The Role of Ninth Type-III Domain of Fibronectin in the Mediation of

Subscriber access provided by UNIV OF DURHAM

New Concepts at the Interface: Novel Viewpoints and Interpretations, Theory and Computations

The Role of Ninth Type-III Domain of Fibronectin in the Mediation of Cell-binding Domain Adsorption on Surfaces with Different Chemistry Tianjie Li, Lijing Hao, Jiangyu Li, Chang Du, and Yingjun Wang Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b01937 • Publication Date (Web): 25 Jul 2018 Downloaded from http://pubs.acs.org on August 2, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

The Role of Ninth Type-III Domain of Fibronectin in the Mediation of Cell-binding Domain Adsorption on Surfaces with Different Chemistry Tianjie Li, †, ‡ Lijing Hao, †, ‡ Jiangyu Li, ⊥ Chang Du, *, †, ‡, § Yingjun Wang *, †, ‡ †

Department of Biomedical Engineering, School of Materials Science and Engineering, South

China University of Technology, Guangzhou 510641, PR China. ‡

National Engineering Research Center for Tissue Restoration and Reconstruction, South China

University of Technology, Guangzhou 510006, PR China. ⊥

Department of Mechanical Engineering, University of Washington, Seattle 98195, WA, USA.

§

Key Laboratory of Biomedical Materials and Engineering, Ministry of Education, South China

University of Technology, Guangzhou 510006, PR China.

KEYWORDS: fibronectin, protein adsorption, molecular dynamic simulation

ABSTRACT: The orientation and conformation of adhesive proteins after adsorption play a central role in cell-binding bioactivity. Fibronectin (Fn) holds two peptide sequences that favor the cell adhesion: the RGD loop on the tenth type-III domain (Fn-III10) and the PHSRN synergy

ACS Paragon Plus Environment

1

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 38

site on the ninth (Fn-III9). Herein, Fn fragments (Fn-III10 and Fn-III9-10) adsorptions on selfassembled monolayers (SAMs) carrying various functional groups (-COOH, -NH2, -CH3 and OH) were investigated by Monte Carlo method and molecular dynamic simulation in order to understand their mediation effect on cell adhesion. The results demonstrated that Fn-III9 could enhance the stiffness of Fn molecule and further fix the adsorption orientation. The RGD site of Fn fragment appeared to be deactivated on hydrophobic surfaces (CH3-SAM) due to the binding of adjacent nonpolar residues on surfaces, whereas charged surfaces (COOH-SAM and NH2SAM) and hydrophilic surfaces (OH-SAM) were conducive to the formation of cell-bindingfavorable orientation for Fn fragments. The cell adhesion capability of Fn fragments was highly improved on positively-charged surfaces (NH2-SAM) and hydrophilic surfaces because of the advantageous steric structure and orientation of both RGD and PHSRN sites. This work will provide insight into the interplay at atomic scale between protein adsorption and surface chemistry for designing biologically responsive substrate surfaces.

Introduction The interaction with extracellular matrix (ECM) proteins of biomaterials surface is crucial to the subsequent cell behaviors.1 When biomaterials are implanted into patient bodies, the adhesive proteins from ECM will be first to adsorb onto the biomaterials surface with a certain orientation prior to cell adhesion.2-3 The orientation and structural completeness of adsorbed proteins and the protein-material interactions decide the biofunctional validity of the protein and thus mediate the cytocompatibility of the biomaterial. The adsorption behavior of protein is affected by many factors like biomaterial surface chemistry,2, 4 protein species,4 the ionic strength and pH of the


2

Page 3 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

solution,5 etc. Therefore, the understanding of the regulation mechanism for protein adsorption is the lighthouse for designing biomaterials with expected cell response. Fibronectin (Fn) is one of the essential multifunctional ECM proteins, which holds two key peptide sequences contributing to cell adhesion on adjacent domains:6-7 RGD tripeptide sequence on the tenth type-III domain (Fn-III10) and PHSRN synergy sequence on the adjacent ninth typeIII domain (Fn-III9). The RGD tripeptide engages in the binding of most integrins like αvβ3 on the cell membrane, thus mediates cell adhesion and migration.8 The mobility, orientation and secondary structure of RGD site are crucial to the cell adhesion capability of Fn.9-10 Alternatively, some other integrins like α5β1 require engagement to both RGD and PHSRN sites with definite orientation and spatial structure to bind.11-12 However, the exact molecular mechanisms of interaction between PHSRN site, even the entire Fn-III9 domain, and the RGD site requires identification to better understand the cell adhesion on Fn matrix for different biomaterial surfaces. Unfortunately, the effect of substrate surface chemistry on the interaction between both cell-binding sites is still vague. Based on repeated random sampling, Monte Carlo (MC) method had been applied in the complicated calculations for numerous biosystems13-15 due to excellent accuracy and speed. Compared with MC, molecular dynamic (MD) simulation demonstrates more practical and detailed features for biomolecule kinetics. Specifically, MD simulation is well-suited for the study of the molecular mechanism on biointerfaces like cell membrane,16 protein-materials interface,17-19 etc. Herein, the biofunctional role of Fn-III9 during the adsorption on the selfassembled monolayers (SAMs) with different surface chemistry was studied by MC and MD simulation successively. The SAMs with different terminal groups served as biomaterial surface models with different surface chemistry. Fn-III10 and Fn-III9-10 all-atom molecular models were


3

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 38

chosen for comparison to investigate the biofunction of Fn-III9. The final orientation, mobility and conformation variation of the adsorbed Fn fragments and the cell-binding sites were studied, as well as the interactions between protein molecules and the SAMs surface. Further statistical analysis of adsorption was carried out to forecast the cell adhesion capability of Fn-III10 and FnIII9-10. The unique biofunction of Fn-III9 in the mediation of cell adhesion on different surface chemistry was properly predicted.

