Correlation between Tribological Properties and the Quantified Structural Changes of Lysozyme on Poly (2-hydroxyethyl methacrylate) Contact Lens

The ocular discomfort is the leading cause of contact lens wear discontinuation. Although the tear proteins as a lubricant might improve contact lens adaptation, some in vitro studies suggested that the amount of adsorbed proteins could not simply explain the lubricating performance of adsorbed proteins. The purpose of this study was to quantify the structural changes and corresponding ocular lubricating properties of adsorbed protein on a conventional contact lens material, poly (2-hydroxyethyl methacrylate) (pHEMA). The adsorption behaviors of lysozyme on pHEMA were determined by the combined effects of protein–surface and protein–protein interactions. Lysozyme, the most abundant protein in tear, was first adsorbed onto the pHEMA surface under widely varying protein solution concentrations to saturate the surface, with the areal density of the adsorbed protein presenting different protein–protein effects within the layer. These values were correlated with the measured secondary structures, and corresponding friction coefficient of the adsorbed and protein covered lens surface, respectively. The decreased friction coefficient value was an indicator of the lubricated surfaces with improved adaptation. Our results indicate that the protein–protein effects help stabilize the structure of adsorbed lysozyme on pHEMA with the raised friction coefficient measured critical for the innovation of contact lens material designs with improved adaptation.


Introduction
The use of contact lenses has become increasingly popular for vision correction and cosmetic reasons over prescribed spectacles [1,2]. However, the insertion of a contact lens into the eye does change the situation at the ocular surface [3] and often causes the wearer contact lens-related discomfort (CLD). CLD is a condition characterized by adverse sensation [4,5] resulting from reduced compatibility between the contact lens and the ocular environment [6], which usually leads to discontinuation of contact lens wear [7]. Various factors that may be related to CLD include lens material designs [8,9] and ocular changes such as varied tear composition and external variations, which include the use of medications, room humidity, or air temperature [10]. Therefore, the management of CLD remains a challenge, and research into CLD aims to determine which factor will improve contact lens adaptation.
A more recent area of interest is lubricity and friction involved during contact lens wear, with current studies suggesting that the reduced friction between the cornea, lens surface, and lid

Method for Protein Adsorption on Contact Lenses
Each contact lens was taken out of the packaging and the excess liquid was removed before using it. The contact lenses were immersed in 30 mL lysozyme solutions of different concentrations (0.5, 0.7, 1.0, 1.5, and 1.9 mg/mL) for two hours. After this, the protein solution was changed to pure physiological saline (pH 7.0-7.2) for another two hours for desorption of loosely adsorbed lysozyme on surfaces [22].

Analysis of the pHEMA Surface
The surface element was analyzed with X-ray photoelectron spectroscopy (XPS, VG ESCALAB250, VG Scientific, East Grinstead, UK). The solid sample was prepared and placed under a ultra-high vacuum (<9 × 10 −8 mBar). The focused X-ray energy was adjusted to 1.5 keV, and the z-axis position was confirmed to focus on the object's surface. The element spectrum was clicked, and the orbit was set. The number of scans was five times each time, and the energy was 1 electron volt (eV). The corresponding raw data are shown in Section S1 of the Supplementary Materials.
The contact angle of the surface was tested with a CA-D type contact angle meter (100 SB Sindatek, Taiwan). The deionized water droplets were placed on the surface of the object, and the angle was measured.
The surface roughness was analyzed with atomic force microscopy (AFM, XE-100, Park System, Ilsan, Korea). The sample surface was required to be flat and attached to the stage, with a scan range of 10 µm × 10 µm. The detailed descriptions are given in Section S2 of the Supplementary Materials.

