Bilirubin Oxidase Adsorption onto Charged Self-Assembled

Jul 25, 2018 - ... charged self-assembled monolayers (SAMs), including COOH-SAM and NH2-SAM with ... Almeida, Carter, Kilgour, Raymonda, and Tejada...
0 downloads 0 Views 5MB Size
Download PDF
Article Cite This: Langmuir XXXX, XXX, XXX−XXX

pubs.acs.org/Langmuir

Bilirubin Oxidase Adsorption onto Charged Self-Assembled Monolayers: Insights from Multiscale Simulations Shengjiang Yang,† Jie Liu,‡ Xuebo Quan,† and Jian Zhou*,† †

Langmuir Downloaded from pubs.acs.org by UNIV OF THE WESTERN CAPE on 08/09/18. For personal use only.

School of Chemistry and Chemical Engineering, Guangdong Provincial Key Lab for Green Chemical Product Technology, South China University of Technology, Guangzhou 510640, P. R. China ‡ Key Laboratory for Green Chemical Process of Ministry of Education, School of Chemical Engineering and Pharmacy, Wuhan Institute of Technology, Wuhan 430073, P. R. China S Supporting Information *

ABSTRACT: The efficient immobilization and orientation of bilirubin oxidase (BOx) on different solid substrates are essential for its application in biotechnology. The T1 copper site within BOx is responsible for the electron transfer. In order to obtain quick direct electron transfer (DET), it is important to keep the distance between the T1 copper site and electrode surface small and to maintain the natural structure of BOx at the same time. In this work, the combined parallel tempering Monte Carlo simulation with the all-atom molecular dynamics simulation approach was adopted to reveal the adsorption mechanism, orientation, and conformational changes of BOx from Myrothecium verrucaria (MvBOx) adsorbed on charged self-assembled monolayers (SAMs), including COOH-SAM and NH2-SAM with different surface charge densities (±0.05 and ±0.19 C·m−2). The results show that MvBOx adsorbs on negatively charged surfaces with a “back-on” orientation, whereas on positively charged surfaces, MvBOx binds with a “lying-on” orientation. The locations of the T1 copper site are closer to negatively charged surfaces. Furthermore, for negatively charged surfaces, the T1 copper site prefers to orient closer to the surface with lower surface charge density. Therefore, the negatively charged surface with low surface charge density is more suitable for the DET of MvBOx on electrodes. Besides, the structural changes primarily take place on the relatively flexible turns, coils, and α-helix. The native structure of MvBOx is well preserved when it adsorbs on both charged surfaces. This work sheds light on the controlling orientation and conformational information on MvBOx on charged surfaces at the atomistic level. This understanding would certainly promote our understanding of the mechanism of MvBOx immobilization and provide theoretical support for BOx-based bioelectrode design.

■

clusters (TNC),12 and the type 1 copper site was utilized to transfer an electron from the substrate to the TNC active site where the reduction reaction takes place. Because of its excellent stability and high activity at neutral pH with high chloride concentrations13 as well as the high redox potential at the T1 copper site,14 BOx has been widely applied in biosensors,15−17 the degradation of excess bilirubin in blood,18,19 dye effluent decolorization,20 and biofuel cells (BFCs),21,22 especially the BFCs utilizing BOx as the cathode that could work efficiently under physiological conditions and at room temperature.21,22 Thus, BOx is an excellent candidate in enzymatic biofuel cells and biosensors. Efficient electron transfer between the redox-active site of an enzyme and the electrode surface is a crucial process in biofuel cells and biosensors, including mediated electron transfer (MET)23,24 and direct electron transfer (DET).25 DET is an

INTRODUCTION Enzyme or “biocatalyst” immobilization on surfaces is one of the most promising techniques for many biotechnological processes including biocatalysis, biosensors, biofuel cells, and biomedicine. Controlling the orientation of protein on material is critical to improving the bioactivity of enzyme;1,2 specifically, the orientation of protein2 can be easily controlled by a charged surface.3−5 Furthermore, enzyme stability is primarily determined by the orientation and conformation6,7 once adsorption occurs. In our previous works, we already pointed out that the orientation of proteins on charged surfaces is mainly dominated by the electric dipole within proteins.3,8−10 Bilirubin oxidase (BOx, EC 1.3.3.5), a member of the family of the blue multicopper oxidases, can efficiently catalyze the oxidation of bilirubin, diphenols, and aryl diamines with the concomitant reduction of O2 to water.11 The active center of BOx contains four redox-active copper atoms that can be classified into three types of sites, including type 1 (T1), type 2 (T2), and type 3 (2 × T3) according to their spectroscopic signatures.11,12 The T2 and T3 copper sites form trinuclear © XXXX American Chemical Society

Received: June 12, 2018 Revised: July 21, 2018 Published: July 25, 2018 A

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

rise in computational power, computer molecular simulation50 provides an alternative to studying the nature of orientation and conformational changes of proteins/peptides adsorbed onto various surfaces.51−53 Conventional Monte Carlo (MC) and molecular dynamics (MD) simulations in canonical ensembles often suffer from the quasi-ergodicity problems;8 i.e., simulations at low temperatures tend to get trapped in local-minimum-energy states and thus fail to sample the whole configuration space. Therefore, when Metropolis MC simulation is applied to probe the lowest-energy orientation, it is not only time-consuming but also may be unreliable. To solve this problem, Xie et al.8 developed a parallel tempering Monte Carlo (PTMC) algorithm based on the united-residue model.3 With this method, both the lowest-energy preferred orientation and orientation distribution could be acquired in a single simulation. The preferred orientation predicted by PTMC as the initial configuration and further mechanisms of the interaction of protein and the surface including conformational changes, binding sites, key adsorption residues, driving forces, and the adsorption free energy could be obtained by the following MD simulations at the atomistic level. Previous studies showed that the surface-induced conformational change was much slower than the orientation adjustment on surfaces.54,55 It is suggested that proteins adsorbed on surfaces with the preferred orientation by translational and rotational motions before subsequent conformational changes. Therefore, obtaining an appropriate initial orientation above the surface is essential before performing MD simulations. The behavior of BOx adsorption on charged surfaces28,39,40,43 and different solid materials31,36−38,41 has been reported by some researchers with diverse experimental methods. The catalytic activity, thermal stability, and DET rate after adsorption have been investigated. Previous works found that the negatively charged surface is better for DET. However, until now, the adsorption mechanism of BOx on charged surfaces was unclear at the atomistic level. In this work, PTMC and all-atom molecular dynamics (AAMD) are employed to investigate the orientation and conformation of BOx adsorbed on charged SAMs, which have been used to investigate DET in previous works.39,40,56 We use PTMC to obtain the preferred preliminary orientation of BOx adsorbed on different charged surfaces. With these predetermined orientations, subsequent AAMD was performed to study the atomistic mechanism of adsorption behavior of BOx on charged SAMs. These amino (NH 2 −)- and carboxyl (COOH−)-terminated SAMs represent positively charged and negatively charged surfaces, respectively. The orientation and conformational changes of BOx adsorbed on these surfaces are explored. BOx solvated in water is investigated as a reference system. The key binding sites, root-mean-square deviation (RMSD), orientation distribution, radius of gyration (Rg), eccentricity, root-mean-square fluctuation (RMSF), definition of the secondary structure of proteins (DSSP), dipole moment, and adsorption energies are discussed in detail.

