanics force field. The coarse-grainrepresentation of the functional groups allows for a decrease inthe degrees of freedom of the system and, consequently, longersampling of the protein configurational space. To evaluate therelative contributions of electrostatics and vdW interactions tothe adsorption of the OPH onto the FMS, two models weredeveloped to represent the negative functional groups commonlyused in experimental studies of OPH immobilization onto FMS.In the model FMScoul, the pore surface was represented by point-charged particles with charge -1e and vdW parameters corre-

532 J. Phys. Chem. B, Vol. 114, No. 1, 2010 Gomes et al.

sponding to an oxyanion. In the model FMSvdW, the pore surfacewas represented by particles with charge -1e and vdWparameters corresponding to acetate groups from the Martiniforce field atom type P3.60 Structure-based characterization ofthe electrostatic potential surface of OPH demonstrates that thesurface opposed to the active site exhibits a very positiveelectrostatic potential, whereas the surface containing theentrance to the active site exhibits a mostly negative electrostaticpotential.17 From the charge distribution along the OPH surface,

it has been proposed that the positively charged surface of OPHis attracted via electrostatic interactions to the negatively chargedsurface of FMS, leaving the entrance to the active sites exposedto the lumen of the mesoporous and freely accessible tosubstrates and products.17 In the present work, the OPH structurewas docked to the FMS pore wall on the basis of thecomplementarity of their electrostatic potential surfaces calculatedwith the program APBS as previously described in ref 17. Thesystem is illustrated in Figure 1 and described in Table 1.

Figure 1. Cartoon representation of the organophosphorous hydrolase structure adsorbed onto the functionalized mesoporous silica surface. (A)Top view of the OPH dimer. Regions of large atomic fluctuations in the MD simulations are indicated in violet (L1, residues 155-165), blue (L2,residues 172-182), yellow (L3, residues 201-215), orange (L4, residues 230-240), purple (L5, residues 255-275), and green (L6, residues306-325) for only one monomer. (B) Close-up view of the active site pocket with residues represented in licorice and Zn2+ cations in theCorey-Pauling-Koltun (CPK) model (yellow). The residue Y309 in the gateway to the active site pocket is highlighted. (C) Side view of the OPHdimer. Positively charged residues in the protein that interact with the functionalized mesoporous silica are represented in licorice. The functionalizedmesoporous silica is represented in CPK (red). The explicit model water molecules are omitted for clarity of visualization.

Dynamics of OPH Adsorbed onto FMS Surfaces J. Phys. Chem. B, Vol. 114, No. 1, 2010 533

Simulation Setup. The systems were placed in a rectangularbox, treated for periodic boundary conditions, and solvated withexplicit SPC model water molecules.61 The systems wereneutralized with Na+ counterions where necessary. Simulationswere carried out in the NPT ensemble, and a time step of 2 fswas used to integrate the equations of motion based on the LeapFrog algorithm.62 The temperature of the solute and solvent wereseparately coupled to the velocity-rescale thermostat at 298.15K with a relaxation time of 0.1 ps. The pressure was maintainedas 1 atm by isotropic coordinate scaling with a relaxation of 1ps. The bond lengths and angles were constrained by using theP-LINCS algorithm,63 and the geometry of the water moleculeswas constrained using the SETTLE algorithm.64 A twin-rangecutoff of 1.0 and 1.2 nm was used for vdW interactions, andlong-range electrostatic interactions were treated by the particlemesh Ewald method.65 The systems were initially minimizedthrough 20 000 iterations of the steepest descent algorithm.Solvent molecules were relaxed during 500 ps at 298.15 K withpositional restraints applied to the heavy atoms of the protein.The full system was equilibrated for 10 ns, followed by theproduction phase of 40 ns. Configurations of the system wererecorded as trajectory files every 1.0 ps. The software packageGROMACS v.4.04 was used for the simulations and propertyanalyses compiled in double precision.66 Protein structures werevisualized with the software VMD 1.86.67 The interface betweenmonomers was computed using our in-house software Surf-inMD. Calculated properties such as atom positional rmsd andfluctuation (RMSF), principal component analysis (PCA), time-dependent secondary structure pattern, solvent number density,radial distribution function, and solvent-accessible surface area(SASA) were computed using the implemented algorithms inthe GROMACS package.66

