Low-Fouling Characteristics of Ultrathin Zwitterionic Cysteine SAMs

Subscriber access provided by The Libraries of the | University of North Dakota

Biological and Environmental Phenomena at the Interface

Low-fouling characteristics of ultrathin zwitterionic cysteine SAMs Peter Lin, Tsung-Liang Chuang, Paul Z. Chen, Chii-Wann Lin, and Frank X. Gu Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b01525 • Publication Date (Web): 30 Jul 2018 Downloaded from http://pubs.acs.org on July 31, 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 34 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

Low-fouling characteristics of ultrathin zwitterionic cysteine SAMs Peter Lin,† Tsung-Liang Chuang,‡ Paul Z. Chen, † Chii-Wann Lin, ‡ and Frank X. Gu*†§ †

Department of Chemical Engineering, Waterloo Institute for Nanotechnology, University of

Waterloo, Waterloo, Ontario N2L 3G1, Canada ‡

Graduate Institute of Biomedical Engineering, Department of Electrical Engineering, National

Taiwan University, Taipei 106, Taiwan §

Department of Chemical Engineering & Applied Chemistry, University of Toronto, Toronto,

Ontario M5T 3A1, Canada KEYWORDS: antifouling surface, implantable bioelectronics, biofouling, cell adsorption, protein adsorption, quartz crystal microbalance, self-assembled monolayer, surface plasmon resonance imaging

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 34

ABSTRACT Surface fouling remains an exigent issue for many biological implants. Unwanted solutes adsorb to reduce device efficiency and hasten degradation while increasing the risks of microbial colonization and adverse inflammatory response. To address unwanted fouling in modern implants in vivo, surface modification with antifouling polymers has become indispensable. Recently, zwitterionic self-assembled monolayers, which contain two or more charged functional groups but are electrostatically neutral and form highly hydrated surfaces, have been the focus of many antifouling coatings. Reports using various compositions of zwitterionic polymer brushes have demonstrated ultralow fouling in the ng/cm2 range. These coatings, however, are thick and can hinder the target application of biological devices. Here, we report an ultrathin (8.52 Å), antifouling self-assembled monolayer composed of cysteine that is amenable to facile


2

Page 3 of 34 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

fabrication. The antifouling characteristics of the zwitterionic surfaces were evaluated against bovine serum albumin, fibrinogen and human blood in real time using quartz crystal microbalance and surface plasmon resonance imaging. Compared to untreated gold surfaces, the ultrathin cysteine coating reduced the adsorption of bovine serum albumin by 95% (43 ng/cm2 adsorbed) after 3 h and 90% reduction after 24 h. Similarly, the cysteine self-assembled monolayer reduced the adsorption of fibrinogen as well as human blood by >90%. The surfaces were further characterized using scanning electron microscopy: protein-enhanced adsorption and cellular adsorption in human blood was found on untreated surfaces but not on the cysteine SAM-protected surfaces. These findings suggest that surfaces can be functionalized with an ultrathin layer of cysteine to resist the adsorption of key proteins, with performance comparable to zwitterionic polymer brushes. As such, cysteine surface coatings are a promising methodology to improve the long-term utility of biological devices.


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 34

INTRODUCTION Recent advances in biomaterials promise novel and more effective implantable devices.1-3 Key challenges that hinder the utility of these implantable devices, however, remain: devices can degrade and must retain long-term biocompatibility and effectiveness in vivo,4-6 and these issues can preclude implantable devices from widespread adoption.7-9 For example, FDA-approved implantable glucose sensors decrease issues with patient compliance but currently see a reduction in performance after only seven days post-implantation.10 Non-specific protein fouling occurs rapidly after implantation and is a key contributor to the aforementioned issues.11-14 Upon surgical insertion, plasma proteins spontaneously cover the surface of the implant, typically triggering a cascade of host responses towards fibrous encapsulation.15-17 In addition, the adsorbed proteins can condition the surface for microbial colonization and subsequent biofilm formation.9,