ideal process in which the electrons are transferred directly from electrodes to the substrate via the redox-active center of the enzyme without any redox mediators.25,26 However, the redox mediators are essential for MET, which may induce higher overpotentials during the electrocatalytic process. 2,2′Azinobis(3-ethylbenzothiazoline-6-sulfonate) (ABTS)27 and metal complexes14 were usually used as redox mediators. It is crucial that the redox-active site is close to the electrode surface for DET. The electron-transfer rate decreases drastically with an increase in the distance between the electrode surface and the redox-active site of the enzymes.25 For BOx, to get better DET, it is necessary to shorten the distance between the T1 copper site and the substrate. Recently, the immobilization of BOx on different kinds of solid materials, such as carbon electrode,28 gold electrodes/ nanoparticles, 17,29−32 carbon nanotubes, 33−36 graphene oxide,37 pyrolytic graphite,38 self-assembled monolayers (SAMs),39,40 and silica nanoparticles41 has been reported by many researchers. By immobilization, the catalytic activity, thermal stability, and reusability are all improved compared to those of the free one.28,34,42 Xia et al.28 investigated the adsorption behavior and catalytic activity of BOx on a charged aromatic-compound-modified carbon electrode, they found that the modification by negatively charged 4-aminobenzoic acid drastically enhances the DET of BOx compared to that of the positively charged compound. They concluded that the favorable orientation and surface charge are key factors influencing the catalytic activity of BOx. They also figured out that the electrostatic interactions between BOx (especially positively charged residues arginine and lysine near the redoxactive center) and negatively charged carbon nanotubes are the most important factors that govern the performance of DETtype bioelectrocatalysis.43 Tominaga et al.40 revealed that the electrocatalytic current of BOx adsorbed onto Au(111) electrodes modified with C3-SO3H and Cn-COOH was strong but did not show any electrocatalytic current when adsorbed onto that modified with C6-NH2, C6-OH, and C5-CH3. They mentioned that the molecular orientation of BOx on electrode surfaces played a dominant role in the DET of BOx on the electrode, and the negatively charged electrode surfaces give a suitable molecular orientation for DET, following the T1 copper site close to electrode surfaces. The T1 copper site is responsible for the electron transfer, and in order to obtain the quick direct electron transfer (DET), it is essential to keep the distance between the T1 cooper site and the electrode surface short. Gutierrez-Sanchez et al.39 showed that BOx adsorbed spontaneously on two oppositely charged self-assembled monolayers (SAMs) with a different catalytic process for O2 reduction, including direct catalysis at negative interfaces and mediated catalysis at positive interfaces, and they assumed that BOx adsorbed on negative SAMs with the orientation of T1 copper close to the surface. It was mainly governed by strong electrostatic interactions between arginine residues close to the T1 copper center and the carboxylic groups on the SAMs. In recent years, more and more experimental techniques have been used to investigate the protein−surface interaction, including circular dichroism spectroscopy (CD),44 solid-state nuclear magnetic resonance (NMR),45 hydrogen−deuterium exchange mass spectroscopy,46 atomic force microscopy (AFM),47 single-molecule force spectroscopy (SMFS),48 and sum frequency generation (SFG) vibrational spectroscopy.49 However, the orientation of enzyme at the atomistic level is very difficult to obtain from experiments. With the exponential

■

EXPERIMENTAL SECTION

The initial configuration of BOx was taken from the high-resolution X-ray structure of BOx from Myrothecium verrucaria12 (MvBOx) available in the protein data bank (PDB ID 2XLL). It is a member of the family of blue multicopper oxidases. The protein has 534 residues with a molecular mass of about 66 kDa. The pI of the enzyme is 4.2. It contains four redox-active copper atom assigned as type 1 (T1), type 2 (T2), and a pair of type 3’s (T3).12 TNC was formed by T2 and T3 B

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

added via the pdb2gmx tool of GROMACS 4.6.7.59 Water molecules were described by the TIP3P model.60 The MvBOx was simulated under physiological conditions, the acidic amino acids including glutamate and aspartate and the C-terminus of MvBOx were deprotonated, and the basic amino acids including lysine and arginine along with the N-terminus were protonated. The four redox-active copper atoms in MvBOx were assigned a charge of +2e in their oxidized state. The net charge of MvBOx was −16e. The potential parameters for protein are from the CHARMM27 force field.61 For SAMS, the (√3 × √3) R30 lattice structure62,63 was adopted for SAMs of HS(CH2)10COOH SAM and HS(CH2)10NH2 in the AAMD simulations. In total, the surface including 396 thiol chains and all the sulfur atoms within SAMs were kept fixed during AAMD simulations. In this work, two surface charge densities (SCD) was considered (i.e., ±0.05 and ±0.19 C·m−2). Twenty-eight (an ∼7% degree of dissociation) and 99 (an ∼25% degree of dissociation) thiol chains were randomly protonated/deprotonated representing SCDs of ±0.05 and ±0.19 C·m−2, respectively. All of the SAMs had dimensions of 8.991 × 9.516 nm2. The potential parameters for SAMs are obtained from the CHARMM27 force field.61 All AAMD simulations were performed with the GROMACS 4.6.7 software package.59 MvBOx was placed 0.5 nm above the SAMs at the beginning of MD simulations. The dimensions of the simulation box were 8.991 × 9.516 × 10.5 nm3. Water molecules were added to the simulation box with the TIP3P model.60 Water molecules within the distance of 0.28 nm from the protein and surfaces were removed. To keep each system neutral, counterion chlorine (Na+) or sodium (Cl−) was added to the simulation boxes for negatively or positively charged SAM systems, respectively. The AAMD simulation was performed in a canonical ensemble with an integration time of 2 fs. The simulated temperature was 300 K, which was controlled by a Nosé−Hoover thermostat64 with a coupling time of 0.5 ps. All bonds, including those to hydrogen atoms, were constrained by the LINCS algorithm.65 A switched potential, with a switch function starting at 0.9 nm and reduced to zero at 1.0 nm, was employed to calculate the nonbonding interactions. Electrostatic interactions were calculated by the particle mesh Ewald66 method with 3dc geometry.67 The neighbor list was updated every 10 steps with a cutoff distance of 1.1 nm. Periodic boundary conditions was used only in the x and y directions; in the z direction, two hard walls67 were used at the top and the bottom of the simulation box, respectively. As a reference, MvBOx from the crystal structure solvated in bulk water was also used. The periodic boundary conditions of MvBOx in the bulk were used in the x, y, and z directions, and the other parameters are the same as those mentioned for surface systems. For each AAMD simulation, the steepest-descent method was first employed to optimize until the minimum energy was lower than 100 kJ·mol−1, which eliminates the steric overlap or inappropriate geometry. Then, in order to equilibrate the solvent and ions around the protein, 1000 ps NVT equilibration was performed with a position restraint on heavy atoms of the protein. Finally, for each system, a 160 ns AAMD simulation was performed without any restrictions at 300 K. In order to confirm the convergence of simulations, the time evolution of total protein−surface interaction energy for the 7% COOH-SAM system during the MD simulation is plotted in Figure S2. It can be found that the total protein−surface interaction energy reached equilibrium after 80 ns, which indicates that 160 ns of simulation is sufficient for this work. The UCSF chimera package68 was used to calculate the surface electrostatic profile. Simulation results were further analyzed by GROMACS59 postprocessing tools, and the visual molecular dynamics (VMD) package69 was employed for structure visualization.

within MvBOx, and T1 copper is separated with TNC and close to the surface of the protein (as shown in Figure 1). The crystal water and copper atoms were reserved in both PTMC and AAMD simulations.