Experimental method Fibronectin segment all-atom molecule. The Fn-III10 all-atom molecular model was from RCSB protein data bank (PDB: 1TTF20) with a net charge at 0. The Fn-III9-10 all-atom molecular model was derived from the Fn-III7-10 molecule (PDB: 1FNF21) by cutting off the peptide bond between Thr1325 and Gly1326, with a negative net charge at -1. As shown in Figure 1, Fn-III10 holds an RGD tripeptide loop near the C-terminal, which located between β-strand F (6th) and G (7th). Fn-III9 holds a PHSRN synergy sequence. The direct distance between RGD loop and PHSRN site is about 30-40 Å and a small rotation between Fn-III9 and Fn-III10 orients the two sites on the same side of Fn molecule.21 Each of the domains contains seven β-strands forming two antiparallel β-sheet layers, of which one contains three and the other consists of four. The orientation angle (θ) was defined as the angle between the normal vector and the dipole moment of a protein molecule and represented as cosine form to describe the orientation of protein molecule. Monte Carlo pre-optimization. To enhance the efficiency of the adsorption process, Monte Carlo pre-optimization was carried out. Electrostatic force is regarded as the driving force for


4

Page 5 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

protein adsorption in the preliminary stage, so the systems were divided into three groups according to the charge of the material surface, i.e. negatively charged (COOH), positively charged (NH2) and non-charged (CH3 and OH). Three endless planes representing each surface was applied to simplify the calculation. The Fn-III10 and Fn-III9-10 molecules were simplified into coarse-grained models by merely remaining the α-carbons on the protein backbone. Only Van der Waals interaction and electrostatic interaction were considered in the MC calculation by using the empirical equations. The equations and corresponding parameters were from Zhou’s work.22-23 The coarse-grained protein molecules were kept rigid during the calculations. The system temperature was adjusted to 300 K. At the beginning of each system, the protein molecule was placed above the material surface plane with a distance of 10 nm. 80,000,000 MC circulations had been conducted for each system by transferring and rotating the protein molecule around the center of mass. The optimized statistic orientation distribution was acquired from the last 40,000,000 MC circulations while the first 40,000,000 circulations were for equalization. The optimized orientations were served as the initial orientation of subsequent MD simulations. SAM surfaces for MD simulation. The SAM surface models with the terminal groups of COOH, -NH2, -CH3 and -OH were constructed by HS(CH2)10COOH, HS(CH2)10NH2, HS(CH2)10CH3 and HS(CH2)10OH, respectively. The (√3 × √3)R30° structures of each alkanethiol on Au(111) were adopted and donated as COOH-SAM, NH2-SAM, CH3-SAM, and OH-SAM accordingly. The scale was adjusted to fit protein size (Table 1). To investigate the mediation of surface charge, a certain number of the terminal groups were protonated randomly for COOH-SAM, representing a surface charge density of around 0.05


5

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 38

C/m2. Similarly, the same number of terminal groups were deprotonated randomly for NH2-SAM (Table 1). The parameters of SAM molecules were derived from the CHARMM36 force field.24 Molecular dynamic simulation. All the sulfur atoms of SAM models were fixed during the MD simulation. Water molecules described by the TIP3P model25 were added to the simulated boxes for solvation. Chlorine and sodium ions were added to keep the system charge neutral. The Fn fragment molecule was placed above the SAM surface by 0.5 nm with the identified orientations from MC calculation. All the MD simulation and subsequent analysis were performed by employing the all-atom model of Fn fragment molecules. The sizes of the simulation boxes and the total atoms of each system were listed in Table 1. The simulations were performed in a canonical ensemble with a time step of 2 fs, and the temperature was controlled at 300 K by a Nosé-Hoover thermostat26 using a coupling time of 0.5 ps. The initial velocity of each atom was assigned from a Maxwell-Boltzmann distribution at 300 K. The bonds containing hydrogen atoms were constrained by the LINCS algorithm.27 The nonbonded interactions were calculated by a switched potential with a switch function starting at 9 Å and reaching zero at 10 Å. Electrostatic interactions were calculated by the particle mesh Ewald method in 3dc geometry28 with a cutoff distance of 12 Å. Periodic boundary conditions were applied only in the x and y directions. Two hard walls were set at both the top and the bottom of the simulation box as implemented in the GROMACS 5.1.2 package.29 All the parameters are from the CHARMM36 force field.24 The systems were first minimized using the steepest descent method to eliminate the steric overlap or inappropriate geometry. Then, a 100 ps NVT equilibration with a position restraint on heavy atoms of the protein was conducted for each system to equilibrate the solvent and ions


6

Page 7 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

around the protein. Finally, a 60 ns MD simulation was performed for each system. For structure visualization, the Visual Molecular Dynamics (VMD) program30 was utilized. The g_mmpbsa tool31-32 was used to analyze the Fn-SAM interfacial free energies (without polarization) during the final stage. The secondary structure of the protein molecules was calculated by dssp tools.3334

The analysis of protein properties, including root mean square deviation (RMSD) of protein

backbone, root mean square fluctuation (RMSF) and solvent-accessible surface areas (SASA) of residues, the gyrate radius and the dipole moment of protein molecules and the hydrogen bonds generated between proteins and surfaces, were all carried out by the use of GROMACS 5.1.2 package.29