Protein Adsorption Areal Density Measurement
The protein concentrations were verified via circular dichroism spectroscopy (CD, Spectropolarimeter J-810, Tokyo, Japan). Different protein solutions were prepared using physiological saline (pH 7.0-7.2) as a solvent, including 0.01 mg/mL, 0.02 mg/mL, 0.03 mg/mL, 0.04 mg/mL, and 0.05 mg/mL under room temperature and measured at a wavelength (λ) of 205 nm [29]. The measured absorbance values vs. protein concentrations calibrated by a BCA assay [30] were noted to construct the standard calibration curve.
The molar extinction coefficient of the protein (ε 205 ) in solution at 205 nm was then determined by recording the background-corrected absorbance at different solution concentrations, as mentioned in Section 2.2, to determine the areal density of the adsorbed protein. Subsequently, the molar extinction coefficient of lysozyme in the saline was obtained from the slope of the absorbance (A 205 ) vs. (Csoln × L) plot. Thus, the areal density of the adsorbed lysozyme was determined by the following equation [31]: where A 250 is the background corrected absorbance at 205 nm, and ε 205 is the molar coefficient that was determined for the protein solution at a wavelength of 205 nm.

Measurement of the Structural Changes of Lysozyme Adsorbed on Soft Contact Lens
The native structure of proteins in physiological saline and the following adsorption-induced secondary structure changes of these adsorbed proteins were determined using circular dichroism (CD). In this study, the CD was operated at room temperature over the wavelength range between λ 1 = 280 nm and λ 2 = 190 nm. The path length was 1 cm, and the scan speed was 50 nm/min in solution status and 100nm/min in adsorbed protein solution, respectively, with standard (100 mdeg) sensitivity. Each spectrum was averaged three times.
The secondary structure in solution and of adsorbed protein was estimated by converting the background-corrected CD signals to molar ellipticity (θ mol ) using the following equations [32]: where θ raw is the background corrected raw CD signal, L is the path length of the cuvette (cm), C soln is the solution concentration of the protein (g/mL), Q ads is the surface density of adsorbed protein(g/cm 2 ), and M is the mean residue molecular weight of 112 g/mol. From the obtained molar ellipticity (θ mol ), the secondary structure content was estimated by DichroWeb, an online database [33,34].

In Vitro Testing System of Friction Coefficient
In vitro contact lens friction tests were measured with a CETR universal micro-tribometer-2 (UMT-2, Bruker, Campbell, CA, USA) using the procedure previously described by Su et al. to simulate the ocular environment [24]. A contact lens adsorbed with lysozyme was fixed on the top stage while the glass was used as the bottom. The contact lens was then rubbed against the quartz glass as the intraocular or natural crystalline lens in the presence of 10 mL of physiological saline [35,36]. A previous study suggested that the eyelid force exhibited towards the eye was in the range of 47-149 milli-Newtons (mN) [37]. The normal force applied in this study was 60 mN due to its most stabilized noise to signal ratio with the rotation speed of 1 rpm. The rotation radius was 10 mm with 5 mm of radius contact area, and the rotation time was 900 s. The quartz glass was cleaned with 75% ethanol after each friction test, and each condition was repeated more than three times.

Statistical Analysis
Differences in friction coefficients and lysosomal structure changes were assessed by Student's t-test to make an allowance of comparisons. A value of p < 0.05 was considered significant.

Characteristics of pHEMA Surfaces
The surface contact angle was measured as 41.2 ± 0.5 (average ± standard deviation, n = 3), similar to the reported values [38,39]. Elemental composition was measured by X-ray photoelectron spectroscopy with the analysis spectrum mentioned in Supplementary Materials Figure S1. Varied surface roughness based on different coating parameters were collected from AFM images and are shown in Supplementary Materials Figure S2. All of the measured values reported fell within the expected range [40]. densities corresponding to the theoretical limits for a saturated surface. Theoretically, lysozyme could be organized in a close-packed side-on orientation (0.17 µg/cm 2 ) and close-packed end-on orientation (0.26 µg/cm 2 ) to saturate the surface with a monolayer [22].