Results

Structural Characterization of OPH in Bulk Water andAdsorbed onto FMS Surfaces. The atom positional rmsd wascalculated from the MD simulations OPHfree, FMSCoul, andFMSvdW with respect to the substrate-bound conformation ofthe X-ray structure (PDB ID 1EZ2 and 1HZY)48,49 (Figure 2).The rmsd from the three simulations converged to a plateau of0.15 nm during the first 15 ns of simulations, indicating thatOPH initially adopted a similar conformation when in bulk wateror adsorbed to the FMS surface. These rmsd values remain fairlystable for the FMSCoul and FMSvdW ensembles during theremaining simulation time. Conversely, the rmsd values fromthe OPHfree ensemble increase further before converging to 0.2nm after 20 ns of simulation. Therefore, the OPH averageconformation deviates more from the substrate-bound conforma-tion in the X-ray structure on bulk water than that adsorbed toFMS surfaces. The atom positional RMSF from the threesimulations exhibit similar profiles (Figure 2). The regions ofhighest atomic fluctuations are the same for OPHfree, FMSCoul,and FMSvdW, and they correspond to loop motifs. The loops L3(residues 201-215), L5 (residues 255-275), and L6 (residue305-325) are situated at the entrance of the active site, whereasloop L1 (residues 155-165) is located at both monomer-

monomer and protein-FMS interfaces (Figure 2). Loop L2(residues 172-182) is also at the monomer-monomer interface,but on a region that is not in contact with the FMS surface.Several trends can be observed when comparing the atompositional RMSF from the three simulations. First, FMS-adsorbed OPH exhibits smaller atomic fluctuations in loops L3and L5, which together delimit the entrance of the active site.Second, it also displays larger atomic fluctuations for residuesin the monomer-monomer interface (residues 100-110,130-140, loops L1 and L2) as compared to OPH in bulk water.Since loop L2 is not close to the FMS surface, the increasedatomic displacement of this region suggests that interfacialinteractions have a long-range influence on the structuraldynamics OPH. Last, RMSF values are particularly large forresidues S308, Y309, and V310 in the FMSvdW simulation. Previouscomputational simulations of wild type and mutant variants ofOPH bound to paraoxon, sarin and soman have also reportedan increased atomic displacement of Y309.56,68 These simulationsshow that the displacement of Y309 leads to the opening/closingof a gateway to the active site pocket, and it has been proposedto assist the exit of the leaving group from the hydrophobicpocket.68 However, experimental studies of the mutation Y309Ffailed to find any significant difference in the magnitude of eitherkcat or kcat/KM when compared to the wild-type enzyme for thehydrolysis of paraoxon or diethyl p-chlorophenyl phosphate.69

No significant changes in the secondary structure content wereobserved in the MD simulations (Table 2). This is in agreementwith Fourier transform infrared spectroscopy experiments forOPH entrapped in FMS.12 Changes in structural dynamics weremore significant in the monomer-monomer interface whosecontact area decreases significantly in the FMSvdW ensemblewhile exhibiting an identical profile in the OPHfree and FMSCoul

simulations (Table 2).Conformational Dynamics of OPH in Bulk Water and

Adsorbed onto FMS Surfaces. PCA of the atom positionaldisplacement from the corresponding average structure wasperformed for OPH in the three MD simulations.70,71 Thismathematical treatment enables the separation of low-frequencymotions that typically determine the kinetics of enzymaticactivity from the much larger number of remaining high-frequency motions. The separation in low- and high-frequencycomponents is obtained via a change of coordinate system fromCartesian to eigenvectors of the covariance matrix, which allowsthe extrapolation of motions in the direction of selectedeigenvectors.70,71 PCA of a molecular dynamics simulationtrajectory does not rest on the assumption of a harmonicpotential, and therefore, modes are usually sorted according tovariance rather than frequency. Nevertheless, the largest-amplitude modes of a PCA usually also represent the slowestdynamical transitions. In this study, PCA is used to describelow-frequency, persistent motions of OPH during the MDsimulations. The eigenvalue amplitudes calculated from the MDtrajectories indicate that only a few modes are required toaccount for most of the large-amplitude motions of OPH in theseensembles (Figure 3). The first three eigenvalues account for40%, 34%, and 24% of the total low frequency motions inOPHfree, FMSCoul, and FMSvdW, respectively. Projection of thecorresponding eigenvectors onto the protein sequence revealsthat residues in the loops L3, L5, or L6 are the primarycontributors to the motions (Figure 3). In the OPHfree andFMSCoul ensembles, the regions of highest atomic displacementare confined to loops L3 and L5 in the entrance of the activesite. These loops, however, do not contribute significantly tothe low-frequency motions in the FMSvdW ensemble, where loop

TABLE 1: Description of the Simulated Systems

no. molecules

system solute water ions FMS total atoms dimensions (nm)

OPHfree 6472 27 765 0 0 89 769 0.8 ! 0.90 ! 13.0FMScoul 6444 51 611 166 168 161 611 15.0 ! 14.0 ! 0.8FMSvdW 6444 51 611 166 168 161 611 15.0 ! 14.0 ! 0.8

534 J. Phys. Chem. B, Vol. 114, No. 1, 2010 Gomes et al.

L6 exhibits the largest atomic displacement. The persistence oflarge displacements of residues in loop L3 along the threeeigenvectors in the OPHfree simulation indicates that the loopmoves in a more disordered way than in the FMSCoul or, mostnotably, in the FMSvdW simulations (Figure 3). In the lattersimulation, loops L3 and L5 appear to adopt one predominantconformation in which the active site is more exposed to thesolvent and, presumably, to OPH substrates (Figure 3).