Results and discussion When cells approach the biomaterial surface, the integrins on the cell membrane build a bridge with identified peptide sequences for cell-cell or cell-ECM interactions.8, 11 RGD tripeptide is identified as a common binding sequence of cellular integrins,8 whereas the proper position of PHSRN peptide could lead to a more advantageous binding with integrins by adding the type of attachable integrins.12 Also, RGD commonly demonstrates a flexible loop conformation.35 Less constrained loop conformation and more mobility would boost cell adhesion.9 Therefore, the orientation mobility and conformation of RGD and PHSRN are commonly regarded as important factors for the evaluation of cell adhesion capability. Initialization with MC pre-optimization. The preferable adsorption orientations of Fn-III10 and Fn-III9-10 molecules were first investigated by MC method as shown in Figure 2. The cosθ of Fn segment is negative on the negatively charged surface, which is the opposite on the positively


7

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 38

charged one. Electrostatic interaction acted as a dominated role in both processes. The orientation distribution of Fn segments on the non-charged resembled the case on the charged surfaces. Since electrostatic interaction was neglected in the non-charged system, Van der Waals (VdW) interaction controlled the adsorption when the protein molecules were close to the surface. The orientation distribution peaks were doublet for Fn-III10 but unimodal for Fn-III9-10. The doublet peaks indicated the alternative possible adsorption orientations, which reflected an unstable adsorption state. This unstable state could be attributed to the round shape of Fn-III10 molecule, which has a flat surface on both sides of Arg93. As a result, Fn-III9 fixed the adsorption orientation of Fn fragment molecule, orienting the cell-binding sites in a definite position. Interactions on Fn-SAM interface. After 60 ns MD simulation, the RMS deviation (RMSD) of protein backbone was calculated to evaluate the stability of the final systems (Figure 3A). Whichever initial orientation was applied, the protein backbone of Fn-III10 and Fn-III9-10 molecules remained complete and steady (RMSD < 0.6 nm) throughout the adsorption process in all systems. All the RMSD curves enter a relatively even state from 55 ns, which indicated the adsorption ended with a stable state. The last-5-ns period was identified as the final stage and used for the subsequent analysis and statistics. The electrostatic interaction, VdW interaction and hydrogen bonds between Fn-III10 and SAMs surface were investigated (Figure 3B&C). Electrostatic interaction dominated the adsorption on the charged surfaces, whereas VdW interaction accounted for the majority of the interactions on the non-charged ones. More hydrogen bonds were formed with charged surface


8

Page 9 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

rather than non-charged surface and none was formed with hydrophobic surface. Fewer hydrogen bonds in O2a-NH2 indicated larger contact area in O3a-NH2, which could be resulted from the more parallel initial orientation to surface. The stronger interaction anchored the protein with such advantageous orientation and was expected to promote the subsequent cell adhesion. The Fn-SAM interactions for Fn-III9-10 were also illustrated in Figure 3B&C. For COOHSAM, Fn-III9-10 generated fewer hydrogen bonds since it tended to be repelled from the surface for the like charge. As shown below, Fn-III9 created a spatial separation for RGD and freed it. Interestingly, the negative electrostatic energies implied that the protein molecule was approaching the surface despite the like charge. This phenomenon confirmed the polarity of FnIII9-10 molecules. Fn-III9-10 was preferably adsorbed on OH-SAM with the Fn-III10 domain keeping the same orientation as in the single system (Figure 4) due to the aggregation of polar residues, causing weak repulsive electrostatic interaction in O2b-OH. Although OH-SAM was denoted as a noncharged surface, a number of the hydroxyl groups at the top would be deprotonated to turn the whole surface into slightly negatively charged. The hydrogen bonds formed between Fn-III10 and OH-SAM surface neutralized the effect of the repulsive electrostatic interaction. For CH3-SAM, all the systems showed negligible electrostatic interaction because of the nonpolar methyl groups. Adsorption orientation. Statistics of the final orientation distribution in the final stage was demonstrated to obtain the accurate final orientation of the protein molecules in Figure 4B. The orientation distribution peak of the Fn-III10 aggregated in the vicinity of 0 on the non-charged surface (CH3 and OH), which could be roughly categorized into two clusters according to the peak value (one consisted of O1a-CH3, O3a-CH3, and O1a-OH, whereas the other consisted of O2a-


9

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 38

CH3, O2a-OH, and O3a-OH). The adsorption distribution of Fn-III9-10 followed a similar trend as Fn-III10. Since MD simulations take account of intramolecular force and molecular deformation, Fn-III10 adsorbed on the surface with an orientation irrelevant to the initial settings. In the later period of adsorption, VdW interaction became in charge of the adsorption. In addition, the doublet in O2b-OH distribution curve indicated the alternative preferable adsorption orientations, as well as the weak interaction between protein molecule and hydrophilic surface. The analysis of visualized systems at the final stage further indicated the details of binding site and possible cell-binding capability of the protein molecules as shown in Figure 4A&C. The cell-binding capability is largely affected by the orientation of RGD tripeptide,10 so the adsorption orientation was named after the position of RGD on the protein molecule relative to the surfaces. Both Fn-III10 and Fn-III9-10 molecules remained a similar orientation after MD simulation as the original orientation (Figure 4B) on charged surfaces. The Fn-III10 molecule was only adsorbed on NH2-SAM surface with an advantageous orientation (“End-on”) due to the negatively charged residue patch (Asp7, Glu9, Asp23) on the opposite side to the RGD loop,36 whereas the RGD was restrained on the COOH-SAM surface with an adverse orientation (“Head-on”). It is interesting to note that the Fn-III9-10 molecule was adsorbed on COOH-SAM with an advantageous orientation (“End-on”). Fn-III9 provided a space between the Fn-III10 and SAM surface to free the RGD site. On NH2-SAM, the Fn-III9-10 molecule was in a favorable final orientation (“End-on”) with both RGD and PHSRN free. The side held the RGD and PHSRN sites stayed upwards, which was likely to activate the binding of other kinds of integrins like α5β1.11-12 Altering the cellular integrin species could increase the global amount of attachable integrins and improved the possibility of cell adhesion on Fn matrix.


