pH-Dependent Ordered Fibrinogen Adsorption on Polyethylene Single

Oct 24, 2016 - Nanostructured surfaces have the potential to influence the assembly as well as the orientation of adsorbed proteins and may, thus, str...
0 downloads 0 Views 5MB Size
Download PDF
Article pubs.acs.org/Langmuir

pH-Dependent Ordered Fibrinogen Adsorption on Polyethylene Single Crystals Christian Helbing,† Robert Stoeßel,† Dominik A. Hering,† Matthias M. L. Arras,† Jörg Bossert,† and Klaus D. Jandt*,†,‡ †

Chair of Materials Science (CMS), Department of Materials Science and Technology, Otto Schott Institute of Materials Research, Faculty of Physics and Astronomy, Friedrich Schiller University Jena, Löbdergraben 32, 07743 Jena, Germany ‡ Jena Center for Soft Matter (JCSM), Friedrich Schiller University Jena, Humboldtstraße 10, 07743 Jena, Germany S Supporting Information *

ABSTRACT: Nanostructured surfaces have the potential to influence the assembly as well as the orientation of adsorbed proteins and may, thus, strongly influence the biomaterials’ performance. For the class of polymeric (bio)materials a reproducible and well-characterized nanostructure is the ordered chain folded surface of a polyethylene single crystal (PE-SC). We tested the hypothesis that the trinodal-rodshaped protein human plasma fibrinogen (HPF) adsorbs on the (001) surface of PESCs along specific crystallographic directions. The PE-SC samples were prepared by isothermal crystallization in dilute solution and characterized by atomic force microscopy (AFM) before as well as after HPF adsorption at different concentrations and pH values. At a physiological pH of 7.4, connected HPF molecules, or e.g., fibrils, fibril networks, or sponge-like structures, were observed on PE-SC surfaces that featured no preferential orientation. The observation of these nonoriented multiprotein assemblies was explained by predominant protein−protein interactions and limited surface diffusion. However, at an increased pH of 9.2, single HPF molecules, e.g., spherical-shaped and trinodal-rod-shaped HPF molecules as well as agglomerates, were observed on the PE-SC surface. The presence of single HPF molecules at increased pH was explained by decreased protein− protein interactions. These single trinodal-rod-shaped HPF molecules oriented preferentially along crystallographic [100] and [010] directions on the PE-SC surface which was explained by an increased amount of intermolecular bonds along these crystallographic directions with increased surface atom density. The study established that HPF molecules can align on chemically homogeneous surface topographies one order of magnitude smaller than the dimension of the protein. This advances the understanding of how to control the assembly and orientation of proteins on nanostructured polymer surfaces. Controlled protein adsorption is a crucial key to improve the surface functionality of future implants and biosensors.

■

pography,15 and crystallinity14,16 as well as the adsorption conditions like pH value14,17,18 and concentration.16 The pH value during the adsorption is an important factor controlling the intra- and interelectrostatic interactions of protein molecules and facilitates protein−protein interactions.19 Protein adsorption has been studied on different materials surfaces such as metals,20 ceramics,21 and polymers.22 On the latter, most of the previous studies investigated the assembly of HPF on amorphous polymers, e.g., polystyrene, polybutadiene, polyurethane, and poly(vinyl chloride) with no specific surface nanostructure, and thus, no preferred orientation of the adsorbed proteins was observed.23−25 The improvement of polymeric biomaterials’ performance requires the investigation and understanding of the interactions between HPF and the nanostructured polymeric surfaces to control and align the HPF adsorption which allows a better control of cell attachment.26 One approach to create nanostructures on polymer surfaces is to utilize a self-organization mechanism, e.g., polymer

INTRODUCTION Protein adsorption is a scientific phenomenon of interest for biointerfaces, surfactants, emulsions, or ultrathin film applications.1,2 The use of materials in applications for life science is often limited by effects resulting from an undesired or unspecific adsorption of proteins on materials’ surfaces.3−6 An important protein is the amphiphilic glycoprotein human plasma fibrinogen (HPF) which further plays a major role in blood coagulation and is thus important for wound healing.7 A negative effect of undesired HPF adsorption is e.g. thrombosis caused by implant-associated blood coagulation.8 Single HPF molecules have a trinodal-rod-like structure formed by the two outer D-domains and the centered E-domain where the two αC-domains are bound9 with a native length of about 47.5 nm (for details see Supporting Information and Figure S1).10 There is thus a great interest to understand and control the adsorption mechanisms and assembly of proteins, especially HPF, on biomaterials surfaces. The assembly, i.e., the arrangement and conformation, of adsorbed proteins depends on the balance of protein−surface11 and protein−protein12 interactions. Parameters that can affect these interactions are the materials surface properties like chemistry,13,14 to© 2016 American Chemical Society

Received: August 22, 2016 Revised: October 21, 2016 Published: October 24, 2016 11868

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir crystallization.27 Polymer crystallization can induce the formation of nanoscaled lamellar crystals, consisting of chains with defined conformations and periodic distances.28 A previous study revealed that shape and size of crystalline polymer nanostructures directly affect and control the assembly of adsorbed proteins.16 A well-known crystalline polymeric biomaterial that forms surface nanostructures is polyethylene (PE). In biomedicine, ultrahigh-molecular-weight PE is used in total hip and knee endoprotheses where it slides against ceramic29 or metallic30 bearing partners. The lifetime of these endoprotheses is limited by the PE wear and friction behavior which can cause mechanical aseptic loosening of the implant.31 The PE wear can be reduced by adsorbed protein layers whereby the extent of the reduction depends on the type, amount, and conformation of the adsorbed proteins.32,33 Although the nanostructured topography (∼10 nm) of meltdrawn PE thin films was shown to induce orientation of adsorbed HPF molecules,16 a more fundamental PE nanostructure is the surface of polyethylene single crystals (PE-SC).27 PE-SC is an archetype of polymer folded chain crystals found in technical applicable PE medicine products. There, the PE-SC surface features a nanoscaled topography in the range of about 10 Å (for details see Supporting Information and Figure S2).34 The nanostructured surface is mainly formed by PE chain folds which align parallel to the PE-SC edges.35 Previous work showed that the PE-SC surface nanostructure induced an orientation of adsorbed polymers and metals such as rod-like microscopic tellurium crystals.36,37 The influence of the PESCs’ surface on the assembly of more complex molecules, e.g. biomolecules, however, was not investigated so far and is the focus of the present study. A further study investigated the HPF adsorption on highly oriented pyrolytic graphite (HOPG) layers.38 The small dimension of this surface structure (∼0.1 nm), determined by the specific arrangement of the carbon atoms, induced an isotropic HPF adsorption behavior on the HOPG layers. It was observed that HPF aligned along edges of different layers. The nanostructured PE-SC surface (∼1 nm) fills the gap between the nanostructured HOPG surface (∼0.1 nm) and the nanostructured melt-drawn PE thin film surfaces (∼10 nm). Therefore, we assume that a nanostructured PE-SC surface with a topography one order of magnitude smaller than the dimension of the protein has the potential for specific protein adsorption, i.e., fibrinogen. We tested the hypothesis that an increase of the pH value reduces protein−protein interactions and, thus, facilitates an oriented adsorption of single HPF molecules on nanostructured PE-SC surfaces along certain crystallographic directions of the PE-SC. The assembly of adsorbed HPF molecules onto nanostructured PE-SC was investigated by atomic force microscopy (AFM) in dependence of the concentration and pH value of the protein solution. The impact of protein−protein and protein−surface interactions on the assembly and orientation of the HPF molecules is discussed. We showed that HPF fiber and network formation on a hydrophobic polymeric surface depend on the pH. Furthermore, we found that single HPF molecules adsorb orientedly on a PE-SC surface; in contrast, HPF fibers did not show an orientation effect. This study was carried out to advance the understanding of protein assembly processes on biomaterials surfaces. Further, it should support the role of surface nanostructures to control the assembly and alignment of adsorbed proteins in future