Figure 1. Secondary structure of MvBOx and illustration of TNC and the orientation angles. The green triangle signified TNC. T1 copper and TNC are displayed in green beads. Arrows a (black), b (blue), and c (magenta) show the direction of the TNC plane normal, surface normal, and dipole, respectively. PTMC. PTMC simulations were performed to obtain the preliminary preferred adsorption orientations of MvBOx on different charged surfaces. Each amino acid of MvBOx was reduced to a single interaction site at the α-carbon of the residue, which well retains the basic structure characters of MvBOx according to the united-residue model.3 The protein structure was kept rigid during simulations, SAMs were treated as a structureless flat surface, and the van der Waals (VDW) and electrostatic interactions were considered. Five replicas, with temperatures distributions of 300, 500, 800, 1500, and 2500 K,8 were simulated in parallel, which ensures sufficient energy overlap between neighboring replicas to allow for the reasonable acceptance of configuration swaps which are performed every 500 cycles. The interaction energies between MvBOx and charged SAMs over the electric dipole orientation and hydrophobic dipole orientation are plotted in Figure S1. In our previous work,57 we proposed that the orientation of an adsorbed protein was controlled by both electric and hydrophobic dipoles. It can be seen that the final preferred MvBOx orientation with the lowest adsorption energy (black area) can be efficiently obtained from PTMC simulations. Each of them was simulated in the canonical ensemble. The parameters were taken from our previous works.3,8,58 The MC simulation was carried out in a box of 10 × 10 × 10.5 nm3 in each replica. MvBOx was initially put in the center of the simulation box with a random orientation, and then the rotation and translation move around its center of mass during the simulation. The displacement of each move was adjusted to ensure an acceptance ratio of 0.5, which is defined as p(E i , T i → E i + 1 , T i + 1 ) = min(1, exp(ΔβΔE)), where

Δβ =

1 kb

(

1 Ti + 1

−

1 Ti

) is the difference between the inverse temper-

atures of neighboring replicas and E = ΔEi+1 − Ei is the energy difference between neighboring replicas. More details could be found from our previous work.8 First, 20 000 000 MC cycles were performed for system equilibrium, and then 20 000 000 MC cycles were carried out for production, a total of 40 000 000 cycles. The final preferred MvBOx orientation with the lowest adsorption energy was chosen as the initial configuration for the following MD simulations. AAMD. With the initial orientation of MvBOx taken from PTMC simulations, the MD simulations were performed to further investigate the atomistic mechanism of adsorption behavior of MvBOx on charged SAMs. Hydrogen atoms within MvBOx were

■

RESULTS AND DISCUSSION In this work, a multiscale simulation method, combining PTMC and AAMD, was employed to investigate the orientation of MvBOx on charged surfaces. The distance between the T1 copper site and the substrate was the key factor in the DET process, so the distance between the T1 copper and SAMs was discussed in detail. The TNC tilt angle C

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir of MvBOx, which is defined as the angles (α) between the TNC plane normal and the surface normal, was also analyzed. Furthermore, the dipole moment, RMSD, RMSF, eccentricity, Rg, adsorption energies, and superimposed structures were calculated to provide valuable insights into the conformational changes of MvBOx. The results are shown in Tables 1 and 2 and in Figures 2−8. Table 1. Interaction Energies between MvBOx and Charged Surfaces by AAMD system

Utot (kJ·mol−1)a

Uele (kJ·mol−1)

Uvdw (kJ·mol−1)

HOOC-SAM (7%) HOOC-SAM9 (25%) NH2-SAM (7%) NH2-SAM (25%)

−574.6 ± 61.8

−468.1 ± 58.0

−106.5 ± 22.3

−569.7 ± 63.0

−383.0 ± 63.1

−186.7 ± 20.0

−650.4 ± 95.8 −755.8 ± 98.2

−455.9 ± 90.5 −636.3 ± 95.3

−194.5 ± 35.9 −119.5 ± 26.1

a

Utot represents the sum of electrostatic and vdW interactions.

Adsorption Orientation of MvBOx and the Distance between the T1 Copper Site and Surfaces. For the DET process of oxidoreductase, the distance between the active center and substrate is usually the key factor for its enzyme activity. In other words, the adsorption orientation of oxidoreductase is vital for exerting their bioactivities. In this work, the effect of different charged surfaces on MvBOx adsorption has been explored in detail. The orientations of MvBOx on both charged SAMs (COOH-SAM and NH2-SAM) obtained from PTMC and AAMD are shown in Figure 2. It can be seen that MvBOx is adsorbed on COOH-SAM surfaces with the back-on orientation (with the principal axis of protein almost antiparallel to the surface normal), whereas on NH2-SAM surfaces the lying-on orientation (with the principal axis of protein relatively parallel to the surface normal) is found. In particular, the orientations of MvBOx underwent almost no changes after the surface charge density increases, which indicates that the surface charge density has no significant influence on the adsorption orientation of MvBOx. The orientation obtained from AAMD is nearly consistent with that from PTMC, which also verifies that the PTMC algorithm is suitable for predicting the orientation of proteins. As seen from Figure 2a,c,e, the dipole direction of MvBOx is almost antiparallel to the surface normal when adsorbed on COOHSAM. The orientation of MvBOx on COOH-SAM is dominated by electrostatic interactions as indicated by the electric dipole. Furthermore, MvBOx adsorbs on COOH-SAM with the single positive patch (as shown in Figure 3a), showing that the orientation is also controlled by the distribution of

Figure 2. Final adsorption configurations of MvBOx adsorbed on differently charged SAMs from PTMC (a, b) and AAMD simulations (c−f): (a) on a negatively charged surface, (b) on a positively charged surface, (c) on a 7% dissociated COOH-SAM, (d) on a 7% dissociated NH2-SAM, (e) on a 25% dissociated COOH-SAM, and (f) on a 25% dissociated NH2-SAM. The magenta arrow indicates the direction of the dipole. Copper sites are represented in green beads. The value of d indicates the distance between the T1 copper site and SAMs.

charged patches above the MvBOx surface.62 For MvBOx adsorbed on a positively charged surface, although it has the reverse charge sign with the surface (a net charge of −16 e) and multiple negative patches are exposed on the surface of MvBOx (as shown in Figure 3a,b), the dipole direction of MvBOx is still almost parallel to the surface normal (Figure 2b,d,f). It indicates again that the orientation of MvBOx adsorbed on charged surfaces is dominated by electrostatic interactions as indicated by the electric dipole, which is consistent with what we had found in our previous works.9,58,62,70

Table 2. Averaged Properties of MvBOx Adsorbed on Charged Surfaces and in Bulk Solution

in bulk on 7% COOH-SAM on 25% COOHSAM on 7% NH2-SAM on 25% NH2-SAM

orientation cos θa

TNC tilt angleb

Rg (nm)

eccentricity

backbone RMSD (nm)

contact residues- RMSD (nm)

dipole moment (D)

−0.97 −0.99

165.9 146.9

2.33 ± 0.006 2.34 ± 0.005 2.37 ± 0.007

0.11 ± 0.004 0.13 ± 0.003 0.13 ± 0.003

0.27 ± 0.01 0.29 ± 0.01 0.37 ± 0.01

0.30 ± 0.02 0.39 ± 0.02

1118 ± 40 1244 ± 79 1168 ± 45

0.92 0.93

35.5 43.4

2.34 ± 0.008 2.31 ± 0.008

0.12 ± 0.003 0.11 ± 0.003

0.30 ± 0.01 0.29 ± 0.02

0.32 ± 0.02 0.33 ± 0.04

1245 ± 35 1357 ± 55

a Orientation refers to the cosine value of the angle between the dipole and the surface normal. bTNC tilt angle refers to the angles (α) between the TNC plane normal and the surface normal.