10

Page 11 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

On the non-charged surfaces, either hydrophobic or hydrophilic, the adsorption orientations of both Fn-III10 and Fn-III9-10 molecules in Figure 4B were different from the initial orientation. Specifically, The Fn-III10 molecules were adsorbed on the CH3-SAMs and OH-SAMs surfaces all with a “Side-on” orientation. The cell adhesion capabilities of “Side-on” orientation is difficult to evaluate since the RGD has the possibility to be caught by the material surface. Therefore, the “Side-on” orientation is speculated to own low cell adhesion capabilities. Presumably, this phenomenon in hydrophobic system (CH3-SAM) was mainly caused by the hydrophobic interaction of the nearby nonpolar residue Pro1497 (Pro82 of Fn-III10) with surface. On the contrary, the polar residue aggregated patch on the same side (Thr14, Thr16, Ser17, Ser43, Gln46, Lys54, Thr58, Ser60, Gly61 and Lys63) contributed to the “Side-on” orientation in hydrophilic system (OH-SAM), which was confirmed by the interaction distribution map in Figure 7. For Fn-III9-10, the molecule was adsorbed on the CH3-SAM with even worse orientation (“Front-on”). The RGD was completely hidden underneath the protein molecule and touched the surface, which was likely to impede cell adhesion. But for hydrophilic system (OH-SAM), the Fn-III9-10 molecule ended with a beneficial orientation (“End-on”). Both cell-binding sites were unrestrained, which may promote the possibility of cell adhesion. But the wide distribution peak (Figure 4B) indicated the weak interaction between Fn-III9-10 and surface. The adsorption was vulnerable and likely to be detached, which would result in the reduction of cell-binding sites. Adsorption conformation. The secondary structure was studied to evaluate the biofunctional validity of the protein after adsorption as demonstrated in Figure 5. In general, all the Fn-III10 molecules kept the number of β-strands due to their stiff structure from high strain energies.37 The antiparallel β-sheet layers reinforce the protein structure. But there were helices formed in O3a-NH2 and O2a-OH, which indicated the possible denaturation of the protein. The helices may


11

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 38

derive from the intense interaction from the huge contacting area with surfaces and may influence the bioactivities. For Fn-III9-10, no helix was formed. Fn-III9 reinforced the structural stiffness of the Fn fragment molecule. Compared to Fn-III10, Fn-III9-10 molecule has two Fn-III domains, which means more antiparallel β-sheet layers making the segment stiff enough. The conformational completeness ensured the original bioactivities throughout the adsorption process. RGD mobility. Compared to Fn-III10, Fn-III9-10 presented a more flexible status according to the RMSF analysis (Figure 6). Particularly, the RGD loop showed greater RMSF on the Fn-III910,

which indicated enhanced mobility to promote cell adhesion capability.9 Each residue

contribution to the attractive Fn-SAM interaction for Fn segments was studied in Figure 7. The contribution to attraction was related to the residues engaging in the binding sites (Figure 4A&C). The mobility and activated status of RGD was evaluated by summarizing from the responding adsorption and conformation analysis. The secondary structure changes of both cell-binding sites varied slightly from system to system (Figure 8). Bend is used for regions of high curvature.38-39 It is a similar loop structure to turns (β-turns), which features hydrogen bonds typical of helices.38-39 The RGD of almost all the Fn-III10 maintained the loop structure, which was advantageous for cell binding. For Fn-III9-10, The RGD also kept the loop structure during adsorption. But the flank residues of RGD in almost all the systems split to coils, which improved the mobility of the RGD sites. Presumably, the destruction of the flank turn structure resulted from the stretching and pressing interaction between the two domains. The enhancement in mobility confirmed the RMSF results and would benefit the subsequent cell binding activities.


12

Page 13 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

The residues may contribute to the binding with SAM surface were labeled in Figure 4A&C, the analysis for the contribution of each residue to the attractive protein-SAM interaction was carried out afterwards. Specifically, Fn-III10 was adsorbed on COOH-SAM mainly with two positively charged residues (Arg30 and Arg78), whereas the adsorption was based on four negatively charged residues (Asp3, Asp7, Glu9, and Asp67) on NH2-SAM. RGD was deactivated for the former, and valid for the latter. The “Side-on” orientation on CH3-SAM and OH-SAM provided a less flexible status of RGD. Fn-III9 created a spatial separation for Fn-III10. The bound with COOH-SAM involved the positively charged residues (Arg1351, Arg1403) on Fn-III9 instead of the arginine on the RGD. The unrestrained RGD was possible for the cellular integrins to bind and promoted cell adhesion afterward. Also, the presence of Fn-III9 screwed the molecule on OH-SAM to improve the mobility of RGD. Promisingly, Fn-III9 could favor the cell adhesion capability on COOH-SAM, NH2-SAM, and OH-SAM with unrestrained RGD and PHSRN sites. For CH3-SAM, the interaction allocation of O2a-CH3 was different from either of the other two systems. The attractive interaction mainly focused on the central part of Fn-III10 for O2a-CH3 (cosθ < 0), whereas on both ends for O1a-CH3 and O3a-CH3 (cosθ > 0). RGD (Res 78-80) was partly restrained by the adjacent nonpolar residue Pro82 (Pro1497 of Fn-III9-10). Therefore, the adsorption orientation of O2a-CH3 was expected to be more favorable for cell adhesion compared to the others. As for OH-SAMs, the Fn-III10 molecules were adsorbed with the middle section, which may promote the cell adhesion capability to a similar level as O2a-CH3. Particularly, FnIII9-10 was adsorbed on CH3-SAM surface by the C-terminal (including the RGD tripeptide). The contribution diagram confirmed the deactivated status of RGD site as well as the possible inhibition of subsequent cell adhesion.