applications. Controlled protein adsorption is relevant for the improvement of materials in contact with body fluids like implants and biosensors.

■

MATERIALS AND METHODS

Polyethylene Single Crystal Preparation. High-density polyethylene (HDPE) (Sigma-Aldrich, Schnelldorf, Germany) was dissolved in concentrations lower than 0.1 wt % at 120 °C in pxylene (synthesis grade, Merck KGaA, Darmstadt, Germany). According to the literature, the formation of PE-SC is induced by isothermal crystallization at a crystallization temperature (Tc) in these dilute solutions. An optimal Tc is in the temperature interval between 80 and 90 °C.39 Thus, we performed the isothermal crystallization at 85 °C for 7 days. The PE-SCs were transferred onto solid surfaces by placing drops (100 μL) of the crystallized polymer solution on pieces of a silicon wafer (1 cm × 1 cm). Subsequently, the solvent was evaporated in a desiccator. Protein Adsorption. Carbonate buffer solution (CBS) was prepared by mixing calcium carbonate (Merck KGaA Darmstadt, Germany) with Millipore water to a pH of 9.2. Phosphate buffered saline (PBS) solution (Biochrom AG, Merck KGaA, Darmstadt, Germany) with a pH of 7.4 was used as received. HPF (Calbiochem, Merck KGaA, Darmstadt, Germany) stock solutions (1 mg/mL) with PBS or CBS were prepared. For protein adsorption, they were subsequently diluted with PBS or CBS to concentrations of 0.5, 2, and 5 μg/mL. Prior to the protein adsorption, the PE-SC samples and the protein solutions were heated up to 37 °C. Subsequently, PE-SC samples were covered with 2 mL of the respective HPF solution and stored for 2 h at 37 °C in an oven to ensure quasi-physiological conditions during the adsorption. Afterward, the PE-SC samples were rinsed with the used buffer (PBS or CBS) and two times with Millipore water to remove the nonadsorbed HPF and residues of the used buffer. Finally, the substrates were dried in a stream of compressed air. The dried samples were characterized by AFM. Characterization of the Samples. The prepared PE-SC samples were characterized by tapping mode atomic force microscopy (AFM) before and after protein adsorption. AFM measurements in air were performed by using a Dimension 3100 and a MultiMode (both from Digital Instruments, Vecco, Santa Barbara, CA) equipped with a Nanoscope IV controller. Measurements were performed at room temperature by using standard tapping mode silicon cantilevers from Bruker (model RTESP, Vecco, Santa Barbara, CA) with a resonance frequency in the range of 315−364 kHz in air, a spring constant in the range of 20−80 N/m, and a typical tip radius of less than 10 nm (typical 7 nm). Topography and phase images of the PE-SC samples were recorded and analyzed. The mean root-mean-square roughness of the PE-SC surface before (Rq,PE‑SC) and after (Rq,CBS‑HPF) HPF adsorption as well as of the underlying silicon substrate (Rq,Si) was determined over an area of 500 nm × 500 nm (n = 5). Analysis of the Alignment of Adsorbed Fibrinogen. AFM height images of the PE-SC samples after protein adsorption were first converted into a binary form and then analyzed to evaluate the orientation of HPF molecules, agglomerates and fibers with respect to the PE-SC surface. The orientation was defined as the angle between the longitudinal axis of a protein structure and a reference axis of the PE-SC, which was chosen as the short diagonal B (Figure S2B). Here, the analysis was simplified by two simple geometric considerations: (i) The surface of a PE-SC can be divided in four sectors (I−IV), where the chain fold orientation in sectors I and III as well as sectors II and IV are similar (see Figure S2B). (ii) The HPF molecule is symmetric and trinodal-rod-like shaped. Thus, every orientation angle between 180° and 360° can be converted to an angle between 0° and 180°. Branches of fibers which were longer than a single molecule were counted as single fibers. The resulting angle distributions were plotted.

■

RESULTS AND DISCUSSION First, the PE-SC surface was characterized as a system for protein adsorption studies. In a second step, the assembly of 11869