18-20

Thus, protein fouling can limit the

interaction of the implanted device with its target host environment, compromising utility, and lead to adverse health effects. These consequences often necessitate secondary treatment or surgery. Recent methodologies extend the lifetime of implantable devices by minimizing the adsorption of non-specific plasma proteins at the initial stage of implantation.12, 18, 21 Often, the surfaces are functionalized with self-assembled zwitterionic molecules, which form a tightly solvated layer of water molecules,22-23 to impart antifouling capabilities.24-26 Compared to conventional hydrophilic polymers, such as polyethylene glycol, many recently developed zwitterionic surfaces exhibit superior performance in resisting protein fouling.9, 27-28 Early work by Jiang et al. have demonstrated the feasibility of using carboxybetaine and sulfobetaine for zwitterionic


4

Page 5 of 34 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

anti-fouling materials.29-31 Chen et al. have reported ultralow protein adsorption (90% reduction in adsorption in all three biological solutions when compared to untreated surfaces), which are comparable to zwitterionic polymer brushes when considering the experimental conditions. Surfaces coated with cysteine SAMs were also imaged using scanning electron microscopy (SEM) and found to avoid cellular adsorption in human blood. These findings demonstrate cysteine to be a promising molecule to form ultrathin antifouling coatings.

MATERIALS & METHODS

Materials. L-cysteine (97% purity), BSA, and Fg from human plasma were used as received from Sigma-Aldrich. Whole human blood in BD Vacutainer plastic tube coated with 7.2 mg of K2 EDTA was provided by National Taiwan University Hospital (NTUH). Phosphate buffered saline (PBS) was purchased from Invitrogen and Sigma-Aldrich. All chemicals were used without further purification. Quartz-Crystal Microbalance. The QCM measurements were performed at room temperature with the AT-cut quartz crystal from ANT technology Co., Ltd. Frequency shift (∆F) was detected by the Affinity Detection System (ADS) type QCM instrument from the same manufacturer. The frequency of the loaded crystal ( ) was 8.97 MHz with a standard deviation of 0.026%. The piezoelectrically active crystal area and reaction chamber volume is 0.1 cm2 and 30 µL respectively, as given by the manufacturer. For all QCM analyses conducted in this study, the samples were first allowed ample time to establish a stable baseline under a continuous flow of PBS before a biological solution (BSA, Fg or human blood sample, 6.6 mg/ml was used for the protein solutions) was introduced to the system. The time required to establish a steady baseline varied among samples (10-45 min), and only the portion of data directly proceeding the


6

Page 7 of 34 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

introduction of a biological solution was presented in the figures to allow for easy viewing of the data. The baseline was considered stable when the change in resonance frequency falls within the machine’s inherent signal drift given by the manufacture as ±0.3 Hz per minute. The signal drift, which is considered as background noise, is removed from the data using a calibration process provided by the manufacture’s signal acquisition software. After exposing the sample surface to a biological solution for the desired amount of time, a PBS solution was again flowed across the sample surface and the resonance frequency is allowed time to stabilize. The change in frequency, , is calculated by subtracting the frequency of PBS baseline from PBS wash. The flow speed was kept at 30 µL/min. Cysteine coated surfaces were fabricated by introducing a 0.1 mM cysteine solution to the gold portion of QCM chip and allowed to react overnight in accordance with our previous work.25 Surface Plasma Resonance Imaging (SPRi). Trapezoidal cyclic olefin copolymer (COC) prisms (n = 1.51) were designed and fabricated with precise angles and dimensions by injection molding at Silitech Technology Co, Inc. A 47 nm gold film was deposited onto the prism using radio-frequency (13.56 MHz) sputtering system at a working pressure of approximately 3 × 10-3 Torr. Double-sided adhesive (F9460PC, 3M, USA) was cut with a laser machine and used to join the prism to the sample holder. The SPRi system was an in-house custom-made model designed for real-time monitoring of reflectivity changes based upon the Kretschmann configuration. The optical module consists of a 1 W 850 nm near-infrared (NIR) LED, an achromatic doublet, and a polarizer that provides a p-polarized collimated beam for irradiating the surface of the 47 nm gold film on the prism. The light was coupled using a COC prism to generate a surface plasmon resonance and was reflected onto a CCD camera connected to a computer through an IEEE 1394 interface. The