Surface-Protein Interfacial Interactions. The MSD of allprotein atoms from their initial position for OPH in bulk waterand in the different FMS surface models was calculated fromthe MD trajectories (Figure 4). These MSD values do not exhibita linear dependence on time, indicating that this property is notfully converged after 50 ns of simulation. Nevertheless, thequalitative inference of these values is still useful for theproposal of comparing the relative differences in diffusionbehavior of OPH in bulk solution and adsorbed to differentrepresentations of the FMS surface at the microscopic level.The MSD curves differ significantly among the simulations(Figure 4). In the presence of the FMS surface, OPH exhibitspredominantly a lateral diffusion along the surface plane (axesx, y) whereas in bulk water, the protein diffuses randomly alongall the Cartesian axes. The MSD of OPH along the FMSCoul

surface was found to be larger than along the FMSvdW surface.Diffusion coefficients calculated from the MSD of the proteinin OPHfree, FMSCoul, and FMSvdW are 3.69 ! 10-7, 2.93 ! 10-7,and 1.3 ! 10-8 cm2 s-1, respectively (Table 2). Experimentallydetermined diffusion coefficients for OPH in bulk water oradsorbed to inorganic surfaces are not available. However,diffusion coefficients estimated for the enzyme lysozymeadsorbed onto oxidized silica and glass surfaces are more than3 orders of magnitude smaller than those in bulk solution, thatis, "10-7 cm2 s-1.72,73 Lysozyme is quite distinct from OPHwith respect to size (163 a.a. versus 662 a.a.), pI (pH 11 versus

pH 8.3), and conformational flexibility (lysozyme is stabilizedby intramolecular disulfide bridges). Therefore, the diffusioncoefficients for the two proteins adsorbed to silicate surfacesare expected to differ in absolute value. However, it can bereasoned from the diffusion behavior of lysozyme that proteinsgenerally exhibit much slower diffusion when adsorbed thanin bulk water. The neglect of van der Waals interactions in theFMSCoul surface model results in a diffusion coefficient for OPHof the same order of magnitude as those in bulk water.

The difference in the diffusional behavior of OPH in theFMSCoul and FMSvdW simulations is further reflected in theaverage FMS-OPH interfacial interatomic distances. Compari-son of the minimum distance between any pair of atoms in OPHand in the FMS surface displays a remarkably distinct behaviorin the two systems (Figure 4). The minimum interfacialatom-atom distance is nearly invariable at 0.25 nm for theFMSvdW ensemble, whereas it fluctuates between 0.09 and 0.64nm in the FMSCoul ensemble within a 50-ns time scale. Thesefindings indicate that although only electrostatic interactionssuffice for the binding of OPH to the FMS surface, van derWaals interactions play a major role in its adhesion. Thedifferent OPH-FMS binding regimes are correlated withdistinguishable hydration patterns in the interfacial regionbetween OPH and the two surface models (Figure 5). Thesolvent structure in the protein-FMS interface was inferredthrough the calculation of the number density of water moleculesas well as the intermolecular radial distribution function (RDF)for atom pairs from the water molecules and the protein surface(Figure 5). The RDF of N-O atom pairs from residues on theFMS binding surface of OPH and water molecules shows thatthe OPH-FMS interfacial surface is less hydrated in the FMSvdW

model. Projection of the solvent number density onto the three-dimensional structure of the solute further demonstrates that thedifferences in hydration are restricted to the protein-FMS

Figure 2. Root-mean-square deviation of backbone atoms of OPH from the X-ray structure (1HYZ) (A) and root-mean-square fluctuation of CRatoms of OPH as a function of residue sequence number (B). OPHfree (black), FMScoul (red), and FMSvdW (green). Rotational and translational fittingof pairs of structures was applied using all backbone atoms. RMSF is averaged for the two monomers over the final 40 ns and a time window of10 ps.

TABLE 2: Averaged structural properties for OPH in the MD simulations

secondary structure (%) SASA (nm2) diffusion coefficients (cm2 s-1)

system R-helix !-strand turn +coil total monomer interface estimated experimentala

OPHfree 45.0 12.6 33.1 455.2 140.7 3.69 ! 10-7

FMScoul 45.0 13.1 32.3 448.8 138.5 2.93 ! 10-7

FMSvdW 44.8 13.4 32.0 439.8 118.6 1.3 ! 10-8

lysozyme "10-7/"10-11

a Diffusion coefficients in bulk water72 and adsorbed onto oxidized silica surfaces,73 respectively.