13

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 38

Fn-III9 changed the adsorption orientation of Fn fragment, which may alter the cell adhesion capability. In this study, the RGD was restrained on the negatively charged surface (COOHSAM) for Fn-III10, but away from the surface for Fn-III9-10. Fn-III9 provided steric separation for Fn-III10 with surfaces. For positively charged surface (NH2-SAM), both unrestrained RGD and PHSRN sites on Fn-III9-10 showed a more favorable orientation possibly to promote cell adhesion. For hydrophobic surfaces (CH3-SAM), Fn-III9 generated additional hydrophobic interaction with the surface. With the help of adjacent proline, the RGD was pressed to be hidden underneath. Similarly, hydrophilic surface (OH-SAM) showed a neutral orientation distribution as hydrophobic surfaces for Fn-III10 (Figure 4B). Fn-III9 lifted up the RGD site and probably endowed the Fn fragment with better cell adhesion capability in spite of low adsorption amount. Briefly, Fn-III9 domain plays an irreplaceable role in mediating the cell adhesion. But the effect of Fn-III9 for the entire Fn molecule is still not identified yet. Although the all-atom structure in high resolution is only obtained for Fn-III7-10 segment rather than the entire Fn, it is commonly regarded as an appropriate alternative to study cell adhesion properties owing to the possession of both key cell-binding sites.40-41 So as to simplify calculation and directly study the role of Fn-III9 in cell adhesion capabilities, the Fn-III9-10 fragment was cut from the Fn-III7-10 molecule with both cell-binding sites remained and used as an alternative. Expectedly, Fn-III9-10 obtained a quite similar adsorption orientation and conformation as Fn-III7-10 simulation (unpublished results). Except for the Fn-III domains, there are two other kinds of domains on Fn: Fn-I and Fn-II. It is not yet possible to simulate these two domains due to the lack of all-atom structure. However, since both Fn-I and Fn-II domains are mechanically stable as stabilized by disulfide bonds,42 it is predicted that the role of Fn-III9 would be similar for the entire Fn molecule.


14

Page 15 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Collectively, the cell adhesion capability of Fn-adsorbed SAMs can be predicted from MD simulation of the cell-binding domain adsorption, which follows the trend: positively charged surfaces (NH2) > negatively charged surfaces (COOH) > hydrophilic surfaces (OH) > hydrophobic surfaces (CH3). This trend was consistent with the cell adhesion and proliferation performance in experimental works43-45 and our cell culture study (unpublished result). It should be noted that the real cell culture conditions are much more complicated and involve multiple proteins and protein-protein interaction. The simulation of the multiple protein system deserves further investigation. Within the scope of current study, the synergistic effect of two functional domains in a single Fn molecule provide some insight on the molecular mechanism of cellular interaction with this typical cell adhesion protein. It has been known that the whole Fn molecule is negatively charged43, and the number of adsorbed Fn and the binding force on COOH and OH surfaces were lower than those on NH2 and CH3 surfaces.44-45 The Fn-adsorbed NH2 surface is favorable for cell migration and proliferation due to advantageous orientation of cell-binding sites and high stability.44 In contrast, the cell migration and proliferation was prohibited on CH3SAM, which would partly due to the adverse orientation of cell-binding sites as described in Figure 4.

Conclusions Overall, Fn-III9 is crucial to the cell adhesion capability of the adsorbed Fn molecule on biomaterial surfaces. Fn-III9 reinforced the orientation of the adsorbed protein as well as the cellbinding sites to ensure a more stable cell adhesion capability.


15

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 38

Fn-III9 is crucial for the adsorption orientation of Fn-III10 fragment molecule. The RGD site was freed from the negatively charged surface by Fn-III9 and endowed the Fn fragment with better cell-binding capability. Both cell-binding sites of adsorbed Fn-III9-10 were flexible on positively charged surfaces to benefit the cell adhesion, whereas the two sites were bound on hydrophobic surfaces to impede the cell adhesion. Fn-III9 is likely to improve the cell adhesion capability of Fn fragments on hydrophilic surfaces in spite of the weak protein-surface interaction. Additionally, Fn-III9 can reinforce the conformation of the Fn molecule to protect it from denaturation and improve the mobility of RGD sequence. Therefore, Fn-III9 enhanced the stability of the protein molecule and facilitate cell adhesion. This work will provide insight into the interplay at atomic scale between protein adsorption and surface chemistry for designing biologically responsive substrate surfaces. It is also expected to provide fundamentals to further shed lights on the adsorption pattern and biofunction of the entire Fn.


16

Page 17 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 1. Fn-III10 (PDB: 1TTF) and Fn-III9-10 molecule (cut from the Fn-III7-10 molecule, PDB: 1FNF). The black arrows stand for the electric dipole moment of each protein; The purple arrows stand for the normal vector. θ stands for the orientation angle (i.e. the angle between the electric dipole moment and the normal vector). Protein molecules are shown in NewCartoon mode. Charged residues are appeared in Licorice mode, with red representing negatively charged and blue representing positively charged.

Figure 2. Statistic orientation distribution of (A) Fn-III10 and (B) Fn-III9-10 molecules adsorbed on positively charged, negatively charged and neutral surface in MC calculation.


17

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 38

Figure 3. (A) RMS deviation of Fn-III10 and Fn-III9-10 protein backbone. The last-5-ns period of MD simulation was identified as the final balanced stage for statistics. (B) The interface energies and (C) the number of hydrogen bonds of Fn-III10 and Fn-III9-10 molecules with SAMs surface during the final stage.