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 34

incidence angle of the light source was manually adjusted so that the value of the reflective intensity was one-third of the difference between the maximum and the minimum reflection. The reflected light was detected by a 12-bit grayscale CCD camera. The aforementioned SPRi subsystems were controlled by a program developed in-house using LabVIEW 8.2. Additional information on this custom-made SPRi system can be found in our previously published paper.39 A liquid injection system was used to first introduce a PBS buffer to the SPR prism in order to acquire a steady baseline before exposing the surfaces to the biological analyte and followed by a PBS buffer wash. The change in reflective intensity (∆RU) is calculated by subtracting the signal of the PBS baseline from that of the post reaction PBS wash. The SPR measurements were completed under zero-flow and, to prevent any dilution effects, at least 3 mL of each solution was injected to flush out the reaction chamber (30 µL). Scanning Electron Microscopy. A field emission scanning electron microscope (FE-SEM, Zeiss Leo 1550) was used for the scanning electron microscopy (SEM) analysis. The electron gun energy was kept at 10.00 kV and the vacuum was kept approximately at 6x10-6 mbar. The samples were subjected to SEM analysis without any additional sample preparation.

RESULTS & DISCUSSION

QCM Calibration Detection of trace amounts of protein adsorbed onto antifouling surfaces is often difficult and requires extremely sensitive instruments. For this study, quartz crystal microbalance was selected as the preferred method to measure surface fouling because the instrument translates real-time changes in resonant frequency directly to changes in adsorbed mass. This section discusses QCM calibration and explains all artifacts produced as a result of the experimental procedure. When


8

Page 9 of 34 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

one face of the quartz crystal resonator is in contact with a liquid, the change in frequency (ΔF) can be expressed by Equation 1,40-42 =−

= −

2

μ 2

μ

−

μ

−

μ

= − −

Equation 1

Where is the density of quartz crystal (2.648 g·cm-3), μ is the shear modulus of quartz (2.947x1011 g·cm−1·s−2), is the frequency of the loaded crystal, is the piezoelectrically active crystal area (given by the manufacturer as 0.1 cm2), is the change in mass, is the viscosity of the liquid and is the density of the liquid. From Eq. 1, the resonant frequency can be disturbed by either a change in surface mass (caused by film deposition or decay), a change in viscosity of the liquid or a change in density of the liquid. If the change to resonator mass is assumed to be zero, thus ignoring the first term of Eq. 1, the equation provides information on the change of resonant frequency when the system is disturbed from a gaseous interface to a liquid interface. Using a crystal with a load frequency ( of 8.96 MHz as an example, when the crystal interface transforms from a vacuum interface to a pure water interface, the resonance frequency is expected to decrease by approximately -1715 Hz (calculated using Equation 1 at 20 °C and assuming zero change in mass). On the reverse, if a gas bubble is introduced to a liquidcrystal interface, an increase in resonance frequency is expected. This phenomenon was observed on some graphs presented in this QCM study when small gaseous bubbles were introduced to the liquid interface amidst the exchange between different solutions. These gaseous bubbles created


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 34

temporary spikes in resonance frequency as the bubbles traveled across the liquid-crystal interface. Operating the QCM under stable conditions, where the change in root product of viscosity and density of the liquid ( ) is negligible, the change in frequency directly reflects the change in surface adsorbents. Therefore, the addition or removal of mass on the surface of the crystal resonator is proportional to the change in resonate frequency by the intrinsic constant ( = − ). Since the intrinsic frequency of the loaded quartz crystal ( varies slightly between each chipset, variation between the intrinsic constants ( ) of each chip was also expected. However, the variance is very low (0.026%) for the chipsets used in this study and deposited mass can be related to the resonate frequency by: #$ = −0.549 ± 0.00030Hz