Dynamics of OPH Adsorbed onto FMS Surfaces J. Phys. Chem. B, Vol. 114, No. 1, 2010 535

interface region. Within a cutoff of 0.45 nm, there are a totalof 139, 144, and 122 water molecules coordinated to side-chain

nitrogen atoms on the FMS-protein interface in the OPHfree,FMSCoul and FMSvdW simulations, respectively. Hence, local

Figure 3. Eigenvalues and atomic displacement along the most representative eigenvectors calculated from the MD simulations OPHfree (circle orbold line), FMScoul (square or solid line), and FMSvdW (triangle or dashed line). (A) Largest eigenvalues. (B) Atomic displacement along the first(i), second (ii), and third (iii) eigenvectors. (C) Projection of atomic displacements along the first eigenvector onto the three-dimensional structureof (i) OPHfree, (ii) FMScoul, and (iii) FMSvdW. The width of the ribbons illustrates the amplitude of the atomic displacement. For clarity, onlyeigenvector components larger than 0.05 ps are shown. (D) Solvent accessibility surface of the trajectory frame at 50 ns for OPH in (i.) OPHfree,(ii) FMScoul, and (iii) FMSvdW. The residue Y309 and Zn2+ cations are represented in CPK.

536 J. Phys. Chem. B, Vol. 114, No. 1, 2010 Gomes et al.

hydration patterns of OPH adsorbed to the FMSCoul surfacemodel and in bulk solvent are comparable, although distinctfrom that observed for the FMSvdW surface model.

Discussion

The representation of FMS surfaces by means of coarse-grained models with distinct interaction potentials can providevaluable insight into the effect of nonbonded interactions withthe FMS surface on protein dynamics. In this approach,interactions between OPH and the FMS surface can be generi-cally grouped into two components: steric due to the confiningnature of the mesoporous surface, and electrostatic/van derWaals due to the functionalization of the mesopore. It is foundthat the steric contact to the FMS surface as well as the natureof the interfacial interactions modulate the conformationaldynamics of OPH toward different ensembles of structures. Thesteric effect imposed by the curvature of the FMS surfaceinduces mainly local structural rearrangements, whereas theinteraction potentials influence local and nonlocal structuralrearrangements. The local rearrangements are predominantlyconfined to loops in the entrance of the active site (Figure 4)that become more accessible to the solvent and, presumably, tosubstrates. The rate-determining step for second-order catalyticrate of enzymes with extremely efficient reaction rates (1010-108

M-1 s-1) corresponds to the association of the free enzyme andthe free substrate. OPH hydrolyzes paraoxon at nearly diffusion-controlled rates ("107 M-1 s-1),74 although it has beendemonstrated that the rate-limiting step of the reaction changesfrom chemical to physical (diffusional) events as the pKa ofthe leaving group of a series of paraoxon analogues is de-creased.18 It has been argued that the slower reaction rate ofOPH compared to superefficient enzymes may be due to aconformational rearrangement upon substrate binding.18 Indeed,our findings suggest that FMS-confined OPH adopts a confor-mational state that may facilitate substrate binding as comparedto OPH free in solution.

Local structural rearrangements are also influenced by thenature of the OPH-FMS interfacial interactions, as evidencedby the distinct atomic displacement patterns of eigenvectorprojection along the protein sequence in the MD ensembles(Figure 4). The regions of largest atomic displacement cor-respond to loops L3 and L5 in the OPHfree and FMSCoul

ensembles and loop L6 in the FMSvdW ensemble. In the latter,loops L3 and L5 are stabilized in one predominant conformation

where the active site is more exposed to the solvent. ResiduesS308, Y309, and V310 are the major contributors to the atomicdisplacement of the loop L6, leading to the opening/closing ofthe gateway to the active site pocket, as previously proposed inrefs 56 and 68. The different potential interactions describingthe FMS surface also affect the structural dynamics of thequaternary structure of OPH. The strong protein-surfaceadhesion in the FMSvdW ensemble results in the rearrangementof the two monomers with respect to each other, which producesa decrease of 20% and 18% in the SASA of the dimer interfacewith respect to the OPHfree and FMSCoul ensembles, respectively(Table 2). These rearrangements of quaternary structure werenot found to produce any significant change in the secondarystructure content of OPH, in agreement with Fourier transforminfrared spectroscopy measurements for OPH entrapped inFMS.12

The role of nonbonded interactions in the adsorption of OPHto FMS surfaces has also been investigated. Although electro-static interactions suffice for the binding of OPH to the FMSsurface, the neglect of van der Waals contributions in therepresentation of the FMS pore model leads to a significantdecrease in protein-surface adhesion and increase in desorptionevents (Figure 4). Diffusion coefficients calculated from theFMSCoul and OPHfree simulations are 1 order of magnitude fasterthan that in the FMSvdW simulation (Table 2). The latter is moreconsistent with experimental measurements of diffusion coef-ficients for lysozyme adsorbed onto oxidized silica and glasssurfaces, which are more than 3 orders of magnitude smallerthan those for the protein in bulk solution; that is, "10-7 cm2

s-1.72,73 The different adhesion regimes of the two surfacemodels are correlated with distinguishable hydration patternsin the interfacial region between OPH and FMS surface (Figure5). The protein-FMS interface is significantly less hydrated inthe FMSvdW ensemble than in the OPHfree, FMSCoul ones, whosehydration patterns are comparable. The binding of OPH to theFMS surface in aqueous solution involves a trade-off betweenthe unfavorable desolvation penalty due to the removal offavorable interactions with water molecules and the generallyfavorable intermolecular interactions made in the bound state.The comparatively lower hydration of the FMSvdW surface modelimplies a smaller desolvation penalty to overcome during theprotein adsorption process. This makes protein adhesion to theFMSvdW surface stronger than to the FMSCoul surface.