D

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 3. Final top-view snapshots of MvBOx adsorbed on negatively charged surfaces (a, c, e) and positively charged surfaces (b, d, f) from AAMD: (a, b) Electrostatic maps of MvBOx (left, side view; right, bottom view); (c) on 7% dissociated COOH-SAM; (d) on 7% dissociated NH2SAM; (e) on 25% dissociated COOH-SAM; and (f) on 25% dissociated NH2-SAM. Residues within 0.35 nm from the surface are shown by the licorice model. Positively and negatively charged and neutral residues are rendered in blue, red, and green, respectively.

copper site close to the surface is a practical and effective way to enhance its catalytic activity. As shown in Figure 2c,e, the distances are 1.61 and 1.71 nm when MvBOx is adsorbed on 7 and 25% dissociated COOH-SAMs, respectively. They are noticeably shorter compared to those (d = 3.84 and 3.99 nm) on 7 and 25% dissociated NH2-SAMs (Figure 2d,f), which indicates that the negatively charged surface is more appropriate for the DET-based MvBOx electrode. In addition, by monitoring the time evolution of the minimum distance between T1 copper and the surface (Figure 4), it can be seen that MvBOx adsorbs on COOH-SAM more stably than on NH2-SAM. Therefore, it is more beneficial to the DET between MvBOx and electrodes by controlling the orientation of MvBOx with the T1 copper site close to an electrode surface modified with negative charges, which is consistent with previous experimental results.28,39,40,43 We conclude that the high catalytic activity of MvBOx on the negatively charged surface is mainly caused by the orientation of the T1 copper site. Orientation Distributions. The orientation and conformation of adsorbed protein are strongly influenced by interface properties between proteins and surfaces (i.e., charge character, surface charge density, hydrophobicity, etc.), which further influence the biological activity of protein. The

Figure 4. Time evolution of the distance (d) between the T1 copper sites within MvBOx and the charged surfaces (i.e., COOH-SAM and NH2-SAM with different surface charge densities).

The distance between the T1 copper site and the electrode surface plays a decisive role in improving the electron-transfer rate for the DET-based MvBOx electrode.28,39,40,43 Thus, immobilized MvBOx on a charged substrate with the T1 E

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 5. Orientation distributions of MvBOx adsorbed on oppositely charged surfaces by PTMC and AAMD simulations for (a) the dipole angle and (b) the TNC tilt angle.

Figure 6. Electrostatic and vdW interaction energies between key contact residues of MvBOx and the oppositely charged surfaces: (a) 7% dissociated COOH-SAM, (b) 7% dissociated NH2-SAM, (c) 25% dissociated COOH-SAM, and (d) 25% dissociated NH2-SAM.

almost vertical to the surface and the peak values of cos θ are −0.97 and −0.99, respectively (Figure 2c,e). Values of cos θ are 0.92 and 0.93 (Figure 2d,f) on NH2-SAMs with lower and higher SCDs, respectively. Narrower orientation distributions can be found on COOH-SAMs when compared with those on NH2-SAMs. The orientation distributions of MvBOx on negatively charged surfaces from AAMD is slightly shifted compared to that from PTMC. However, relatively larger shifts are present on the positively charged surfaces because the side chains of protein residues are ignored and only the locations of α-C are considered in PTMC simulations. Moreover, in AAMD simulations, all atoms including flexible side chains are taken into account. Thus, a slight deviation in the orientation distributions occurs. Furthermore, MvBOx has a single positive patch and multiple negative patches on the protein surface; therefore, the orientation of MvBOx adsorbed on negatively charged surfaces is also controlled by the electric dipole and the distribution of charged patches. However, on positively charged surfaces, it is dominated only by the electric dipole, which contributes to the larger shifts in the orientation

orientation of proteins adsorbed on a substrate is usually quantitatively characterized by the orientation angles (θ),9,70 as shown in Figure 1, which is defined as the angle between the unit vector normal to the surface (b) and the unit vector along the protein dipole (c). Usually, the orientation of adsorbed protein is quantitatively represented by the cosine value of this angle (cos θ). Furthermore, the angles (α) between the TNC plane normal and the surface normal (shown in Figure 1) are also calculated to further verify the orientation of MvBOx adsorbed on charged surfaces. The results are shown in Figure 5b. Moreover, the stability of protein adsorbed on surfaces can be quantitatively described by the orientation distribution. In this work, the effect of different charged surfaces (i.e., COOHSAM and NH2-SAM) on MvBOx adsorption has been investigated in detail with multiscale simulations. The corresponding preferred orientation and orientation distribution of MvBOx on charged surfaces derived from PTMC and AAMD simulations are shown in Figures 2 and 5, respectively. As can be seen in Figure 5a, on the 7 and 25% dissociated COOH-SAM surfaces, the dipole direction of MvBOx is F

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

that the orientations of MvBOx adsorbed on charged surfaces with different SCDs are not exactly the same although they have almost the same dipole directions. This indicates that the unique orientation of MvBOx on charged surfaces can be determined by the dipole angle and the TNC tilt angle. It is obvious that the trends in the orientation distribution of MvBOx adsorbed on both charged surfaces with various degrees of dissociation are basically the same. The final configurations of MvBOx on both charged surfaces from PTMC are also similar to those from AAMD (shown in Figures 2 and 3). Interaction Energy and Key Binding Sites. To clarify the adsorption strength and modes of MvBOx on charged surfaces, the total protein−surface interaction energies (the interaction energies between the protein and surfaces) and key binding residues over the last 60 ns of the AAMD simulation are summarized in Table 1 and Table S1. As shown in Table 1, the adsorption process of MvBOx on charged surfaces is controlled by the synergy of electrostatic and vdW interactions. Obviously, the adsorption behavior is mainly dominated by electrostatic interactions, and vdW interactions cannot be ignored, which is consistent with the finding from previous work in that proteins adsorbed on charged surfaces are usually dominated by electrostatic interactions.9,62,70 For a negatively charged surface, the electrostatic interaction decreases with the increase in SCD as the vdW interaction evidently increases. For positively charged surfaces, as the SCD increases, the electrostatic interactions also simultaneously increase but the vdW interaction decreases. Nevertheless, because MvBOx carries net negative charges, the total interaction energies decrease with the increase in SCD on negative SAMs. However, it obviously increases with the increase in SCD on positive SAMs. The total interaction energies are stronger on NH2-SAM compared to those on COOH-SAMs. Even so, the adsorption orientation is not favorable for the DET on positive surfaces as the T1 copper atom is far away from surfaces. It can also be seen from Figure 2 that the T1 copper site is closer to the surface when MvBOx is adsorbed on 7% dissociated COOH-SAM rather than on 25% dissociated COOH-SAM, which indicates that negatively charged surfaces with a low SCD are suitable for the immobilization of MvBOx. It also agrees with the previous experiential result.37 To explore the pivotal residues and adsorption stability of MvBOx adsorbed on both charged surfaces, the residue contact maps are plotted in Figure S3. The key residues of MvBOx in contact with SAMs in the final configuration of AAMD simulations are identified. The results are shown in Figure 3 and Table S1. A residue is thought to be in contact with the surface if the distance between any residue atom and the surface is less than 0.35 nm.62 The results indicate that MvBOx was steadily adsorbed on both charged surfaces (shown in Figure S3). As can be seen from Figure 3 and Table S1, for MvBOx on COOH-SAMs, the major adsorbed residues are Thr357, Gly358, Gly365, Asn394, and three positively charged residues (i.e., Arg353, Arg356, and Arg437). While on NH2-SAMs, the key adsorbed residues include three negatively charged residues (Asp53, Asp322, and Asp323) and Pro52, Thr325, Gln505, Ala506, and Gln507. On both charged surfaces, the charged residues of the MvBOx surface strongly interact with the charged surfaces, as shown in Figure 3a,b, and it is obvious that the single positive patch is close to the COOH-SAM. However, this positive patch is far away from the NH2-SAM; meanwhile, the multiple negatively charged

Figure 7. Simulated structure (red) of MvBOx in bulk water (a) and adsorbed on 7% dissociated COOH-SAM (b), 7% dissociated NH2SAM (c), 25% dissociated COOH-SAM (d), and 25% dissociated NH2-SAM (e) superimposed on its crystal structure (green).