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir

the detailed analysis in the literature, we expect an identical surface nanostructure for our PE-SCs. Subsequently, the PE-SC samples were used for the HPF adsorption experiments. After HPF adsorption the root-meansquared roughness (Rq,CBS‑HPF) increased to 0.87 ± 0.05 nm. In addition, trinodal-rod-shaped structures were observed on the surface of the PE-SC after HPF adsorption as displayed in the higher magnified inset of Figure 1b. The distinctive shape of these structures indicated the presence of single trinodal-rodshaped HPF molecules at the PE-SC surface. The imaging of single HPF molecules was enabled by the PE-SC low roughness (Rq,PE‑SC = 0.43 ± 0.02 nm) which is lower than the height of a single HPF molecule (hHPF = 1.4 ± 0.4 nm).16 The observation of distinctive trinodal-rod-shaped HPF molecules on PE-SC (Figure 1b) indicated that PE-SC surfaces are smooth enough to investigate protein−protein interactions at surfaces and protein−surface interactions in dependence of the adsorption conditions. In the following, we investigated the assembly of HPF at the surface of these smooth nanostructured hydrophobic PE-SC in dependence of the adsorption conditions, i.e., HPF concentration and buffer of the HPF solution. Influence of the Adsorption Conditions on the Assembly of HPF on PE-SC. HPF adsorption experiments were performed for varying adsorption conditions, i.e., HPF concentrations (0.5, 2.0, and 5.0 μg/mL) and buffer of the protein solution (PBS with pHPBS = 7.4 and CBS with pHCBS = 9.2) to evaluate the influence of protein−protein and protein− surface interactions on the HPF assembly on PE-SC surfaces. Representative AFM height images for each adsorption condition are shown in Figure 2. HPF adsorption from PBS solution (pHPBS = 7.4) induced the formation of (i) fibril-like structures (Figure 2a, 0.5 μg/ mL), (ii) networks of fibril-like structures (Figure 2b, 2.0 μg/ mL), or (iii) sponge-like structures (Figure 2c, 5.0 μg/mL). The dimensions of the HPF assemblies are shown in Table 1. The size and shape of the fibril-like HPF structures as well as of the HPF fibril network structures suggested that they consisted of several connected HPF molecules. The formation of HPF fibrils and fibril network structures on hydrophobic surfaces is well described in the literature and is in agreement with our observations.15,16,24 Their formation can be explained by a combination of protein−surface and protein−protein interactions during the adsorption process at physiological pH value of 7.4.15,24,44,45 A further increase of the HPF concentration enhanced the probability for protein−protein interactions and, thus, facilitated the formation of thicker and denser HPF layers on the PE-SC surface. Similar dense HPF layers were also reported for the adsorption of HPF on hydrophobic UHMWPE and nanostructured titanium oxide.15,16,46 In contrast to the PBS solutions, we found for HPF adsorption from CBS solution (pHCBS = 9.2) a distinctively different adsorption behavior of the HPF at the PE-SC surfaces. For concentrations of 0.5 μg/mL (Figures 2d and 3) and 2.0 μg/mL (Figure 2e), we observed two different HPF structures at the PE-SC surface: (iv) trinodal-rod-shaped structures and (v) spherical-shaped structures. The trinodal-rod-shaped structures are exemplarily highlighted in the AFM height image on Figure 3. The dimensions of the HPF adsorbed from CBS are shown in Table 1.

adsorbed HPF molecules on PE-SC surfaces was investigated in dependence of the concentration and pH value of the protein solution. Finally, the influence of the PE-SC surface nanostructure on the orientation of HPF assemblies was analyzed. PE-SC Before and After Protein Adsorption. The aim of the current study was to investigate the adsorption and assembly of HPF on the native hydrophobic nanostructured PE-SC surfaces. First, it was ensured that the PE-SC roughness is smooth enough to track single HPF molecules or small HPF assemblies by AFM height imaging. PE-SC samples were characterized by AFM before and after HPF adsorption. Figure 1 illustrates the AFM height images of the same PE-SC before

Figure 1. AFM height image of a PE-SC before (a) and after (b) adsorption of HPF from CBS solution with a pH of 9.2 and a concentration of 0.5 μg/mL. The PE-SC showed a smooth surface before HPF adsorption (a). The inset shows the schematically drawn ideal nanostructured PE-SC chain fold surface (not in scale). After HPF adsorption trinodal-rod-like and spherical structures were found on the PE-SC surface (b). The insets show enlarged images of a spherical (bottom left), a schematically drawn trinodal-rod-like HPF molecule (bottom right), and trinodal-rod structure which is highlighted (top).

(Figure 1a) and after adsorption of HPF from CBS solution (Figure 1b). The root-mean-squared roughness of the PE-SC surface (Rq,PE‑SC) and the underlying silicon substrate (Rq,Si) were calculated to be 0.43 ± 0.02 and 0.13 ± 0.01 nm, respectively. The native PE-SC surface showed an increased surface roughness compared to the underlying silicon substrate. This observation can be explained by the nanostructured surface morphology of the PE-SC which is composed of chain folds. In the current study, it was not possible to image individual chain folds of the PE-SC surface with the given AFM setup. However, the morphology and chain fold orientation at the PE-SC surface under our chosen conditions are well described in the literature.40,41 There, it is shown that the chain folds, which are aligned parallel to the edges of the rhombus, can form the (001) or (112) plane. In the second case, (112) plane, a hollow pyramid will be formed during the crystallization, which collapses after drop casting the PE-SC on a substrate. This leads to a crack in the middle of the pyramid. Such a crack was not visible on our PE-SC surfaces, so the chain folds are aligned in the (001) plane. Additionally, chain folds on the surface of PE-SC have been imaged in multiple previous studies,41,42 and it was shown that 75% of the surface is formed by sharp folds (adjacent reentry) in direction of the close-packed crystal plane ([110]).34,43 On the basis of 11870

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir

Figure 2. AFM height images of HPF structures after adsorption of different HPF concentrations (0.5, 2.0, and 5.0 μg/mL) from PBS (a−c) and CBS (d−f). The insets show magnified pictures of the HPF structures and geometrical models of the arranged HPF molecules. After adsorption from PBS of 0.5 μg/mL HPF (a) fibril-like structures were formed, of 2.0 μg/mL HPF (b) a network of fibril-like structures, and of 5.0 μg/ mL HPF (c) a dense layer was formed. After adsorption from CBS of 0.5 μg/mL (d) and 2.0 μg/mL (e) HPF trinodal-rod-shaped and spherical-shaped structures were formed and of 5.0 μg/mL HPF (f) a dense layer was formed.