18

Page 19 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 4. (A) Final orientation of Fn-III10 molecules after MD simulation. (B) Final orientation distribution of Fn-III10 and Fn-III9-10 molecules during the final stage of MD simulation. (C) Final orientation of Fn-III9-10 molecules after MD simulation. e stands for the electric dipole moment calculated from the all-atom model. RGD site is indicated in green transparent Surf mode. PHSRN synergy site is indicated in blue transparent Surf mode.


19

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 38

Figure 5. The secondary structure of the Fn-III10 and Fn-III9-10 molecules in the final stage. Newly formed helix structure on NH2 and OH-terminal SAM surface implied the possible denaturation of Fn-III10, whereas the slight change in the structure throughout the simulation implied the stability of Fn-III9-10.


20

Page 21 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 6. The RMS fluctuation of each residue on Fn-III10 and Fn-III9-10 molecules during the final stage. The green area represents the RMSF of RGD, whereas the blue one represents the RMSF of PHSRN.

Figure 7. The average contribution of each residue in the Fn-III10 and Fn-III9-10 molecules to the Fn-SAM interface attractive interactions during the final stage. The residue labels were demonstrated in colors according to the properties: red for residues with a negatively charged side chain; blue for residues with a positively charged side chain; green for residues with a noncharged polar side chain; yellow for residues with a non-polar side chain.


21

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 38

Figure 8. The secondary structure variation of Fn-III10 and Fn-III9-10 during adsorption process. The black rectangle marks the secondary structure of RGD and flank adjacent residues. The blue rectangle marks the secondary structure of PHSRN and flank adjacent residues.

Table 1. The scale of MD simulations.

Fn-III10 Total alkyl chains SAM

Plane dimension

surfaces Total protonated (or deprotonated) chains MD simulation box dimension Total atoms

Fn-III9-10

132 5.196 × 5.500 nm

320 2

10 (7.58%)

8.660 × 8.000 nm2 24 (7.50%)

5.196 × 5.500 × 8.000 nm3 8.660 × 8.000 × 10.000 nm3 ≈ 23,000

≈ 70,000

AUTHOR INFORMATION Corresponding Author


22

Page 23 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

*School of Materials Science and Engineering, South China University of Technology, Guangzhou 510641, China. E-mail addresses: [email protected] (C. Du), [email protected] (Y. Wang) Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes The authors declare no competing financial interest. ACKNOWLEDGMENT This work was financially supported by National Key R&D Program of China (2017YFC1105000), Science and Technology Planning Project of Guangdong Province (2017B030314008), National Natural Science Foundation of China (51572087, 31700823), Shenzhen Science and Technology Innovation Committee (JCYJ20170818160503855), Natural Science Foundation of Guangdong Province of China (2017A030310335) and the Fundamental Research Funds for the Central Universities (2017BQ001). ABBREVIATIONS Fn, fibronectin; SAMs, self-assembled monolayers; ECM, extracellular matrix; MC, Monte Carlo; MD, molecular dynamic; RMSD, root mean square deviation; RMSF, root mean square fluctuation; VdW, Van der Waals. REFERENCES (1) Wilson, C. J.; Clegg, R. E.; Leavesley, D. I.; Pearcy, M. J., Mediation of biomaterial–cell


23

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 38

interactions by adsorbed proteins: a review. Tissue engineering 2005, 11, 1-18. (2) Nel, A. E.; Mädler, L.; Velegol, D.; Xia, T.; Hoek, E. M.; Somasundaran, P.; Klaessig, F.; Castranova, V.; Thompson, M., Understanding biophysicochemical interactions at the nano–bio interface. Nature materials 2009, 8, 543-557. (3) Anselme, K., Osteoblast adhesion on biomaterials. Biomaterials 2000, 21, 667-681. (4) Faucheux, N.; Schweiss, R.; Lützow, K.; Werner, C.; Groth, T., Self-assembled monolayers with different terminating groups as model substrates for cell adhesion studies. Biomaterials 2004, 25, 2721-2730. (5) Jones, K. L.; O’Melia, C. R., Protein and humic acid adsorption onto hydrophilic membrane surfaces: effects of pH and ionic strength. Journal of Membrane Science 2000, 165, 31-46. (6) Kao, W. J.; Lee, D.; Schense, J. C.; Hubbell, J. A., Fibronectin modulates macrophage adhesion and FBGC formation: the role of RGD, PHSRN, and PRRARV domains. Journal of Biomedical Materials Research Part A 2001, 55, 79-88. (7) Mardon, H. J.; Grant, K. E., The role of the ninth and tenth type III domains of human fibronectin in cell adhesion. FEBS letters 1994, 340, 197-201. (8) Kimizuka, F.; Ohdate, Y.; Kawase, Y.; Shimojo, T.; Taguchi, Y.; Hashino, K.; Goto, S.; Hashi, H.; Kato, I.; Sekiguchi, K., Role of type III homology repeats in cell adhesive function within the cell-binding domain of fibronectin. Journal of Biological Chemistry 1991, 266, 3045-3051. (9) Liu, J.; Yang, X.-W.; Chen, W.-Q.; Tang, J.-G., Effect of disulfide bond on the conformation and anticoagulant activity of an Arg-Gly-Asp motif displayed on a mutant insulin protein framework. Letters in Peptide Science 2002, 9, 179-186. (10) Kimura, R. H.; Levin, A. M.; Cochran, F. V.; Cochran, J. R., Engineered cystine knot peptides that bind alphavbeta3, alphavbeta5, and alpha5beta1 integrins with low-nanomolar