Figure 8. RMSF of each residue of MvBOx in bulk water and adsorbed on dissociated COOH-SAMs (7%, 25%) and NH2-SAMs (7%, 25%). Residues 1 and 533 correspond to the N-terminal and Cterminal, respectively.

distribution on positively charged surfaces. The orientation distributions from AAMD are wider than those from PTMC, which is caused by the adopted rigid united-residue protein model3 in PTMC simulations. As shown in Figure 5a,b, the distribution of the dipole angle on the 7% dissociated COOH-SAM almost overlaps with that on the 25% dissociated COOH-SAM, and the same trends are found on NH2-SAMs. However, the distributions of the TNC tilt angle on the 7% dissociated COOH-SAM present slight shifts compared to that on 25% dissociated COOH-SAM, and similar results can be found on NH2-SAMs. The results show G

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

adsorbed on negatively charged surfaces. On NH2-SAMs, the values of RMSD are 0.3 and 0.29 nm for 7 and 25% dissociated NH2-SAM, respectively. They are almost the same, and no obvious conformational change occurs compared to that in bulk water (0.27 nm). It indicates that the positively charged surfaces do not induce a significant conformational change in MvBOx. Furthermore, the final simulated structures of MvBOx adsorbed on charged SAMs and in bulk solution are superimposed on its crystal structure by VMD69 and shown in Figure 7. To further observe the local conformational change, the RMSF value of each residue is analyzed according to the last 60 ns trajectory of MvBOx adsorbed on SAMs and in bulk water. The residue-based RMSF of a backbone atom can specifically identify regions of protein that undergo high structural changes and fluctuations during the simulation. It is

regions are close to NH2-SAM. It can be seen that from Figure 3 and Table S1 the number of contact residues is almost the same, indicating that although MvBOx carries a net charge of −16e it can still be adsorbed on negatively charged surfaces stably. Just as mentioned in section 3.2, the orientation distributions of MvBOx adsorbed on the COOH-SAMs are narrower than those on the NH2-SAMs (shown in Figure 5a). To evaluate the electrostatic and vdW contributions from each contact residue, electrostatic and vdW interaction energies between contact residues of MvBOx and surfaces are also calculated. The results are summarized in Figure 6. As shown in Figure 6a,c, for the 7 and 25% dissociated COOHSAMs, MvBOx adsorption energy is primarily contributed by the strong electrostatic interaction between positively charged residues Arg353 and Arg437 and the surfaces. The interaction energies are −171.2 and −182.3 kJ·mol−1 on the 7% dissociated COOH-SAM and −186.0 and −178.6 kJ·mol−1 on the 25% dissociated COOH-SAM. Simultaneously, the electrostatic repulsion interaction of negatively charged residue Asp370 and COOH-SAM is presented, which is another reason that the total interaction energies decrease with the increase in SCD on negative SAMs. As illustrated in Figure 6b,d, for NH2-SAM surfaces, MvBOx adsorption is mainly mediated by negatively charged residues Asp53, Asp322, and Asp323. The strong electrostatic interactions of the three residues with NH2-SAM surfaces dominate the adsorption of MvBOx. The interaction energies are −81.1, −8.37, and −110.3 kJ·mol−1 on the 7% dissociated NH2-SAM and −47.1, −67.2, and −122.6 kJ·mol−1 on the 25% dissociated NH2SAM. As shown in Figure 3d,f, it can be found that the distance between Asp322 and 7% dissociated NH2-SAM is longer than that on 25% dissociated NH2-SAM, which leads to the weaker interaction energy between Asp322 and 7% dissociated NH2SAM. Conformational Changes. Protein conformation plays an essential role in determining its stability and catalytic efficiency and further influences its biotechnological applications. Proteins/enzymes undergo conformational changes when adsorbed onto different surfaces (i.e., charged surfaces, hydrophobic or hydrophilic materials, etc.). To quantitatively investigate the overall conformational change in MvBOx adsorbed on charged SAMs, the root-mean-square deviation (RMSD), root-mean-square fluctuation (RMSF), superimposed structures, definition of secondary structure of proteins (DSSP), dipole moment, eccentricity, and Rg are analyzed by averaging over the last 60 ns trajectory of the MD simulations. RMSD, RMSF, Superimposed Structures, and DSSP. The overall conformational change in protein adsorbed surfaces can be quantitatively analyzed by RMSD, which is ÄÅ 1 N ÉÑ1/2 d e fi n e d a s RMSD(t1 , t 2) = ÅÅÅÅ M ∑i = 1 miri(t1) − ri(t1)2 ÑÑÑÑ Ç Ö N where M = ∑i = 1 mi and ri(t) is the position of atom i at time t. The average RMSD values of MvBOx adsorbed on charged SAMs are calculated on the basis of the structures of simulated and crystal MvBOx, and only backbone atoms are considered. The results are illustrated in Table 2. It can be seen that the RMSD value of MvBOx in bulk water is 0.27 nm, which is almost equal to that on 7% dissociated COOH-SAMs (0.29 nm), while on 25% dissociated COOH-SAM the value is 0.37 nm, which indicates that MvBOx undergoes a larger structural change when adsorbed on COOH-SAM with high SCD. This indicates that low SCD is more suitable for MvBOx

defined asRMSF =

∑(ri − ri,ref )2 T

where ri, ri,ref, and T are the

instantaneous position vector, the reference position vector of the ith atom, and the total simulation frames for statistics, respectively. The results are shown in Figure 8. It can be seen that the RMSF values of most residues of MvBOx in bulk water and on charged surfaces are less than 0.2 nm, with only some residues exhibiting relatively larger values. For MvBOx in bulk water, only residues 50−56 (turn) and the C-terminal (residues 1 and 2) of MvBOx exhibit larger RMSF values. Similarly, residues 193−196 (turn), residues 368−374 (coil), residues 488−492 (turn), and the C-terminal (residues 532− 533) exhibit large RMSF fluctuations when MvBOx is adsorbed on 7% dissociated COOH-SAM, whereas for MvBOx on 25% dissociated COOH-SAM, only residues 477−487 (turn and coil), the N-terminal (residue 1) and Cterminal (residues 531−533) exhibit large RMSF values. Simultaneously, residues 337−344 (coil) and the C-terminal (residues 530−533) for MvBOx adsorbed on 7% dissociated NH2-SAM also present larger RMSF values; however, for MvBOx adsorbed on 25% dissociated NH2-SAM, more residues, including residues 159−161 (turn), residues 193− 196 (turn), residues 337−347 (coil), residues 476−482 (helix), and residues 503−508 (helix), display a great value of the RMSF. The results show that the slight structural changes mainly take place in the relatively flexible turns, coils, and α-helix. On the other hand, it can also be seen that from Figure 7 the overall secondary structures of MvBOx are well kept. The information about the stability of secondary structures of MvBOx adsorbed on both charged surfaces was also illustrated by DSSP analysis, and from the results shown in Figure S4, it can be seen that the properties of secondary structures of MvBOx were retained during whole MD simulations. It indicates that MvBOx retains its biological activity when adsorbed on charged surfaces and in bulk water. Radius of Gyration, Eccentricity, and Dipole Moment. The overall structure of MvBOx adsorbed on charged SAMs can be quantitatively analyzed by Rg and the eccentricity. Rg is defined as the mass-weighted root-mean-square average distance of all atoms in a protein from its center of mass and the eccentricity, which quantitatively characterized the ellipsoidal degree of a protein. The eccentricity can be calculated with 1 − Iave/Imax, in which Iave represents the average of three principal moments of inertia and Imax is the maximal principal moment of inertia. If the value of the eccentricity is close to zero, the protein is spherelike; the protein is ellipsoidlike when the eccentricity approaches 1. H

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

latter plays the dominant role. The single positive patch of MvBOx is close to the negatively charged surfaces with the back-on orientation. For positively charged surfaces, the multiple negative patches of MvBOx are close to the surfaces with the lying-on orientation. The simulation results show that the negatively charged surface is beneficial for DET with the T1 copper site close to the electrode surface, while it is always far away from the surface once MvBOx adsorbs on positively charged surfaces. Furthermore, the negatively charged surface with low surface charge density is more suitable for the DET between MvBOx and the electrode. Although MvBOx is a negatively charged protein, it can stably adsorb on negatively charged surfaces and obtain a better adsorption orientation for DET. The RMSD, RMSF, superimposed structures, Rg, eccentricity, and dipole moment are employed to further investigate the conformational change of MvBOx during the adsorption process, the simulation results reveal that the structural changes primarily take place on the relatively flexible turns, coils, and α-helix, and the native overall second structure of MvBOx is well preserved when it adsorbs on both charged surfaces or in bulk water. This work sheds light on the orientation and conformation of MvBOx adsorbed on charged surfaces at the atomistic level. It can also provide guidelines for the preparation of BOx-based bioelectrodes and biosensors with excellent performance.