The Areal Density of Adsorbed Proteins and Protein-Protein Effects
Polymers 2020, 12, x FOR PEER REVIEW 5 of 10 Figure 1 presents a plot of the areal density of the adsorbed lysozyme on the pHEMA surface for each protein solution concentration. As shown, protein solutions resulted in distributed areal densities, with the amount of adsorbed protein from higher solution concentrations falling within the areal densities corresponding to the theoretical limits for a saturated surface. Theoretically, lysozyme could be organized in a close-packed side-on orientation (0.17 μg/cm 2 ) and close-packed end-on orientation (0.26 μg/cm 2 ) to saturate the surface with a monolayer [22].
The interaction of a protein with the surface may occur as a dynamic exchange process with adsorption rate and desorption rate represented by kon and koff,, respectively. The adsorbed proteins may undergo surface-induced unfolding and spread out over the surface to increase their footprint on the surface [41] with the unfolding rate expressed as ks. (b) The adsorption capacity represented by the areal density of the adsorbed lysozyme protein on the pHEMA surface was measured at different protein solution concentrations. The error bars denote the mean ± SD for n = 3.

The Protein-Protein Effects and Protein Secondary Structure
The influence of protein-protein interactions on the secondary structure of adsorbed lysozyme is presented in Figure 2. As shown, secondary structures of lysozyme adsorbed from increased solution concentration were comparatively retained on the pHEMA surface. There was a significant reduction in the helical structure. In contrast, a substantial increase in percent sheet content was observed for the lysozyme layers adsorbed from solutions at lower protein concentration when compared to native protein structure in physiological saline. The interaction of a protein with the surface may occur as a dynamic exchange process with adsorption rate and desorption rate represented by k on and k off " respectively. The adsorbed proteins may undergo surface-induced unfolding and spread out over the surface to increase their footprint on the surface [41] with the unfolding rate expressed as k s . (b) The adsorption capacity represented by the areal density of the adsorbed lysozyme protein on the pHEMA surface was measured at different protein solution concentrations. The error bars denote the mean ± SD for n = 3.

The Protein-Protein Effects and Protein Secondary Structure
The influence of protein-protein interactions on the secondary structure of adsorbed lysozyme is presented in Figure 2. As shown, secondary structures of lysozyme adsorbed from increased solution concentration were comparatively retained on the pHEMA surface. There was a significant reduction in the helical structure. In contrast, a substantial increase in percent sheet content was observed for the lysozyme layers adsorbed from solutions at lower protein concentration when compared to native protein structure in physiological saline.
Polymers 2020, 12, x FOR PEER REVIEW 5 of 10 Figure 1 presents a plot of the areal density of the adsorbed lysozyme on the pHEMA surface for each protein solution concentration. As shown, protein solutions resulted in distributed areal densities, with the amount of adsorbed protein from higher solution concentrations falling within the areal densities corresponding to the theoretical limits for a saturated surface. Theoretically, lysozyme could be organized in a close-packed side-on orientation (0.17 μg/cm 2 ) and close-packed end-on orientation (0.26 μg/cm 2 ) to saturate the surface with a monolayer [22].

The Protein-Protein Effects and Protein Secondary Structure
The influence of protein-protein interactions on the secondary structure of adsorbed lysozyme is presented in Figure 2. As shown, secondary structures of lysozyme adsorbed from increased solution concentration were comparatively retained on the pHEMA surface. There was a significant reduction in the helical structure. In contrast, a substantial increase in percent sheet content was observed for the lysozyme layers adsorbed from solutions at lower protein concentration when compared to native protein structure in physiological saline.  Secondary structural contents (%) of adsorbed lysozyme on pHEMA surface from different solution concentrations. The secondary structure of lysozyme measured in physiological saline solution was studied as the control. The error bars denote the mean ±SD for n = 3. *and ** represent p < 0.05 and p < 0.01, respectively. Figure 3 presents a plot of the percent helical content of the adsorbed lysozyme on pHEMAbased contact lens (Polymacon) with corresponding friction coefficient values for each protein solution concentration. As shown, the coefficient of friction values was reduced when the lysozyme was adsorbed from the diluted protein solution concentrations, with the increased secondary structural changes as indicated by the increased helical content (%).