(1.2 ± 0.3 nm), width (18 ± 3 nm), and length (51 ± 3 nm) of the trinodal-rod-shaped HPF structures also indicated the presence of HPF molecules with trinodal-rod conformation (Table 1). After adsorption from CBS solution with a HPF concentration of 5.0 μg/mL a dense layer (Figure 2f) was observed on the PE-SC surface. The height of this layer of 1.5 ± 0.5 nm (Table 1) agrees with the height of a single HPF molecule (1.4 ± 0.4 nm).16 This indicates that the dense HPF layer was also

On the PE-SC surface, the amount of both HPF assemblies, the trinodal-rod-shaped and the spherical-shaped structures, increased for a concentration of 2.0 μg/mL (Figure 2e) compared to a concentration of 0.5 μg/mL (Figure 2d). The height (2.0 ± 0.4 nm) and diameter (27 ± 6 nm) (Table 1) of the spherical-shaped HPF structures correspond to values reported for height and diameter of single globular-shaped HPF molecules adsorbed on a hydrophilic silicon surfaces.47 The distinctive geometrical shape16 and dimensions,16 i.e., height 11871

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir Table 1. HPF Assemblies at the PE-SC and their Geometry for Different Concentrations and pHs pH 7.4 (PBS) concn [μg/mL]

a

pH 9.2 (CBS) dimensionsa

HPF morphology

0.5

i

fibril-like structure

2

ii

network of fibril-like structures

5

iii

sponge-like structure

h = 1.9 ± 0.5 nm w = 30 ± 5 nm l = 100−1000 nmb h = 7 ± 3 nm w = 45 ± 10 nm l = 82 ± 25 nmc h = 6 ± 1 nmd

HPF morphology

dimensionsa

iv

trinodal-rod-shaped structure

v

spherical-shaped structure

h = 1.2 ± 0.3 nm w = 18 ± 3 nm l = 51 ± 3 nm h = 2.0 ± 0.4 nm d = 27 ± 6 nm

vi

dense layer

h = 1.5 ± 0.5 nmd

Height (h), width (w), and length (l) or diameter (d) (n = 30). Length of the fibril-like structures was not uniform. Length between cross-links. Based on the assumption that the holes in the sponge-like layer represent the PE-SC surface. b

c

d

molecules in solution. An increase of the pH above the isoelectric point of the HPF increases its overall charge and, thus, also the repulsive forces between the protein molecules in solution. Consequently, protein−protein interactions are reduced.51 (II) The conformation of HPF molecules changes during the adsorption on surfaces. The conformational reorganization is strongly influenced by the interactions between protein and surface which depends on the physical and chemical properties of the surface, i.e., hydrophilicity/hydrophobicity, surface charge, and/or functional groups. The adsorption of HPF molecules on hydrophilic surfaces results from van der Waals interactions between polar αCdomains and the polar surface. Thus, αC-domains attach to the surface and are not available for the interaction with other HPF molecules.23,52However, the adsorption of HPF molecules on hydrophobic surfaces results from the hydrophobic effect.13,14 Nonpolar parts of the HPF molecules, i.e., the D-domains and the central E-domain, attach to the nonpolar surface.24,44,52 Polar αC-domains do not attach to the surface and are thus available for interaction with other HPF molecules. Consequently, single HPF molecules were preferentially observed on hydrophilic surfaces while hydrophobic surfaces promote the formation of HPF fibrils and fibril networks. However, single adsorbed HPF molecules have been observed on hydrophobic materials surfaces when HPF protein−protein interactions were strongly reduced.16,44 In the present study this was realized by using CBS buffer which has a pH value of 9.2. We assume that the increased pH facilitated the deprotonation of the HPF molecules in solution. In consequence, the trinodal-rod conformation was destabilized and other conformation changes were enabled. Thus, a part of the HPF molecules adapted a globular conformation while the other part of the HPF molecules remained in the trinodal-rod conformation. In addition to the conformation changes, the deprotonation also inactivated the αC-domains because they were bound to the central E-domain (model Figures 2d and 2e). Because of the inactivation of the αC-domains, the probability for protein−protein interactions was strongly reduced and the aggregation of several HPF molecules inhibited. In consequence, globular and trinodal-rod-shaped single HPF molecules adsorbed on the PE-SC surface instead of fibrils or fibril networks. Both, HPF fibrils and single trinodal-rod HPF have an anisotropic shape; i.e., their length is higher than their width. An anisotropic nanostructured surface may influence the orientation of these anisotropic shaped objects during adsorption. In the following, we will investigate and discuss

Figure 3. AFM height image of a PE-SC surface after HPF adsorption from CBS solution (pH = 9.2) at a concentration of 0.5 μg/mL. On the PE-SC surface trinodal-rod-shaped HPF structures and sphericalshaped HPF structures were observed. The inset shows a magnification of the marked area without and with an overlay, which highlights the trinodal-rod-shaped HPF structures and their orientation.

formed by single molecules. In comparison to lower HPF concentrations (0.5 and 2.0 μg/mL) of the CBS solution, trinodal-rod-shaped structures or spherical-shaped HPF structures were not distinguishable. Thus, we conclude that the CBS solution induced the adsorption of single globular-shaped and trinodal-rod-shaped HPF molecules at the PE-SC surface instead of the HPF fibrils or fibril networks as found after the adsorption from PBS solution at similar HPF concentrations. The observation of these two totally different HPF assemblies can be explained by buffer-dependent HPF protein−protein interactions which were further influenced by interactions of the HPF proteins with the hydrophobic PE-SC surface. HPF protein−protein interactions strongly depend on the presence of free αC-domains and are thus directly linked to the HPFs conformation. Free αC-domains of a HPF molecule can bind to other HPF molecules which facilitates the formation of HPF assemblies consisting of several connected proteins.15,16,24 The HPF conformation and the resulting availability of free αC-domains for HPF aggregation can be triggered by (I) the conditions in the buffer solution49 and by (II) the interactions with the surface:15,16,24 (I) The conditions in the buffer solution, i.e., pH value and ionic strength, affect the conformation as well as the overall charge of the HPF proteins in the solution.11,13,24 A high or low pH value of the buffer solution induces the binding of HPF αCdomains to the central E-domain. Consequently, the amount of unbound αC-domains is reduced and the probability for HPF protein−protein interactions decreases.11,50 Additionally, high or low pH values also influence the overall charge of the HPF 11872

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir

Figure 4. Angle distribution of fibril-like (left) and trinodal-rod-shaped (right) structures on the PE-SC surface. For the trinodal-rod-shaped max max structures maxima at Θmax 1 = 0 ± 5°, Θ2 = 85 ± 5°, and Θ3 = 175 ± 5° were found. The scheme (bottom) illustrates the angle measurements and shows differently oriented single HPF molecules and the trinodal-rod-shaped structure maxima are highlighted.