Figure 4. Mean square displacement of OPH atoms from their initial positions over 50 ns of simulation (A) and minimum atom-atom distancesbetween OPH atoms and the FMS surface (B) in simulations FMScoul (black) and FMSvdW (gray).

Dynamics of OPH Adsorbed onto FMS Surfaces J. Phys. Chem. B, Vol. 114, No. 1, 2010 537

It has been shown that OPH is optimally immobilized ontoFMS surfaces functionalized with negatively charged groupsand low coverage (2%).15 In consonance with that, OPHimmobilization efficiency is very low for FMS surfaces func-tionalized with positively charged groups at high coverage(20%).15 On the basis of these findings, a predominantlyelectrostatic-driven mechanism has been proposed to underliethe immobilization of OPH onto functionalized FMS surfaces.However, such a mechanism is not entirely reconcilable withthe observation that under low surface coverage by positivelycharged groups, OPH is also immobilized onto FMS surfaceswith significant efficiency, and that high surface coverage bynegatively charged groups causes a sharp decrease in theimmobilization efficiency of the system.15 Our simulations ofOPH adsorbed onto FMS surface models described by distinctinteraction potentials show that interfacial adhesion is signifi-

cantly enhanced by an increase in the van der Waals contribu-tions and support an alternative interpretation of the aforemen-tioned macroscopic measurements. Therefore, we propose thatthe adsorption of OPH onto FMS surfaces functionalized witheither negatively or positively charged groups is modulated notonly by attraction between opposite charges but also from anincrease in surface hydrophobicity due to the presence ofmethylene groups linking the functional groups to the meso-porous wall. In summary, whereas charge-charge interactionsare detrimental to the binding of OPH to the FMS surface, abalance between hydrophobic and electrostatic interactionsappears necessary to facilitate the desolvation of the FMSfunctional groups and OPH adhesion to the surface.

Conclusion

The successful design of mesoporous support for proteinimmobilization requires a detailed understanding of the con-formational and adsorption properties of the target protein.10,11

It can provide important insights into the enhancement of proteinloading and specific activity as well as into the modulation ofthe hydrophobicity or amphiphilicity of the support for effectivesubstrate diffusion through the pores for instance. Explicitsolvent MD simulations have been performed for the enzymeOPH free in solution as well as adsorbed onto two models ofthe FMS functional carboxyl groups to investigate the role ofnonbonded interactions in the conformational dynamics andadsorption of the protein. In the FMSCoul model, van der Waalsinteractions were made negligible by representing the functionalgroup by interaction potentials corresponding to an oxyanion,whereas in the FMSvdW model, van der Waals interactions wereaccounted for by interaction potentials representative of anacetate group. It has been found that the steric contact to theFMS surface as well as the nature of the interfacial interactionsmodulate the structural dynamics of OPH toward differentensembles of structures. Structural rearrangements of loops inthe entrance of the active site increase its accessibility to thesolvent and presumably to substrates and could facilitateproductive binding between substrate and enzyme. In fact, thehydrolysis rate of paraoxon by OPH appears to be limited bychemical events, among which are substrate-induced confor-mational changes, as proposed by Caldwell et al.18 The differentFMS surface models also affect the structural dynamics of thequaternary structure of OPH, with a significant decrease in theSASA of the dimer interface for the FMSvdW ensembles (Table2). These rearrangements of quaternary structure were not foundto produce any significant change in the secondary structurecontent of OPH, in agreement with Fourier transform infraredspectroscopy measurements for OPH entrapped in FMS.12 It hasalso been found that although electrostatic interactions sufficedfor the binding of OPH to the FMS surface, the neglect of vander Waals contributions in the representation of the FMS poremodel led to a significant decrease in protein-surface adhesionand increase in desorption events. The comparatively lowerhydration of the FMSvdW surface model implies a smallerdesolvation penalty to overcome during the protein adsorptionprocess and stronger protein adhesion to the FMSvdW ascompared to the FMSCoul surface models. Furthermore, thediffusion coefficient calculated from the FMSvdW simulation isconsistent with experimental measurements of diffusion forlysozyme adsorbed onto oxidized silica and glass surfaces.73 Inaddition, when fully adsorbed onto the FMSvdW surface model,OPH displays a conformational dynamics in which the activesite of the enzyme is unobstructed by loop L6. This findingsuggests that the observed enhanced catalytic activity of FMS-

Figure 5. (A) Radial distribution function g(r) of intermolecular N-Opairs in the side-chains of residues at the FMS interface of OPH andin the water molecules and (B) the average solvent number densityaround OPH in the FMScoul (a) and FMSvdW (b) ensembles. Solventdensity representation along the longitudinal y-axis. OPHfree (black),FMScoul (red), and FMSvdW (green).