Discussion
The pHEMA is a biocompatible and water-absorbing material used to make contact lenses or medical delivery systems [42,43]. The presence of carbonyl (C=O) and one hydroxyl (-OH) groups on each side chain in pHEMA subsequently reduce the hydrophobicity of the surface [44].
As indicated in Figure 1b, the interaction of the lysozyme with the pHEMA occurs rapidly, with proteins adsorbed from different solution concentrations resulting in the widely distributed areal densities, as shown in Figure 1a. The initial interaction of a protein with a surface may occur as a dynamic exchange process with adsorption rate and desorption rate represented by kon and koff, respectively. However, adsorbed proteins may undergo surface-induced unfolding and spread out over the surface to increase their footprint on the surface [41] with the unfolding rate expressed as ks, which is usually an irreversible process [20]. When protein is adsorbed from higher solution concentration, there is a subsequent increase in adsorption rate. When the adsorption rate is faster

Discussion
The pHEMA is a biocompatible and water-absorbing material used to make contact lenses or medical delivery systems [42,43]. The presence of carbonyl (C=O) and one hydroxyl (-OH) groups on each side chain in pHEMA subsequently reduce the hydrophobicity of the surface [44].
As indicated in Figure 1b, the interaction of the lysozyme with the pHEMA occurs rapidly, with proteins adsorbed from different solution concentrations resulting in the widely distributed areal densities, as shown in Figure 1a. The initial interaction of a protein with a surface may occur as a dynamic exchange process with adsorption rate and desorption rate represented by k on and k off , respectively. However, adsorbed proteins may undergo surface-induced unfolding and spread out over the surface to increase their footprint on the surface [41] with the unfolding rate expressed as k s , which is usually an irreversible process [20]. When protein is adsorbed from higher solution concentration, there is a subsequent increase in adsorption rate. When the adsorption rate is faster than the unfolding rate of an adsorbed protein on the surface (i.e., k on >> k s ), the degree of spreading of a protein over the surface will be inhibited by the neighboring adsorption sites being occupied by other adsorbing proteins. This condition is indicated by the significant protein-protein interactions with the saturated monolayer coverage of protein on the surface (i.e., minimized footprint). In this study, the areal densities of adsorbed lysozyme correspond to the theoretical limits for a saturated surface for a monolayer of protein organized in a close-packed side-on orientation (0.17 µg/cm 2 ) and close-packed end-on orientation (0.26 µg/cm 2 ) [22]. On the other hand, when protein is adsorbed from diluted solution concentrations, there is a relatively higher unfolding rate when compared to the adsorption rate (i.e., k on << k s ), in which case the proteins on the surface tend to spread out and reach their maximum footprint depending on the protein-surface interactions involved before the neighboring sites being occupied, with a relatively lower amount of protein being adsorbed, which presents the minimized protein-protein interactions [23]. The broad range of areal density observed in Figure 1, which could be controlled by varying the protein solution concentrations from which the protein is adsorbed, thus could be proportionally correlated with a broad distribution of protein-protein interactions involved. For example, when adsorbed from the diluted solution concentration, there will be a lower areal density of adsorbed proteins with subsequent weaker protein-protein interactions.
The data from Figure 2 further present a much clearer correlation between the protein-protein interactions and the stability of adsorbed lysozyme on a pHEMA surface. The increased protein-protein interactions were represented by the higher real density of adsorbed proteins, from which it is clearly shown that protein-protein interactions tend to stabilize the structure of adsorbed lysozyme on pHEMA. Little structural changes were observed when compared to the native lysozyme structure in physiological saline. For example, the highest areal density (i.e., representing the most significant protein-protein interactions involved) was observed when adsorbing the protein from a solution concentration of 1.9 mg/mL. There is only a 15% decrease in the native-state percent sheet (38% to 33%) and a 15% increase in the native-state percent helicity (30% to 35%), while at the lowest areal density observed when protein was adsorbed from 0.5 mg/mL solution concentrations, the protein-protein interaction was minimized.
The protein-surface interactions further induced the sheet and helical structures to decrease to only 28% (i.e., 25% loss in percent sheet) and increase to 58% (i.e., more than 90% increase in percent helicity), respectively.
Based on these results observed, the pHEMA surface with a large density of hydrogen bondable groups may interact with lysozyme by competing with the hydrogen bonds that stabilize the protein secondary structures on the surface [11,45], from which the lysozyme structure is destabilized due to the protein-surface interactions involved. On the other hand, the protein-protein interactions tend to inhibit the unfolding of adsorbed lysozyme from neighbor proteins, thus helping to stabilize the protein structures on a pHEMA surface.
Finally, an apparent correlation was observed between the coefficient of friction measured from the protein covered contact lens surface (Polymacon) and percent helical content within the adsorbed lysozyme layer. Polymacon is a pHEMA-based material with a nonionic hydrophilic 2-hydroxyethylmethacrylate (HEMA) cross-linked with ethylene glycol dimethacrylate [46]. Our results suggest that the friction coefficient is primarily being influenced by the structures of adsorbed proteins, with the friction coefficient reduced from the increased helical content from the native state structures, reflecting the reduced friction forces between two solid surfaces at a constant load, in which case, when compared to the control group (i.e., pure physiological saline at the interface), the lysozyme in the native structure was shown to be a poor boundary lubricant, which might be considered as a high-shear-strength layer at the interface. On the other hand, the lubricating properties of the unfolded lysozyme upon adsorption on a Polymacon surface might be attributed to the hydrophilic moieties exposed to the aqueous environment. Thus, the water molecules trapped at the surface might provide a lubricious, low-shear-strength, fluid film [11,12], as shown in Figure 4.
Therefore, when these data are compared with results from Figure 2, they further suggest that protein-protein interactions primarily influence the friction coefficient involved by preserving the structures of adsorbed lysozyme on a pHEMA, which in turn increases the friction coefficient observed. However, a complete study considering more complex model fluids, including the ionic effects, interactions between multiple proteins, and physiological temperature settings to present the biomolecules within the tear film, is required in our future study. structures of adsorbed lysozyme on a pHEMA, which in turn increases the friction coefficient observed. However, a complete study considering more complex model fluids, including the ionic effects, interactions between multiple proteins, and physiological temperature settings to present the biomolecules within the tear film, is required in our future study.