Because of the symmetry of the HPF molecule, Θ 1max corresponds to Θmax 3 , so there are only two preferred directions. A total of 87% of the 1513 analyzed single HPF molecules featured an orientation in the angle ranges from 70° to 110° and from 160° to 20°. This observation indicates that the surface nanostructure of the PE-SC influenced the orientation of adsorbed single HPF molecules but interestingly not of HPF fibrils. The oriented adsorption of single HPF molecules can be explained by considering crystallographic directions of high surface chain densities at the PE-SC surface. Here, the ∼ 90° and Θmax ∼ 180° maximum orientation angles of Θmax 1 2 correspond to an orientation of the single HPF molecules along the [100] and [010] directions of the PE-SC. Both the [100] and the [010] directions are characterized by increased linear surface chain densities (SCD) compared to other crystallographic directions on the PE-SC surface. We define the SCD as the number of chain fold start points per distance (nm) with respect to the crystallographic direction.53 Based on these assumptions, the highest SCD was found for the [010] direction to be 2.02 chain fold start points per nm (2.02 nm−1). Also, the [100] direction features a higher SCD with 1.35 nm−1 as compared to the [110] direction with 1.12 nm−1. Because of the high SCD, the HPF molecules can develop an increased amount of intermolecular bonds along specific crystallographic directions which, thus, induce an orientation of the protein. This effect also explained the preferential adsorption of HPF on the crystalline lamellae of semicrystalline PE as compared to the amorphous PE regions.16 This observed orientation of trinodal-rod-like HPF molecules on PE-SC is different to that of other rod-like assemblies reported in the literature. Te nanorods (rod-like crystals) and evaporated PE nanorods (rod-like crystals) on PE-SC surfaces orientated preferably along the [110] directions.36,37 The

the influence of the hydrophobic PE-SC surface nanostructure on the orientation of HPF fibrils and single HPF molecules. Orientation of HPF Assemblies on Nanostructured PE-SC Surfaces. The PE-SC has an anisotropic surface nanostructure (see Figure S2C) which is determined by the crystal lattice parameters and the chain folds. We investigated the influence of this PE-SC surface nanostructure on the orientation of HPF assemblies. We selected the fibril-like structures adsorbed from PBS solution and trinodal-rod-shaped structures adsorbed from CBS solution because they allow to meaningfully determine orientation by AFM in contrast to the other structures like networks and dense layers. The orientation of both structures, fibril and trinodal-rod-shaped, was determined by the angle between the structures’ long axis and the short diagonal B of the PE-SC (see Figure S1B for details on the geometry). Figure 4a shows the statistical analysis of the angle distribution for fibril-like structures. The frequency of the categorized orientation angles Θ ranges from 2.5% for 165 ± 5° to 7.5% for 135 ± 5°. Maxima were found for Θmax 1 = 135 ± 5° min = 165 ± 5°. However, and Θmax 2 = 65 ± 5° and minima for Θ a statistical evidence for a preferential orientation of the adsorbed fibril-like structures cannot be deduced due to the small differences in angle frequency of 5% between Θmax 1 and 2 and Θmin. After the adsorption from CBS solutions (pHCBS = 9.2) at 0.5 and 2.0 μg/mL HPF trinodal-rod-shaped structures were observed on the PE-SC surface (see Figures 1b and 2d, e). Figure 4b shows the statistical analysis of the angle distribution for trinodal-rod-shaped structures in relation to the short diagonal B of the PE-SC. The frequency of categorized orientation angles Θ is in the range of 1%−22%. In comparison to the PBS adsorption experiments, three distinct maxima Θmax 1 = 0 ± 5°, Θmax = 85 ± 5°, and Θmax = 175 ± 5° were found. 2 3 11873

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir

Figure 5. Model (true to scale) for the orientation of α-helix structures dependent on crystallographic directions. A fibrinogen model is shown on top where an α-helix-rich region is highlighted, the coiled-coil region. The α-helix is illustrated as a simple cylinder with a width of ∼3 nm (amino acid residues not considered) and a pitch length of ∼5.4 nm;55 the turns of the helix are marked (helix angle ∼30°). Below: the PE-unit cell is shown with the dimensions of a = 0.494 nm and b = 0.741 nm.53 The middle, sectors I and IV of a PE-SC with differently oriented HPF molecules are shown. The insets show possible orientation of a corresponding α-helix. Observed preferred orientations are highlighted.

(see Table 1), this is correlated to the parallel orientation of the α-helix chains to the PE-SC chain folds which minimizes their Gibbs free energy. Under this assumption, the orientation of the chains in the α-helix is identical to the aligned PE nanorods and Te nanorods found in previous studies.36,37 Also, the dimensions (width and pitch length) of an α-helix are in the same order of the PE unit cell (Figure 5). The model in Figure 5 shows that the turns of an α-helix can stack between two chain-folds in [010] direction of sector I and [100] direction of sector IV. These directions are similar to the observed preferred orientation of single HPF molecules (Figure 4). The stacking is facilitated by the similar orientation of chain folds on the PESC surface and chains in the helix. A change in the orientation of an α-helix leads to an unstacking of the turns. Further, a helix aligned in [010] direction covers the highest number of chain folds (Figure 5) and also has the highest SCD; thus, it can exhibit the most intermolecular interactions. Beside the [010] orientation, we also observed HPF molecules orientated along the [100] direction. This direction has the second highest SCD, as shown in Figure 5. This leads to an increase of formed intermolecular interactions between chain folds and the helix chain. Both effects together, the two highest SCDs and the possible stacking of the α-helix chain between the chain folds, dependent on the sector of the PE-SC, can apparently influence the orientation of HPF molecules. The observation of HPF orientation angles beside the preferred ones can be explained by defects of the PE-SC surface, i.e., loose loops. These defects

orientation of small PE molecules and rod-like Te crystals compared to a HPF molecule was assumed to be driven by short intermolecular interactions between the chain folds and the molecules. In both cases the orientation was induced by the alignment of the chains (PE) and the anisotropic nuclei (Te) during the crystallization along or perpendicular to the chain fold direction.36,37 During the adsorption on hydrophobic surfaces the E- and D-domains of an HPF molecule start to spread to increase the contact area.44,52 Therefore, the structures inside the E- and Ddomains have to change their orientation.47 The HPF molecule consists of six polypeptide chains with formed α-helix and βsheet structures. On hydrophobic surfaces like the PE-SC the αhelix structures are dominant.54 An α-helix-rich region in the HPF molecule is the coiled-coil region, but the structure is also present in the E- and D-domains where they are located in the outer parts.47 The observed orientations of trinodal-rod-shaped HPF molecules on PE-SC can be explained as follows: We assume that the α-helix structures of the chains which are located in the outer part of the protein47 influence the orientation because the dimensions (width and pitch length) and the pitch angle of the α-helix are similar to the PE unit cell.53,55 The α-helix has a pitch angle, the angle between pitch and helix axis, being approximately 30°. This is similar to the angle of the [010] and the [110] direction, which is 33.7°. We propose that when HPF’s helical structures align along the [010] direction, as observed for single HPF molecules in case iv 11874