538 J. Phys. Chem. B, Vol. 114, No. 1, 2010 Gomes et al.

confined OPH may sprawl from a combined effect of confine-ment and the adoption of a conformation induced uponadsorption. It has also been argued, with basis on the presentedsimulations, that the adsorption of OPH onto FMS surfacesfunctionalized with either CH2-CH2-COO- or CH2-CH2-NH+

3 groups is modulated not only by attraction betweenopposite charges but also from an increase in surface hydro-phobicity due to the presence of methylene (-CH2) groupslinking the functional groups to the mesoporous.

Acknowledgment. This work was supported by the NIHNational Institute of General Medical Sciences (Grant no.R01GM080987), the Laboratory Directed Research & Develop-ment program “Modeling Protein-Nanomaterial Interactions”funded by the Pacific Northwest National Laboratory, and theBrazilian National Council for Research and Development(CNPq). Computational resources were provided by the Envi-ronmental Molecular Sciences Laboratory at Pacific NorthwestNational Laboratory. Pacific Northwest National Laboratory isoperated for DOE by Battelle.

References and Notes

(1) Kane, R. S.; Stroock, A. D. Biotechnol. Prog. 2007, 23, 316.(2) Rusmini, F.; Zhong, Z. Y.; Feijen, J. Biomacromolecules 2007, 8,

1775.(3) Penn, S. G.; He, L.; Natan, M. J. Curr. Opin. Chem. Biol. 2003, 7,

609.(4) Zhao, Y. N.; Trewyn, B. G.; Slowing, I. I.; Lin, V. S. Y. J. Am.

Chem. Soc. 2009, 131, 8398.(5) Slowing, I. I.; Trewyn, B. G.; Lin, V. S. Y. J. Am. Chem. Soc.

2007, 129, 8845.(6) Trewyn, B. G.; Giri, S.; Slowing, I. I.; Lin, V. S. Y. Chem. Commun.

2007, 3236.(7) Slowing, I. I.; Trewyn, B. G.; Giri, S.; Lin, V. S. Y. AdV. Funct.

Mater. 2007, 17, 1225.(8) Weitkamp, J.; Hunger, M.; Rymsa, U. Microporous Mesoporous

Mater. 2001, 48, 255.(9) Eggers, D. K.; Valentine, J. S. Protein Sci. 2001, 10, 250.

(10) Hanefeld, U.; Gardossi, L.; Magner, E. Chem. Soc. ReV. 2009, 38,453.

(11) Hudson, S.; Cooney, J.; Magner, E. Angew. Chem., Int. Ed. 2008,47, 8582.

(12) Lei, C. H.; Shin, Y.; Magnuson, J. K.; Fryxell, G.; Lasure, L. L.;Elliott, D. C.; Liu, J.; Ackerman, E. J. Nanotechnology 2006, 17, 5531.

(13) Deere, J.; Magner, E.; Wall, J. G.; Hodnett, B. K. J. Phys. Chem.B 2002, 106, 7340.

(14) Liu, J.; Shin, Y.; Nie, Z. M.; Chang, J. H.; Wang, L. Q.; Fryxell,G. E.; Samuels, W. D.; Exarhos, G. J. J. Phys. Chem. A 2000, 104, 8328.

(15) Lei, C.; Shin, Y.; Liu, J.; Ackerman, E. J. J. Am. Chem. Soc. 2002,124, 11242.

(16) Lei, C.; Shin, Y.; Liu, J.; Ackerman, E. J. Nano Lett. 2007, 7, 1050.(17) Lei, C.; Soares, T. A.; Shin, Y.; Liu, J.; Ackerman, E. J.

Nanotechnology 2008, 19, 125102.(18) Caldwell, S. R.; Newcomb, J. R.; Schlechtand, K. A.; Raushel, F. M.

Biochemistry 1991, 30, 7438.(19) Chen-Goodspeed, M.; Sogorb, M. A.; Wu, F. Y.; Hong, S. B.;

Raushel, F. M. Biochemistry 2001, 40, 1325.(20) Chen-Goodspeed, M.; Sogorb, M. A.; Wu, F. Y.; Raushel, F. M.

Biochemistry 2001, 40, 1332.(21) Gopal, S.; Rastogi, V.; Ashman, W.; Mulbry, W. Biochem. Biophys.