24

Page 25 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

affinity. Proteins 2009, 77, 359-69. (11) Nagai, T.; Yamakawa, N.; Aota, S.-i.; Yamada, S. S.; Akiyama, S. K.; Olden, K.; Yamada, K. M., Monoclonal antibody characterization of two distant sites required for function of the central cell-binding domain of fibronectin in cell adhesion, cell migration, and matrix assembly. The Journal of cell biology 1991, 114, 1295-1305. (12) Kao, W. J.; Lee, D., In vivo modulation of host response and macrophage behavior by polymer networks grafted with fibronectin-derived biomimetic oligopeptides: the role of RGD and PHSRN domains. Biomaterials 2001, 22, 2901-2909. (13) Narambuena, C. F.; Varretti, F. O. S.; Ramirez-Pastor, A. J., Adsorption thermodynamics of two-domain antifreeze proteins: theory and Monte Carlo simulations. Physical Chemistry Chemical Physics 2016, 18, 24549-24559. (14) Jin, T.; Stanciulescu, I., Numerical simulation of fibrous biomaterials with randomly distributed fiber network structure. Biomechanics and modeling in mechanobiology 2016, 15, 817-830. (15) Yao, X.; Peng, R.; Ding, J., Cell–material interactions revealed via material techniques of surface patterning. Advanced Materials 2013, 25, 5257-5286. (16) Tu, Y.; Lv, M.; Xiu, P.; Huynh, T.; Zhang, M.; Castelli, M.; Liu, Z.; Huang, Q.; Fan, C.; Fang, H., Destructive extraction of phospholipids from Escherichia coli membranes by graphene nanosheets. Nature nanotechnology 2013, 8, 594-601. (17) Liu, J.; Liao, C.; Zhou, J., Multiscale Simulations of Protein G B1 Adsorbed on Charged Self-Assembled Monolayers. Langmuir 2013, 29, 11366-11374. (18) Peng, C.; Liu, J.; Zhou, J., Molecular simulations of Cytochrome c adsorption on a bare gold surface: insights for the hindrance of electron transfer. The Journal of Physical Chemistry C


25

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 38

2015, 119, 20773-20781. (19) Kohler, S.; Schmid, F.; Settanni, G., Molecular Dynamics Simulations of the Initial Adsorption Stages of Fibrinogen on Mica and Graphite Surfaces. Langmuir 2015, 31, 13180-90. (20) Dickinson, C. D.; Veerapandian, B.; Dai, X.-P.; Hamlin, R. C.; Xuong, N.-h.; Ruoslahti, E.; Ely, K. R., Crystal structure of the tenth type III cell adhesion module of human fibronectin. Journal of molecular biology 1994, 236, 1079-1092. (21) Leahy, D. J.; Aukhil, I.; Erickson, H. P., 2.0 Å crystal structure of a four-domain segment of human fibronectin encompassing the RGD loop and synergy region. Cell 1996, 84, 155-164. (22) 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, 3472-3478. (23) Zhou, J.; Zheng, J.; Jiang, S., Molecular simulation studies of the orientation and conformation of cytochrome c adsorbed on self-assembled monolayers. Journal of Physical Chemistry B 2004, 108, 17418-17424. (24) MacKerell, A. D.; Bashford, D.; Bellott, M.; Dunbrack, R.; Evanseck, J.; Field, M. J.; Fischer, S.; Gao, J.; Guo, H.; Ha, S. a., All-atom empirical potential for molecular modeling and dynamics studies of proteins. The journal of physical chemistry B 1998, 102, 3586-3616. (25) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L., Comparison of simple potential functions for simulating liquid water. The Journal of chemical physics 1983, 79, 926-935. (26) Hoover, W. G., Canonical dynamics: equilibrium phase-space distributions. Physical review A 1985, 31, 1695. (27) Hess, B.; Bekker, H.; Berendsen, H. J.; Fraaije, J. G., LINCS: a linear constraint solver for molecular simulations. Journal of computational chemistry 1997, 18, 1463-1472.


26

Page 27 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(28) Yeh, I.-C.; Berkowitz, M. L., Ewald summation for systems with slab geometry. The Journal of Chemical Physics 1999, 111, 3155-3162. (29) Abraham, M.; Van Der Spoel, D.; Lindahl, E.; Hess, B., the GROMACS development team. GROMACS User Manual version 5.0.7 2015, 5. (30) Humphrey, W.; Dalke, A.; Schulten, K., VMD: visual molecular dynamics. Journal of molecular graphics 1996, 14, 33-38. (31) Baker, N. A.; Sept, D.; Joseph, S.; Holst, M. J.; McCammon, J. A., Electrostatics of nanosystems: Application to microtubules and the ribosome. Proceedings of the National Academy of Sciences 2001, 98, 10037-10041. (32) Kumari, R.; Kumar, R.; Lynn, A., g_mmpbsa—A GROMACS Tool for High-Throughput MM-PBSA Calculations. Journal of Chemical Information and Modeling 2014, 54, 1951-1962. (33) Wakamura, K.; Hirokawa, K.; Orita, K., Dictionary of protein secondary structure: Pattern recognition of hydrogen‐bonded and geometrical features. Biopolymers 1983, 22, 2577–2637. (34) Joosten, R. P.; te Beek, T. A. H.; Krieger, E.; Hekkelman, M. L.; Hooft, R. W. W.; Schneider, R.; Sander, C.; Vriend, G., A series of PDB related databases for everyday needs. Nucleic Acids Research 2011, 39, D411-D419. (35) Yamada, T.; Matsushima, M.; Inaka, K.; Ohkubo, T.; Uyeda, A.; Maeda, T.; Titani, K.; Sekiguchi, K.; Kikuchi, M., Structural and functional analyses of the Arg-Gly-Asp sequence introduced into human lysozyme. Journal of Biological Chemistry 1993, 268, 10588-10592. (36) Koide, A.; Jordan, M. R.; Horner, S. R.; Batori, V.; Koide, S., Stabilization of a fibronectin type III domain by the removal of unfavorable electrostatic interactions on the protein surface. Biochemistry 2001, 40, 10326-10333. (37) Panos, M.; Sen, T. Z.; Ahunbay, M. G., Molecular Simulation of Fibronectin Adsorption