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir

sible for the HPF orientation at higher pH. This will be part of further studies.

will lead to a change in the orientation of the PE chain folds and, thus, influences the orientation of the single HPF molecules. This orientation effect, however, was only observed for single HPF molecules that adsorbed on the nanostructured PE-SC surface. HPF fibrils featured no preferential orientation and were formed due to the conformation changes of single HPF molecules that adsorbed on the PE-SC surface. This observation can be explained by the increased protein−protein interactions when adsorbing from PBS solution (pHCBS = 7.4) compared to the CBS solution (pHCBS = 9.2). As mentioned before, the lower pH increased the protein−protein interactions between adsorbed neighboring HPF molecules and, thus, induced the fibril formation. The formation process itself influenced the orientation of the fibrils stronger than the increased SCD of specific crystallographic directions. In addition, due to the increased molar mass, the surface diffusion of HPF fibrils was decreased compared to the surface diffusion of single HPF molecules. Thus, a combination of surface induced fibril formation and a limited surface diffusion of these HPF fibrils inhibited their orientation. From the literature it is known that the lateral diffusion of proteins on surfaces is strongly influenced by protein−protein interactions as well as the surface hydrophilicity/hydrophobicity.56,57 A significantly lower rate of surface diffusion facilitated by protein−protein interactions was reported for the adsorption of bovine serum albumin on poly(methyl methacrylate) surfaces.58 For HPF proteins adsorbed on nanostructured melt-drawn poly(ethylene) films it was observed that the surface diffusion of the molecules decreased with increasing protein−protein interactions. Besides, also the already weaker protein−surface interactions will be further decreased by increasing protein−protein interactions.56,57 In addition, the increased protein−protein interactions also lead to longer residence times of HPF molecules on the surface.46 In previous studies preferential orientation of proteins has been induced by either a heterogeneous surface chemistry, e.g., by using amphiphilic diblock copolymers,59 or by a defined surface topography with dimensions in the range of the protein, like rippled structures.60 For diblock copolymers it was found that proteins preferentially adsorbed on domains of hydrophobic blocks due to the preferential adsorption.59 In contrast, on rippled silica substrates it was shown that the orientations of adsorbed single HPF molecules depend on the rippled wavelength due the surface adsorption energy that decreased with increasing surface curvature.60 Furthermore, an orientation of HPF proteins as well as HPF nanofibrils was observed at nanosteps of HOPG layers.38 In the previous studies the used structures to orientate HPF were of the order of 50 nm. However, in the present study the orientation of the HPF molecules was induced by the subnanoscaled surface structure of the PE-SC. The PE-SC has a chemically homogeneous surface yet an anisotropic topography below the length scale of the protein dimension. Here, we propose that the single HPF molecules orient on the PE-SC surface to increase the amount of surface−protein intermolecular bonds along crystallographic directions with an increased surface chain density. However, a preferential alignment of HPF fibrils was not observed and this is explained by increased protein−protein interactions and decreased protein diffusion compared to the single HPF molecules. Further investigation of HPF adsorption on PE-SC surfaces at higher pH values and lower concentrations can give additional information on the interactions which are respon-

■

CONCLUSION Nanostructured polymer surfaces can be used to understand and control the adsorption of proteins on biomaterials surfaces. A suitable model system to study protein interactions and surface assembly are hydrophobic polyethylene single crystals which feature a defined surface nanostructure. The study established that HPF molecules can align on chemically homogeneous surface topographies one order of magnitude smaller than the dimension of the protein. By orienting on the PE-SC surface, the single HPF molecules increase the amount of surface−protein intermolecular bonds which can be higher along crystallographic directions with an increased surface chain density. In addition, the similar dimensions of the α-helix and chain folds support the alignment. The understanding of protein assembly and orientation is crucial for the application of biomaterials in contact with a biological system. This work presents a strategy to control and orient proteins on nanostructured polymeric biomaterials to improve the surface functionalities, like cell adhesion, of future implants and biosensors.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.6b03110. Figures S1 and S2 (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]; Tel +49 (0) 3641 94 77 30; Fax +49 (0) 3641 94 77 32. Present Addresses

D.A.H.: Department of Clinical Radiology, University Hospital Münster, Albert-Schweitzer-Straße 33, 48149 Münster, Germany. M.M.L.A.: Biology and Soft Matter Division, Neutron Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, TN 37831. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge the partial financial support of the Deutsche Forschungsgemeinschaft (DFG) project: “Novel functional materials based on self-assembled protein nanofibers: creating and understanding nanofibers”, AOBJ: 609403.

■

REFERENCES

(1) Liu, X.; Deng, J.; Ma, L.; Cheng, C.; Nie, C.; He, C.; Zhao, C. Catechol Chemistry Inspired Approach to Construct Self-CrossLinked Polymer Nanolayers as Versatile Biointerfaces. Langmuir 2014, 30, 14905−14915. (2) Thyparambil, A. A.; Wei, Y.; Latour, R. A. Evaluation of the Effectiveness of Surfactants and Denaturants to Elute and Denature Adsorbed Protein on Different Surface Chemistries. Langmuir 2015, 31, 11814−11824. (3) Kasemo, B. Biological surface science. Surf. Sci. 2002, 500, 656− 677.