Their values are shown in Table 2. It can be seen that the value of Rg and the eccentricity of MvBOx in bulk water are 2.32 nm and 0.11. When adsorbed on the COOH-SAM and NH2-SAM, Rg and the eccentricity are almost equal to those in the bulk. It demonstrates that no significant conformational change occurs during the adsorption process of MvBOx on charged surfaces. The dipole moment, which defined N asμ⃗ = ∑i = 1 qi(ri − rCOM), where qi is the partial charge of each atom in a protein and rCOM is the position of center of mass of the protein, can be used to quantitatively analyze the distribution of protein charges and further characterize the conformational changes of MvBOx adsorbed on charged surfaces. It can be seen from Table 2 that the dipole moments of MvBOx adsorbed on negatively charged surfaces are 1244 and 1168 D for 7 and 25% dissociated COOH-SAM, respectively. They are 11.3 and 4.5% larger than that of MvBOx in bulk solution (1118 D), and the dipole moments of MvBOx on 7 and 25% dissociated NH2-SAMs are 1245 and 1357 D, which is 11.4 and 21.4% larger than that in bulk solution. This is mainly caused by the electrostatic interactions between charged residues on the MvBOx surface and charged SAMs. Furthermore, MvBOx carries a net charge of −16 e, which leads to the electrostatic repulsion interaction to the COOH-SAM, and thus it tends to adsorb on COOH-SAM with low SCD. It can be seen from Figure 8 that structural changes primarily take place on the relatively flexible turns, coils, and α-helix and that charged residues Glu195, Asp370, Arg374, Asp488, Glu491, Glu492, and Asp533 of MvBOx adsorbed on 7% dissociated COOH-SAM exhibit the relatively larger RMSF values. On 25% dissociated COOH-SAM, only residues Asp478 and Asp533 present large RMSF values. Simultaneously, on 25% dissociated NH2-SAMs, charged residues Glu160, Asp161, Glu195, Arg337, Asp338, Asp478, and Glu503 exhibit relatively larger RMSF values, and only residues Arg337, Asp338, and Asp533 show a large value of the RMSF. In other words, these charged residues within MvBOx experience slight structural changes. This is the reason that the dipole moment of MvBOx on 7% dissociated COOH-SAM is larger than that on 25% dissociated COOH-SAM while it is larger on 25% dissociated NH2-SAM than on 7% dissociated NH2-SAM. The dipole moment changes of MvBOx are induced by the movement of the flexible turns, coils, and αhelix. Nevertheless, the backbone structure of MvBOx is well preserved, as shown in Figure 7 and Table 2, and the superimposed structure, RMSD, Rg, and eccentricity results also indicate that the overall configuration of MvBOx is not influenced by the charged surfaces.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.8b01974. Key adsorption residues of MvBOx adsorbed on charged surfaces; interaction energies versus the electric dipole and hydrophobic dipole of MvBOx in PTMC sampling; the time evolution of the total protein−surface interaction energy during the MD simulation; contact maps between MvBOx and charged SAMs during simulations and the time evolution of the secondary structures of MvBOx adsorbed on differently charged surfaces (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Fax: +86 20 87114069. Tel: +86 20 87114069.

■

ORCID

Shengjiang Yang: 0000-0002-9920-8906 Jie Liu: 0000-0002-5138-5348 Xuebo Quan: 0000-0003-3738-9059 Jian Zhou: 0000-0002-3033-7785

CONCLUSIONS In this work, the all-atom molecular dynamics simulation combined with the parallel tempering Monte Carlo simulation method have been employed to explore the orientation, conformation, and efficient DET of MvBOx on oppositely charged SAM surfaces with different surface charge densities (±0.05 C·m−2 and ±0.19 C·m−2). The simulation results show that negatively charged surfaces promote the DET process while positively charged surfaces prevent the DET process, which is consistent with previous experimental results, and more molecular-level details are found. MvBOx can well adsorb on both positively and negatively charged surfaces. The adsorption behaviors are controlled by the synergy of the vdW and electrostatic interactions, while the

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Support from the National Natural Science Foundation of China (nos. 21776093, 21376089, and 21706197), the National Key Basic Research Program of China (no. 2013CB733500), the Guangdong Science Foundation (no. 2014A030312007), and the Fundamental Research Funds for the Central Universities (SCUT-2015ZP033) is gratefully I

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

(20) Zhang, X.; Liu, Y.; Yan, K.; Wu, H. Decolorization of anthraquinone-type dye by bilirubin oxidase-producing nonligninolytic fungus Myrothecium sp. IMER1. J. Biosci. Bioeng. 2007, 104 (2), 104−110. (21) Tsujimura, S.; Fujita, M.; Tatsumi, H.; Kano, K.; Ikeda, T. Bioelectrocatalysis-based dihydrogen/dioxygen fuel cell operating at physiological pH. Phys. Chem. Chem. Phys. 2001, 3 (7), 1331−1335. (22) Tsujimura, S.; Tatsumi, H.; Ogawa, J.; Shimizu, S.; Kano, K.; Ikeda, T. Bioelectrocatalytic reduction of dioxygen to water at neutral pH using bilirubin oxidase as an enzyme and 2, 2′-azinobis (3ethylbenzothiazolin-6-sulfonate) as an electron transfer mediator. J. Electroanal. Chem. 2001, 496 (1−2), 69−75. (23) Mao, F.; Mano, N.; Heller, A. Long tethers binding redox centers to polymer backbones enhance electron transport in enzyme “wiring” hydrogels. J. Am. Chem. Soc. 2003, 125 (16), 4951−4957. (24) Habrioux, A.; Napporn, T.; Servat, K.; Tingry, S.; Kokoh, K. Electrochemical characterization of adsorbed bilirubin oxidase on Vulcan XC 72R for the biocathode preparation in a glucose/O2 biofuel cell. Electrochim. Acta 2010, 55 (26), 7701−7705. (25) Léger, C.; Bertrand, P. Direct electrochemistry of redox enzymes as a tool for mechanistic studies. Chem. Rev. 2008, 108 (7), 2379−2438. (26) Ghindilis, A. L.; Atanasov, P.; Wilkins, E. Enzyme-catalyzed direct electron transfer: Fundamentals and analytical applications. Electroanalysis 1997, 9 (9), 661−674. (27) Trifonov, A.; Herkendell, K.; Tel-Vered, R.; Yehezkeli, O.; Woerner, M.; Willner, I. Enzyme-capped relay-functionalized mesoporous carbon nanoparticles: effective bioelectrocatalytic matrices for sensing and biofuel cell applications. ACS Nano 2013, 7 (12), 11358−11368. (28) Xia, H.; Kitazumi, Y.; Shirai, O.; Kano, K. Enhanced direct electron transfer-type bioelectrocatalysis of bilirubin oxidase on negatively charged aromatic compound-modified carbon electrode. J. Electroanal. Chem. 2016, 763, 104−109. (29) Salaj-Kosla, U.; Pö ller, S.; Beyl, Y.; Scanlon, M. D.; Beloshapkin, S.; Shleev, S.; Schuhmann, W.; Magner, E. Direct electron transfer of bilirubin oxidase (Myrothecium verrucaria) at an unmodified nanoporous gold biocathode. Electrochem. Commun. 2012, 16 (1), 92−95. (30) Murata, K.; Kajiya, K.; Nakamura, N.; Ohno, H. Direct electrochemistry of bilirubin oxidase on three-dimensional gold nanoparticle electrodes and its application in a biofuel cell. Energy Environ. Sci. 2009, 2 (12), 1280−1285. (31) Pankratov, D.; Sotres, J.; Barrantes, A.; Arnebrant, T.; Shleev, S. Interfacial behavior and activity of laccase and bilirubin oxidase on bare gold surfaces. Langmuir 2014, 30 (10), 2943−2951. (32) Ramírez, P.; Mano, N.; Andreu, R.; Ruzgas, T.; Heller, A.; Gorton, L.; Shleev, S. Direct electron transfer from graphite and functionalized gold electrodes to T1 and T2/T3 copper centers of bilirubin oxidase. Biochim. Biophys. Acta, Bioenerg. 2008, 1777 (10), 1364−1369. (33) Navaee, A.; Salimi, A.; Jafari, F. Electrochemical Pretreatment of Amino-Carbon Nanotubes on Graphene Support as a Novel Platform for Bilirubin Oxidase with Improved Bioelectrocatalytic Activity towards Oxygen Reduction. Chem. - Eur. J. 2015, 21 (13), 4949−4953. (34) Lalaoui, N.; Le Goff, A.; Holzinger, M.; Cosnier, S. Fully Oriented Bilirubin Oxidase on Porphyrin-Functionalized Carbon Nanotube Electrodes for Electrocatalytic Oxygen Reduction. Chem. Eur. J. 2015, 21 (47), 16868−16873. (35) Mazurenko, I.; Monsalve, K.; Rouhana, J.; Parent, P.; Laffon, C.; Goff, A. L.; Szunerits, S.; Boukherroub, R.; Giudici-Orticoni, M.T.; Mano, N. How the intricate interactions between carbon nanotubes and two bilirubin oxidases control direct and mediated O2 reduction. ACS Appl. Mater. Interfaces 2016, 8 (35), 23074− 23085. (36) Milton, R. D.; Wang, T.; Knoche, K. L.; Minteer, S. D. Tailoring Biointerfaces for Electrocatalysis. Langmuir 2016, 32 (10), 2291−2301.