Conclusions
In this study, the relationship between the adsorption responses of lysozyme from varied protein solution concentrations and the friction between the surfaces of protein-covered contact lens material was studied at a molecular level. The low friction between two solid surfaces might be attributed to the fluidity of the hydrated and unfolded lysozyme layers from which the degree to which proteinprotein interactions limit the unfolding of lysozyme on a pHEMA surface can thus be related to protein lubricating properties and be simply controlled by adjusting the concentration of the protein in solution, which influences the CLD of lens wear.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Figure S1: X-ray photoelectron spectrometer of pHEMA used in this study. Figure S2: AFM images and surface roughness analysis for commercially available contact lenses and pHEMA surface made with different coating parameters.

Conclusions
In this study, the relationship between the adsorption responses of lysozyme from varied protein solution concentrations and the friction between the surfaces of protein-covered contact lens material was studied at a molecular level. The low friction between two solid surfaces might be attributed to the fluidity of the hydrated and unfolded lysozyme layers from which the degree to which protein-protein interactions limit the unfolding of lysozyme on a pHEMA surface can thus be related to protein lubricating properties and be simply controlled by adjusting the concentration of the protein in solution, which influences the CLD of lens wear.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4360/12/8/1639/s1, Figure S1: X-ray photoelectron spectrometer of pHEMA used in this study. Figure S2: AFM images and surface roughness analysis for commercially available contact lenses and pHEMA surface made with different coating parameters.