11875

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir (4) Subramani, K.; Jung, R. E.; Molenberg, A.; Hammerle, C. H. Biofilm on dental implants: a review of the literature. Int. J. Oral Maxillofac Implants 2009, 24, 616−26. (5) Brash, J. L., Horbett, T. A. Eds.; Proteins at Interfaces. Physicochemical and Biochemical Studies. ACS Symp. Ser. 1987, 343, doi 10.1021/bk-1987-0343.pr001. (6) Huisman, I. H.; Prádanos, P.; Hernández, A. The effect of protein−protein and protein−membrane interactions on membrane fouling in ultrafiltration. J. Membr. Sci. 2000, 179, 79−90. (7) Colman, R. W.; Hirsh, J.; Marder, V. J.; Clowes, A. W.; George, J. N. Hemostasis and Thrombosis: Basic Principles and Clinical Practice; Lippincott Williams & Wilkins: Philadelphia, PA, 2001. (8) Tang, L. P.; Eaton, J. W. Fibrin(ogen) mediates acute inflammatory responses to biomaterials. J. Exp. Med. 1993, 178, 2147−56. (9) Doolittle, R. F. Fibrinogen and Fibrin. Annu. Rev. Biochem. 1984, 53, 195−229. (10) Cacciafesta, P.; Humphris, A. D. L.; Jandt, K. D.; Miles, M. J. Human plasma fibrinogen adsorption on ultraflat titanium oxide surfaces studied with atomic force microscopy. Langmuir 2000, 16, 8167−8175. (11) Tsapikouni, T. S.; Missirlis, Y. F. Protein−material interactions: From micro-to-nano scale. Mater. Sci. Eng., B 2008, 152, 2−7. (12) Kastantin, M.; Langdon, B. B.; Schwartz, D. K. A bottom-up approach to understanding protein layer formation at solid−liquid interfaces. Adv. Colloid Interface Sci. 2014, 207, 240−252. (13) Roach, P.; Farrar, D.; Perry, C. C. Interpretation of Protein Adsorption: Surface-Induced Conformational Changes. J. Am. Chem. Soc. 2005, 127, 8168−8173. (14) Barrias, C. C.; Martins, C. L.; Miranda, C. S.; Barbosa, M. A. Adsorption of a therapeutic enzyme to self-assembled monolayers effect of surface chemistry and solution pH on the amount and activity of adsorbed enzyme. Biomaterials 2005, 26, 2695−2704. (15) Keller, T. F.; Reichert, J.; Thanh, T. P.; Adjiski, R.; Spiess, L.; Berzina-Cimdina, L.; Jandt, K. D.; Bossert, J. Facets of protein assembly on nanostructured titanium oxide surfaces. Acta Biomater. 2013, 9, 5810−5820. (16) Keller, T. F.; Schonfelder, J.; Reichert, J.; Tuccitto, N.; Licciardello, A.; Messina, G. M. L.; Marletta, G.; Jandt, K. D. How the surface nanostructure of polyethylene affects protein assembly and orientation. ACS Nano 2011, 5, 3120−3131. (17) Ortega-Vinuesa, J. L.; Tengvall, P.; Lundstrom, I. Molecular packing of HSA, IgG, and fibrinogen adsorbed on silicon by AFM imaging. Thin Solid Films 1998, 324, 257−273. (18) Höök, F.; Rodahl, F.; Kasemo, B.; Brzezinski, P. Structural changes in hemoglobin during adsorption to solid surfaces: Effects of pH, ionic strength, and ligand binding. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 12271−12276. (19) Jones, K. L.; O’Melia, C. R. Protein and humic acid adsorption onto hydrophilic membrane surfaces: effects of pH and ionic strength. J. Membr. Sci. 2000, 165, 31−46. (20) Anand, G.; Zhang, F.; Linhardt, R. J.; Belfort, G. ProteinAssociated Water and Secondary Structure Effect Removal of Blood Proteins from Metallic Substrates. Langmuir 2011, 27, 1830−1836. (21) Lin, K.; Xia, L.; Gan, J.; Zhang, Z.; Chen, H.; Jiang, X.; Chang, J. Tailoring the Nanostructured Surfaces of Hydroxyapatite Bioceramics to Promote Protein Adsorption, Osteoblast Growth, and Osteogenic Differentiation. ACS Appl. Mater. Interfaces 2013, 5, 8008−8017. (22) Hahm, J.-I. Fundamentals of Nanoscale Polymer−Protein Interactions and Potential Contributions to Solid-State Nanobioarrays. Langmuir 2014, 30, 9891−9904. (23) Cao, L.; Sukavaneshvar, S.; Ratner, B. D.; Horbett, T. A. Glow discharge plasma treatment of polyethylene tubing with tetraglyme results in ultralow fibrinogen adsorption and greatly reduced platelet adhesion. J. Biomed. Mater. Res., Part A 2006, 79A, 788−803. (24) Koo, J.; Galanakis, D.; Liu, Y.; Ramek, A.; Fields, A.; BA, X.; Simon, M.; Rafailovich, M. H. Control of Anti-Thrombogenic Properties: Surface-Induced Self-Assembly of Fibrinogen Fibers. Biomacromolecules 2012, 13, 1259−1268.