Res. Commun. 2000, 279, 516.(22) Li, W. A.; Aubert, S. D.; Raushel, F. M. J. Am. Chem. Soc. 2003,

125, 7526.(23) Li, W. A.; Lum, K. T.; Chen-Goodspeed, M.; Sogorb, M. A.;

Raushel, F. M. Bioorg. Med. Chem. 2001, 9, 2083.(24) Omburo, G. A.; Kuo, J. M.; Mullins, L. S.; Raushel, F. M. J. Biol.

Chem. 1992, 267, 13278.(25) Raushel, F. M. Curr. Opin. Microbiol. 2002, 5, 288.(26) Roodveldt, C.; Tawfik, D. S. Protein Eng., Des. Sel. 2005, 18, 51.(27) Cho, C. M. H.; Mulchandani, A.; Chen, W. Appl. EnViron.

Microbiol. 2004, 68, 2026.(28) Cho, C. M. H.; Mulchandani, A.; Chen, W. Protein Eng., Des. Sel.

2006, 19, 99.(29) Hill, C. M.; Li, W. S.; Thoden, J. B.; Holden, H. M.; Raushel,

F. M. J. Am. Chem. Soc. 2004, 125, 8990.

(30) Lei, C. H.; Valenta, M. M.; Saripalli, K. P.; Ackerman, E. J. J.EnViron. Qual. 2007, 36, 233.

(31) Ghanem, E.; Raushel, F. M. Toxicol. Appl. Pharmacol. 2005, 207,459.

(32) Deere, J.; Serantoni, M.; Edler, K. J.; Hodnett, B. K.; Wall, J. G.;Magner, E. Langmuir 2004, 20, 532.

(33) Essa, H.; Magner, E.; Cooney, J.; Hodnett, B. K. J. Mol. Catal. B:Enzym. 2007, 49, 61.

(34) Goradia, D.; Cooney, J.; Hodnett, B. K.; Magner, E. J. Mol. Catal.B: Enzym. 2005, 32, 231.

(35) Haynes, C. A.; Norde, W. J. Colloid Interface Sci. 1995, 169, 313.(36) Chong, A. S. M.; Zhao, X. S. Appl. Surf. Sci. 2004, 237, 398.(37) He, J.; Liu, Z. J.; Hai, C. X. Aiche J. 2008, 54, 2495.(38) Vinu, A.; Miyahara, M.; Ariga, K. J. Phys. Chem. B 2005, 109,

6436.(39) Vinu, A.; Streb, C.; Murugesan, V.; Hartmann, M. J. Phys. Chem.

B 2003, 107, 8297.(40) Katiyar, A.; Ji, L.; Smirniotis, P.; Pinto, N. G. J. Chromatogr., A

2005, 1069, 119.(41) Katiyar, A.; Ji, L.; Smirniotis, P. G.; Pinto, N. G. Microporous

Mesoporous Mater. 2005, 80, 311.(42) Vinu, A.; Murugesan, V.; Tangermann, O.; Hartmann, M. Chem.

Mater. 2004, 16, 3056.(43) Diaz, J. F.; Balkus, K. J. J. Mol. Catal. B: Enzym. 1996, 2, 115.(44) Adcock, S. A.; McCammon, J. A. Chem. ReV. 2006, 106, 1589.(45) van Gunsteren, W. F.; Bakowies, D.; Baron, R.; Chandrasekhar,

I.; Christen, M.; Daura, X.; Gee, P.; Geerke, D. P.; Glattli, A.; Hunenberger,P. H.; Kastenholz, M. A.; Ostenbrink, C.; Schenk, M.; Trzesniak, D.; vander Vegt, N. F. A. Angew. Chem., Int. Ed. 2006, 45, 4064.

(46) van Gunsteren, W. F.; Bakowies, D.; Burgi, R.; Chandrasekhar, I.;Christen, M.; Daura, X.; Gee, P.; Glattli, A.; Hansson, T.; Oostenbrink, C.;Peter, C.; Pitera, J.; Schuler, L.; Soares, T. A.; Yu, H. B. Chimia 2001, 55,856.

(47) Feng, X.; Fryxell, G. E.; Wang, L. Q.; Kim, A. Y.; Liu, J.; Kemner,K. M. Science 1997, 276, 923.

(48) Benning, M. M.; Hong, S. B.; Raushel, F. M.; Holden, H. M. J. Biol.Chem. 2000, 275, 30556.

(49) Benning, M. M.; Shim, H.; Raushel, F. M.; Holden, H. M.Biochemistry 2001, 40, 2712.

(50) Rodriguez, R.; Chinea, G.; Lopez, N.; Pons, T.; Vriend, G.Bioinformatics 1998, 14, 523.

(51) Bas, D.; Rogers, D. M.; Jensen, J. H. Proteins: Struct., Funct.,Bioinf. 2008, 73, 765.