acknowledged. An allocation of time from the SCUTGrid at South China University of Technology is gratefully acknowledged.

■

REFERENCES

(1) Li, Y.; Ogorzalek, T. L.; Wei, S.; Zhang, X.; Yang, P.; Jasensky, J.; Brooks, C. L.; Marsh, E. N. G.; Chen, Z. Effect of immobilization site on the orientation and activity of surface-tethered enzymes. Phys. Chem. Chem. Phys. 2018, 20 (2), 1021−1029. (2) Zhang, A.; Xie, Y.; Zhou, J. Experimental control and characterization of protein orientation on surfaces. Progress in Chemistry 2009, 21 (7−8), 1408−1417. (3) Zhou, J.; Chen, S.; Jiang, S. Orientation of adsorbed antibodies on charged surfaces by computer simulation based on a united-residue model. Langmuir 2003, 19 (8), 3472−3478. (4) Liu, J.; Liao, C.; Zhou, J. Multiscale simulations of protein G B1 adsorbed on charged self-assembled monolayers. Langmuir 2013, 29 (36), 11366−11374. (5) Xie, Y.; Li, Z.; Zhou, J. Hamiltonian replica exchange simulations of glucose oxidase adsorption on charged surfaces. Phys. Chem. Chem. Phys. 2018, 20 (21), 14587−14596. (6) Secundo, F. Conformational changes of enzymes upon immobilisation. Chem. Soc. Rev. 2013, 42 (15), 6250−6261. (7) Roach, P.; Farrar, D.; Perry, C. C. Interpretation of protein adsorption: surface-induced conformational changes. J. Am. Chem. Soc. 2005, 127 (22), 8168−8173. (8) Xie, Y.; Zhou, J.; Jiang, S. Parallel tempering Monte Carlo simulations of lysozyme orientation on charged surfaces. J. Chem. Phys. 2010, 132 (6), 065101. (9) Zhou, J.; Zheng, J.; Jiang, S. Molecular simulation studies of the orientation and conformation of cytochrome c adsorbed on selfassembled monolayers. J. Phys. Chem. B 2004, 108 (45), 17418− 17424. (10) Chen, S.; Liu, L.; Zhou, J.; Jiang, S. Controlling antibody orientation on charged self-assembled monolayers. Langmuir 2003, 19 (7), 2859−2864. (11) Solomon, E. I.; Sundaram, U. M.; Machonkin, T. E. Multicopper oxidases and oxygenases. Chem. Rev. 1996, 96 (7), 2563−2606. (12) Cracknell, J. A.; McNamara, T. P.; Lowe, E. D.; Blanford, C. F. Bilirubin oxidase from Myrothecium verrucaria: X-ray determination of the complete crystal structure and a rational surface modification for enhanced electrocatalytic O2 reduction. Dalton Transactions 2011, 40 (25), 6668−6675. (13) Mano, N.; Edembe, L. Bilirubin oxidases in bioelectrochemistry: Features and recent findings. Biosens. Bioelectron. 2013, 50, 478− 485. (14) Christenson, A.; Shleev, S.; Mano, N.; Heller, A.; Gorton, L. Redox potentials of the blue copper sites of bilirubin oxidases. Biochim. Biophys. Acta, Bioenerg. 2006, 1757 (12), 1634−1641. (15) Chauhan, N.; Rawal, R.; Hooda, V.; Jain, U. Electrochemical biosensor with graphene oxide nanoparticles and polypyrrole interface for the detection of bilirubin. RSC Adv. 2016, 6 (68), 63624−63633. (16) Feng, Q.; Du, Y.; Zhang, C.; Zheng, Z.; Hu, F.; Wang, Z.; Wang, C. Synthesis of the multi-walled carbon nanotubes-COOH/ graphene/gold nanoparticles nanocomposite for simple determination of Bilirubin in human blood serum. Sens. Actuators, B 2013, 185, 337− 344. (17) Pita, M.; Gutierrez-Sanchez, C.; Toscano, M. D.; Shleev, S.; De Lacey, A. L. Oxygen biosensor based on bilirubin oxidase immobilized on a nanostructured gold electrode. Bioelectrochemistry 2013, 94, 69− 74. (18) Kimura, M.; Matsumura, Y.; Miyauchi, Y.; Maeda, H. A new tactic for the treatment of jaundice: an injectable polymer-conjugated bilirubin oxidase. Exp. Biol. Med. 1988, 188 (3), 364−369. (19) Dennery, P. A.; Seidman, D. S.; Stevenson, D. K. Neonatal hyperbilirubinemia. N. Engl. J. Med. 2001, 344 (8), 581−590. J