27

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 38

onto Polyurethane Surfaces. Langmuir 2012, 28, 12619-12628. (38) Touw, W. G.; Baakman, C.; Black, J.; te Beek, T. A.; Krieger, E.; Joosten, R. P.; Vriend, G., A series of PDB-related databanks for everyday needs. Nucleic acids research 2014, 43, D364D368. (39) Kabsch, W.; Sander, C., Dictionary of protein secondary structure: pattern recognition of hydrogen‐bonded and geometrical features. Biopolymers 1983, 22, 2577-2637. (40) Wang, H.; He, Y.; Ratner, B. D.; Jiang, S., Modulating cell adhesion and spreading by control of FnIII7-10 orientation on charged self-assembled monolayers (SAMs) of alkanethiolates. Journal of biomedical materials research. Part A 2006, 77, 672-8. (41) Wilson, K.; Stuart, S. J.; Garcia, A.; Latour, R. A., Jr., A molecular modeling study of the effect of surface chemistry on the adsorption of a fibronectin fragment spanning the 7-10th type III repeats. Journal of biomedical materials research. Part A 2004, 69, 686-98. (42) Hynes, R. O.; Yamada, K. M., Fibronectins: multifunctional modular glycoproteins. The Journal of cell biology 1982, 95, 369-377. (43) Kristine, L.; Edward, P.; N., S. C.; A., C. L., Cell-Type-Specific adhesion mechanisms mediated by fibronectin adsorbed to chemically derivatized substrata. Journal of Biomedical Materials Research 1992, 26, 1343-1363. (44) Lin, M.; Wang, H.; Ruan, C.; Xing, J.; Wang, J.; Li, Y.; Wang, Y.; Luo, Y., Adsorption Force of Fibronectin on Various Surface Chemistries and Its Vital Role in Osteoblast Adhesion. Biomacromolecules 2015, 16, 973-984. (45) Lee, M. H.; Ducheyne, P.; Lynch, L.; Boettiger, D.; Composto, R. J., Effect of biomaterial surface properties on fibronectin–α5β1 integrin interaction and cellular attachment. Biomaterials 2006, 27, 1907-1916.


28

Page 29 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir


29

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 38

GRAPHIC ABSTRACT


30

Page 31 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 1. Fn-III10 (PDB: 1TTF) and Fn-III9-10 molecule (cut from the Fn-III7-10 molecule, PDB: 1FNF). The black arrows stand for the electric dipole moment of each protein; The purple arrows stand for the normal vector. θ stands for the orientation angle (i.e. the angle between the electric dipole moment and the normal vector). Protein molecules are shown in NewCartoon mode. Charged residues are appeared in Licorice mode, with red representing negatively charged and blue representing positively charged. 227x87mm (300 x 300 DPI)


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. Statistic orientation distribution of (A) Fn-III10 and (B) Fn-III9-10 molecules adsorbed on positively charged, negatively charged and neutral surface in MC calculation. 316x123mm (300 x 300 DPI)


Page 32 of 38

Page 33 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 3. (A) RMS deviation of Fn-III10 and Fn-III9-10 protein backbone. The last-5-ns period of MD simulation was identified as the final balanced stage for statistics. (B) The interface energies and (C) the number of hydrogen bonds of Fn-III10 and Fn-III9-10 molecules with SAMs surface during the final stage. 501x263mm (300 x 300 DPI)


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. (A) Final orientation of Fn-III10 molecules after MD simulation. (B) Final orientation distribution of Fn-III10 and Fn-III9-10 molecules during the final stage of MD simulation. (C) Final orientation of Fn-III910 molecules after MD simulation. e stands for the electric dipole moment calculated from the all-atom model. RGD site is indicated in green transparent Surf mode. PHSRN synergy site is indicated in blue transparent Surf mode. 1813x1019mm (72 x 72 DPI)


Page 34 of 38

Page 35 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 5. The secondary structure of the Fn-III10 and Fn-III9-10 molecules in the final stage. Newly formed helix structure on NH2 and OH-terminal SAM surface implied the possible denaturation of Fn-III10, whereas the slight change in the structure throughout the simulation implied the stability of Fn-III9-10. 476x240mm (300 x 300 DPI)


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6. The RMS fluctuation of each residue on Fn-III10 and Fn-III9-10 molecules during the final stage. The green area represents the RMSF of RGD, whereas the blue one represents the RMSF of PHSRN. 316x133mm (300 x 300 DPI)


Page 36 of 38

Page 37 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 7. The average contribution of each residue in the Fn-III10 and Fn-III9-10 molecules to the Fn-SAM interface attractive interactions during the final stage. The residue labels were demonstrated in colors according to the properties: red for residues with a negatively charged side chain; blue for residues with a positively charged side chain; green for residues with a non-charged polar side chain; yellow for residues with a non-polar side chain. 548x250mm (300 x 300 DPI)


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 8. The secondary structure variation of Fn-III10 and Fn-III9-10 during adsorption process. The black rectangle marks the secondary structure of RGD and flank adjacent residues. The blue rectangle marks the secondary structure of PHSRN and flank adjacent residues. 845x336mm (300 x 300 DPI)


Page 38 of 38

The Role of Ninth Type-III Domain of Fibronectin in the Mediation of

Recommend Documents