(52) Oostenbrink, C.; Soares, T. A.; van der Vegt, N. F. A.; vanGunsteren, W. F. Eur. Biophys. J. Biophys. 2005, 34, 273.

(53) Soares, T. A.; Daura, X.; Oostenbrink, C.; Smith, L. J.; vanGunsteren, W. F. J. Biomol. NMR 2004, 30, 407.

(54) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Kollman, P. A. J. Am.Chem. Soc. 1993, 115, 9620.

(55) Bylaska, E. J.; de Jong, W. A.; Kowalski, K.; Straatsma, T. P.;Valiev, M.; Wang, D.; Apra, E.; Windus, T. L.; Hirata, S.; Hackler, M. T.;Zhao, Y.; Fan, P.-D.; Harrison, R. J.; Dupuis, M.; Smith, D. M. A.;Nieplocha, J.; Tipparaju, V.; Krishnan, M.; Auer, A. A.; Nooijen, M.;Brown, E.; Cisneros, G.; Fann, G. I.; Frucht, H.; Garza, J.; Hirao, K.;Kendall, R.; Nichols, J. A.; Tsemekhman, K.; Wolinsk K.; Anchell, J.;Bernholdt, D.; Borowski, P.; Clark, T.; Clerc, D.; Dachsel, H.; Deegan,M.; Dyall, K.; Elwood, D.; Glendening, E.; Gutowski, M.; Hess, A.; Jaffe,J.; Johnson, B.; Ju, J.; Kobayashi, R.; Kutteh, R.; Lin, Z.; Littlefield, R.;Long, X.; Meng, B.; Nakajima, T.; Niu, S.; Pollack, L.; Rosing, M.;Sandrone, G.; Stave, M.; Taylor, H.; Thomas, G.; Lenthe, J. v.; Wong, A.;Zhang. Z. NWChem, A Computational Chemistry Package for ParallelComputers, Version 5.0; Pacific Northwest National Laboratory: Richland,Washington 99352-0999, 2006; a modified version.

(56) Soares, T. A.; Osman, M.; Straatsma, T. P. J. Chem. TheoryComput. 2007, 3, 1569.

(57) Chandrasekhar, I.; Kastenholz, M.; Lins, R. D.; Oostenbrink, C.;Schuler, L. D.; Tieleman, D. P.; van Gunsteren, W. F. Eur. Biophys.J. Biophys. 2003, 32, 67.

(58) Lins, R. D.; Hunenberger, P. H. J. Comput. Chem. 2005, 26, 1400.(59) Soares, T. A.; Hunenberger, P. H.; Kastenholz, M. A.; Krautler,

V.; Lenz, T.; Lins, R. D.; Oostenbrink, C.; van Gunsteren, W. F. J. Comput.Chem. 2005, 26, 725.

(60) Marrink, S. J.; Risselada, H. J.; Yefimov, S.; Tieleman, D. P.; deVries, A. H. J. Phys. Chem. B 2007, 111, 7812.

(61) Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.;Hermans, J. Interaction models for water in relation to protein hydration.In Intermolecular Forces; Pullman, B., Ed.; Reidel: Dordrecht, 1981; p331.

(62) van Gunsteren, W. F.; Berendsen, H. J. C. Mol. Simul. 1988, 1,173.

(63) Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M.J. Comput. Chem. 1997, 18, 1463.

Dynamics of OPH Adsorbed onto FMS Surfaces J. Phys. Chem. B, Vol. 114, No. 1, 2010 539

(64) Miyamoto, S.; Kollman, P. A. J. Comput. Chem. 1992, 13, 952.(65) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.;

Pedersen, L. G. J. Chem. Phys. 1995, 103, 8577.(66) Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. J. Chem.

Theory Comput. 2008, 4, 435.(67) Humphrey, W.; Dalke, A.; Schulten, K. J. Mol. Graphics 1996,

14, 33.(68) Koca, J.; Zhan, C. G.; Rittenhouse, R. C.; Ornstein, R. L. J. Am.

Chem. Soc. 2001, 123, 817.(69) Aubert, S. D.; Li, Y. C.; Raushel, F. M. Biochemistry 1994, 43,

5707.

(70) Amadei, A.; Linssen, A. B.; Berendsen, H. J. Proteins 1993, 17,412.

(71) Garcıa, A. E. Phys. ReV. Lett. 1992, 68, 2696.(72) Dubin, S. B.; Clark, N. A.; Benedek, G. B. J. Chem. Phys. 1971,

54, 5158.(73) Hanasaki, I.; Takahashi, H.; Sazaki, G.; Nakajima, K.; Kawano, S.

J. Phys. D: Appl. Phys. 2008, 41.(74) Dumas, D. P.; Caldwell, S. R.; Wild, J. R.; Raushel, F. M. J. Biol.

Chem. 1989, 264, 19659.

JP9083635

540 J. Phys. Chem. B, Vol. 114, No. 1, 2010 Gomes et al.