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

(54) Gray, J. J. The interaction of proteins with solid surfaces. Curr. Opin. Struct. Biol. 2004, 14 (1), 110−115. (55) Agashe, M.; Raut, V.; Stuart, S. J.; Latour, R. A. Molecular simulation to characterize the adsorption behavior of a fibrinogen γchain fragment. Langmuir 2005, 21 (3), 1103−1117. (56) Li, J.; Cheng, G.; Dong, S. Direct electron transfer to cytochrome c oxidase in self-assembled monolayers on gold electrodes. J. Electroanal. Chem. 1996, 416 (1), 97−104. (57) Peng, C.; Liu, J.; Zhao, D.; Zhou, J. Adsorption of Hydrophobin on Different Self-Assembled Monolayers: The Role of the Hydrophobic Dipole and the Electric Dipole. Langmuir 2014, 30 (38), 11401−11411. (58) Zhao, D.; Peng, C.; Zhou, J. Lipase adsorption on different nanomaterials: a multi-scale simulation study. Phys. Chem. Chem. Phys. 2015, 17 (2), 840−850. (59) Hess, B.; Kutzner, C.; Van Der Spoel, D.; Lindahl, E. GROMACS 4: algorithms for highly efficient, load-balanced, and scalable molecular simulation. J. Chem. Theory Comput. 2008, 4 (3), 435−447. (60) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. Comparison of simple potential functions for simulating liquid water. J. Chem. Phys. 1983, 79 (2), 926−935. (61) MacKerell, A. D., Jr; Bashford, D.; Bellott, M.; Dunbrack, R. L., Jr; Evanseck, J. D.; Field, M. J.; Fischer, S.; Gao, J.; Guo, H.; Ha, S. All-atom empirical potential for molecular modeling and dynamics studies of proteins. J. Phys. Chem. B 1998, 102 (18), 3586−3616. (62) Liu, J.; Xie, Y.; Peng, C.; Yu, G.; Zhou, J. Molecular Understanding of Laccase Adsorption on Charged Self-Assembled Monolayers. J. Phys. Chem. B 2017, 121 (47), 10610−10617. (63) Liu, J.; Yu, G.; Zhou, J. Ribonuclease A adsorption onto charged self-assembled monolayers: A multiscale simulation study. Chem. Eng. Sci. 2015, 121, 331−339. (64) Hoover, W. G. Canonical dynamics: equilibrium phase-space distributions. Phys. Rev. A: At., Mol., Opt. Phys. 1985, 31 (3), 1695. (65) Hess, B.; Bekker, H.; Berendsen, H. J.; Fraaije, J. G. LINCS: a linear constraint solver for molecular simulations. J. Comput. Chem. 1997, 18 (12), 1463−1472. (66) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.; Pedersen, L. G. A smooth particle mesh Ewald method. J. Chem. Phys. 1995, 103 (19), 8577−8593. (67) Yeh, I.-C.; Berkowitz, M. L. Ewald summation for systems with slab geometry. J. Chem. Phys. 1999, 111 (7), 3155−3162. (68) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. UCSF Chimeraa visualization system for exploratory research and analysis. J. Comput. Chem. 2004, 25 (13), 1605−1612. (69) Humphrey, W.; Dalke, A.; Schulten, K. VMD: visual molecular dynamics. J. Mol. Graphics 1996, 14 (1), 33−38. (70) Liu, J.; Peng, C.; Yu, G.; Zhou, J. Molecular simulation study of feruloyl esterase adsorption on charged surfaces: effects of surface charge density and ionic strength. Langmuir 2015, 31 (39), 10751− 10763.

(37) Filip, J.; Andicsová-Eckstein, A.; Vikartovská, A.; Tkac, J. Immobilization of bilirubin oxidase on graphene oxide flakes with different negative charge density for oxygen reduction. The effect of GO charge density on enzyme coverage, electron transfer rate and current density. Biosens. Bioelectron. 2017, 89, 384−389. (38) Dos Santos, L.; Climent, V.; Blanford, C. F.; Armstrong, F. A. Mechanistic studies of the ‘blue’Cu enzyme, bilirubin oxidase, as a highly efficient electrocatalyst for the oxygen reduction reaction. Phys. Chem. Chem. Phys. 2010, 12 (42), 13962−13974. (39) Gutierrez-Sanchez, C.; Ciaccafava, A.; Blanchard, P. Y.; Monsalve, K.; Giudici-Orticoni, M. T.; Lecomte, S.; Lojou, E. Efficiency of Enzymatic O2 Reduction by Myrothecium verrucaria Bilirubin Oxidase Probed by Surface Plasmon Resonance, PMIRRAS, and Electrochemistry. ACS Catal. 2016, 6 (8), 5482−5492. (40) Tominaga, M.; Ohtani, M.; Taniguchi, I. Gold single-crystal electrode surface modified with self-assembled monolayers for electron tunneling with bilirubin oxidase. Phys. Chem. Chem. Phys. 2008, 10 (46), 6928−6934. (41) Batra, B.; Lata, S.; Rana, J.; Pundir, C. S. Construction of an amperometric bilirubin biosensor based on covalent immobilization of bilirubin oxidase onto zirconia coated silica nanoparticles/chitosan hybrid film. Biosens. Bioelectron. 2013, 44, 64−69. (42) Chen, J.-P.; Wang, H.-Y. Improved properties of bilirubin oxidase by entrapment in alginate-silicate sol-gel matrix. Biotechnol. Tech. 1998, 12 (11), 851−853. (43) Xia, H.-q.; Kitazumi, Y.; Shirai, O.; Ozawa, H.; Onizuka, M.; Komukai, T.; Kano, K. Factors affecting the interaction between carbon nanotubes and redox enzymes in direct electron transfer-type bioelectrocatalysis. Bioelectrochemistry 2017, 118, 70−74. (44) Read, M. J.; Burkett, S. L. Asymmetric α-helicity loss within a peptide adsorbed onto charged colloidal substrates. J. Colloid Interface Sci. 2003, 261 (2), 255−263. (45) Shaw, W. J.; Ferris, K.; Tarasevich, B.; Larson, J. L. The structure and orientation of the C-terminus of LRAP. Biophys. J. 2008, 94 (8), 3247−3257. (46) Jones, T. T.; Fernandez, E. J. α-Lactalbumin tertiary structure changes on hydrophobic interaction chromatography surfaces. J. Colloid Interface Sci. 2003, 259 (1), 27−35. (47) Sorci, M.; Dassa, B.; Liu, H.; Anand, G.; Dutta, A. K.; Pietrokovski, S.; Belfort, M.; Belfort, G. Oriented covalent immobilization of antibodies for measurement of intermolecular binding forces between zipper-like contact surfaces of split inteins. Anal. Chem. 2013, 85 (12), 6080−6088. (48) Srinivasan, K.; Banerjee, S.; Parimal, S.; Sejergaard, L.; Berkovich, R.; Barquera, B.; Garde, S.; Cramer, S. M. Single Molecule Force Spectroscopy and Molecular Dynamics Simulations as a Combined Platform for Probing Protein Face-Specific Binding. Langmuir 2017, 33 (41), 10851−10860. (49) Ding, B.; Jasensky, J.; Li, Y.; Chen, Z. Engineering and characterization of peptides and proteins at surfaces and interfaces: A case study in surface-sensitive vibrational spectroscopy. Acc. Chem. Res. 2016, 49 (6), 1149−1157. (50) Zheng, J.; Li, L.; Chen, S.; Jiang, S. Molecular simulation study of water interactions with oligo (ethylene glycol)-terminated alkanethiol self-assembled monolayers. Langmuir 2004, 20 (20), 8931−8938. (51) Wei, S.; Brooks, C. L., III Stability and orientation of cecropin P1 on maleimide self-assembled monolayer (SAM) surfaces and suggested functional mutations. Chin. Chem. Lett. 2015, 26 (4), 485− 490. (52) Zhao, J.; Wang, Q.; Liang, G.; Zheng, J. Molecular dynamics simulations of low-ordered Alzheimer β-amyloid oligomers from dimer to hexamer on self-assembled monolayers. Langmuir 2011, 27 (24), 14876−14887. (53) Wang, Q.; Wei, S.; Wu, J.; Zou, X.; Sieggreen, O.; Liu, Y.; Xi, C.; Brooks, C. L., III; Chen, Z. Interfacial Behaviors of Antimicrobial Peptide Cecropin P1 Immobilized on Different Self-Assembled Monolayers. J. Phys. Chem. C 2015, 119 (39), 22542−22551. K

DOI: 10.1021/acs.langmuir.8b01974 Langmuir XXXX, XXX, XXX−XXX