(25) Scott, E. A.; Elbert, D. L. Mass Spectrometric Mapping of Fibrinogen Conformations at Poly(ethylene terephthalate) Interfaces. Biomaterials 2007, 28, 3904−3917. (26) George, P. A.; Doran, M. R.; Croll, T. I.; Munro, T. P.; CooperWhite, J. J. Nanoscale presentation of cell adhesive molecules via block copolymer self-assembly. Biomaterials 2009, 30, 4732−4737. (27) Strobl, G. The Physics of Polymers. Concepts for Understanding Their Structure and Behavior, 3rd ed.; Springer: Berlin, 2007. (28) Muthukumar, M. Shifting Paradigms in Polymer Crystallization. Progress in Understanding of Polymer Crystallization Lecture Notes in Physics 2007, 714, 1−18. (29) Granchi, D.; Ciapetti, G.; Amato, I.; Pagani, S.; Cenni, E.; Savarino, L.; Avnet, S.; Peris, J. L.; Pellacani, A.; Baldini, N.; Giunti, A. The Influence of Alumina and Ultra-High Molecular Weight Polyethylene Particles on Osteoblast-Osteoclast Cooperation. Biomaterials 2004, 25, 4037−4045. (30) Gonzalez-Mora, V. A.; Hoffmann, M.; Stroosnijder, R.; Gil, F. J. Wear Tests in a Hip Joint Simulator of Different CoCrMo Counterfaces on UHMWPE. Mater. Sci. Eng., C 2009, 29, 153−158. (31) Ren, W. P.; Markel, D. C.; Zhang, R.; Peng, X.; Wu, B.; Monica, H.; Wooley, P. H. Association between UHMWPE Particle- Induced Inflammatory Osteoclastogenesis and Expression of RANKL, VEGF, and Flt-1. Biomaterials 2006, 27, 5161−5169. (32) Heuberger, M. P.; Widmer, M. R.; Zobeley, E.; Glockshuber, R.; Spencer, N. D. Protein-Mediated Boundary Lubrication in Arthroplasty. Biomaterials 2005, 26, 1165−1173. (33) Berglin, M.; Pinori, E.; Sellborn, A.; Andersson, M.; Hulander, M.; Elwing, H. Fibrinogen Adsorption and Conformational Change on Model Polymers: Novel Aspects of Mutual Molecular Rearrangement. Langmuir 2009, 25, 5602−5608. (34) Allegra, G.; Famulari, A. Chain statistics in polyethylene crystallization. Polymer 2009, 50, 1819−1829. (35) Reneker, D. H.; Geil, P. H. Morphology of Polymer Single Crystals. J. Appl. Phys. 1960, 31, 1916−1925. (36) Faghihi, S.; Hoffmann, Th.; Petermann, J.; Martinez-Salazar, J. Decoration of the Fold Surfaces of Polyethylene Single Crystals with Tellurium and Tin. Macromolecules 1992, 25, 2509−2512. (37) Wittmann, J. C.; Lotz, B. Polymer Decoration: The Orientation of Polymer Folds as Revealed by the Crystallization of Polymer Vapors. J. Polym. Sci., Polym. Phys. Ed. 1985, 23, 205−226. (38) Reichert, J.; Wei, G.; Jandt, K. D. Formation and Topotactical Orientation of Fibrinogen Nanofibrils on Graphite Nanostructures. Adv. Eng. Mater. 2009, 11, 177−181. (39) Tian, M.; Dosière, M.; Hocquet, S.; Lemstra, P. J.; Loos, J. Novel Aspects Related to Nucleation and Growth of Solution Grown Polyethylene Single Crystals. Macromolecules 2004, 37, 1333−1341. (40) Geil, P. H. Polymer Deformation. II. Drawing of Polyethylene Single Crystals. J. Polym. Sci., Part A: Gen. Pap. 1964, 2, 3813−3833. (41) Patil, R.; Reneker, D. H. Molecular Folds in Polyethylene Observed by Atomic-force Microscopy. Polymer 1994, 35, 1909−1914. (42) Magonov, S. N.; Reneker, D. H. Characterization of Polymer Surfaces with Atomic Force Microscopy. Annu. Rev. Mater. Sci. 1997, 27, 175−222. (43) Hoffman, J. D.; Miller, R. L. Kinetics of crystallization from the melt and chain folding in polyethylene fractions revisited: theory and experiment. Polymer 1997, 38, 3151−3212. (44) Gettens, R. T. T.; Bai, Z. J.; Gilbert, J. L. Quantification of the kinetics and thermodynamics of protein adsorption using atomic force microscopy. J. Biomed. Mater. Res., Part A 2005, 72A, 246−257. (45) Marchin, K. L.; Berrie, C. L. Conformational Changes in the Plasma Protein Fibrinogen upon Adsorption to Graphite and Mica Investigated by Atomic Force Microscopy. Langmuir 2003, 19, 9883− 9888. (46) Kastantin, M.; Keller, T. F.; Jandt, K. D.; Schwartz, D. K. Singlemolecule tracking of fibrinogen dynamics on nanostructured poly(ethylene) films. Adv. Funct. Mater. 2012, 22, 2617−23. (47) Tunc, S.; Maitz, M. F.; Steiner, G.; V́ azquez, L.; Pham, M. T.; Salzer, R. In situ conformational analysis of fibrinogen adsorbed on Si surfaces. Colloids Surf., B 2005, 42, 219−225. 11876

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877

Article

Langmuir (48) Wei, G.; Reichert, J.; Jandt, K. D. Controlled self-assembly and templated metallization of fibrinogen nanofibrils. Chem. Commun. 2008, 3903−3905. (49) Wei, G.; Reichert, J.; Bossert, J.; Jandt, K. D. Novel Biopolymeric Template for the Nucleation and Growth of Hydroxyapatite Crystals Based on Self-Assembled Fibrinogen Fibrils. Biomacromolecules 2008, 9, 3258−3267. (50) Mihalyi, E.; Tercero, J. C.; Diaz-Mauriño, T. Changes in pH associated with clotting of fibrinogen. Kinetic studies of the pH shift and correlation of the pH change with the release of fibrinopeptides and the ensuing polymerization. Biochemistry 1991, 30, 4753−4762. (51) Weisel, J. W.; Medved, L. The Structure and Function of the αC Domains of Fibrinogen. Ann. N. Y. Acad. Sci. 2001, 936, 312−327. (52) Sit, P. S.; Marchant, R. E. Surface-Dependent Conformations of Human Fibrinogen Observed by Atomic Force Microscopy under Aqueous Conditions. Thromb. Haemostasis 1999, 82, 1053−1060. (53) Bunn, C. W. Chemical Crystallography; Oxford University Press: Oxford, 1945. (54) Steiner, G.; Tunc, S.; Maitz, M.; Salzer, R. Conformational Changes during Protein Adsorption. FT-IR Spectroscopic Imaging of Adsorbed Fibrinogen Layers. Anal. Chem. 2007, 79, 1311−1316. (55) Pauling, L.; Corey, R. B.; Branson, H. R. The Structure of Proteins: Two Hydrogen-bonded Helical Configurations of the Polypeptide chain. Proc. Natl. Acad. Sci. U. S. A. 1951, 37, 205−211. (56) Langdon, B. B.; Kastantin, M.; Schwartz, D. K. Apparent Activation Energies Associated with Protein Dynamics on Hydrophobic and Hydrophilic Surfaces. Biophys. J. 2012, 102, 2625−2633. (57) Langdon, B. B.; Kastantin, M.; Walder, R.; Schwartz, D. K. Interfacial Protein−Protein Associations. Biomacromolecules 2014, 15, 66−74. (58) Tilton, R. D.; Gast, A. P.; Robertson, C. R. Surface diffusion of interacting proteins: Effect of concentration on the lateral mobility of adsorbed bovine serum albumin. Biophys. J. 1990, 58, 1321−1326. (59) Matsusaki, M.; Omichi, M.; Kadowaki, K.; Kim, B. H.; Kim, S. O.; Maruyama, I.; Akashi, M. Protein nanoarrays on a highly-oriented lamellar surface. Chem. Commun. 2010, 46, 1911−1913. (60) Sommerfeld, J.; Richter, J.; Niepelt, R.; Kosan, S.; Keller, T. F.; Jandt, K. D.; Ronning, C. Protein Adsorption on Nano-scaled, Rippled TiO2 and Si Surfaces. Biointerphases 2012, 7, 1−7.

11877

DOI: 10.1021/acs.langmuir.6b03110 Langmuir 2016, 32, 11868−11877