Equation 2

The change in resonance frequency was continuously monitored and recorded in real-time. The QCM figures are composed of three portions (PBS baseline, exposure and PBS wash) and the portions are separated by dotted red lines. is calculated by subtracting the stable frequency value acquired during PBS baseline from that acquired during PBS wash; therefore, the wash procedures employ the same buffer solution as is used during the for accurate calculation. Any biological substances that are weakly adsorbed to the surface are also been removed during the post-exposure PBS wash. The flow speed was transiently increased when switching between solutions to prevent a splash-over effect by ensuring that the new solution was introduced to the sample surface as soon as possible. If the flow speed was not increased, the newly introduced biomolecules may diffuse into the previously introduced buffer solution in the machine tubing and create an unnecessary concentration gradient. The necessity of increased flow speed combined with the introduction of


10

Page 11 of 34 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

small air bubbles oftentimes created significant spikes or dips in frequency readings that shortly follow the introduction of a new solution. These artifacts of the experimental procedure resolve themselves, and samples were allowed ample amount of time to return to an equilibrium reading.

Monolayer Fabrication The fabrication of a thin protective monolayer is essential in maximizing the interaction of biological devices with their host environment while reducing unwanted surface fouling. QCM was used to monitor the formation of the SAM overnight as well as determine the final coating thickness. Cysteine coatings were fabricated by self-assembling cysteine onto the gold portion of a QCM chipset in accordance with our previous work, which demonstrated cysteine-coated surfaces to be highly zwitterionic,25 and the reaction was monitored overnight under a continuous flow (Figure 1). The system was allowed ample time to established a steady baseline under a continuous flow of PBS before the cysteine solution was introduced. To aide visualization, Figure 1 incorporates two modifications to the x-axis: First, only the last 5 min of the PBS baseline is shown as additional data points do not contribute any new information. Second, only the initial 3 h of the overnight reaction is shown in order to preserve the adsorption trend; after 3 h of exposure, the change in frequency plateaus and additional changes are very gradual with the exception of spikes caused by the introduction of air bubbles. Subsequent to the overnight reaction, a PBS wash was employed to remove any physisorbed layers of cysteine. The PBS wash portion of Figure 1 illustrates the signal approaching a plateau after 60 min, indicating that most of the secondary layer are removed. Nevertheless, the data trend suggests that lengthening the PBS wash step will remove additional physisorbed molecules. Indeed, almost 12 h are required to completely remove the physisorbed molecules (Figure S1). The dip in signal observed when switching from the PBS to cysteine solution is the result of an artifact of the


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 34

experimental process. The total change in frequency, calculated by subtracting the reading during PBS baseline from the plateau portion of PBS wash, is -19.75 Hz which translates to 10.84 ng of cysteine adsorption in accordance with Eq. 2. A typical Au(111) surface is reported to have 4.5x1014 adsorption sites/cm2 when maximum coverage is obtained37, 43 or 9.05 ng of cysteine on a 0.1 cm2 gold surface. Based on these findings, the fabricated cysteine surfaces have a complete coverage and the thickness is 1.20 monolayers. An upright cysteine molecule chemisorbed through the thiol group onto a gold surface is reported to have a height of 7.1Å;44-46 a 1.20 monolayer coating translates to an average coating thickness of 8.52 Å.

Figure 1. Fabrication of a cysteine surface. To improve data visualization, this figure only incorporates data from the last 5 min of PBS baseline, initial 3 h of thiol reaction, and 60 min of post-exposure PBS wash. The x-axis is rescaled for the PBS portions to preserve the data trend. of −19.75 Hz was observed, translating to 10.84 ng cysteine adsorption or a coating thickness of 8.52 Å.

Biological Fouling


12

Page 13 of 34 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

This study employs a dynamic QCM to investigate the antifouling property of cysteine SAMs against single protein solutions composed of BSA or Fg as well as human blood. Untreated gold surfaces were used as the control in this study. BSA was selected as a model protein to assess nonspecific fouling primarily because many of its characteristics such as size, charge, and amino acid sequence are well understood47-48 and it closely resembles human serum albumin (HSA), the predominant plasma protein in the blood. Furthermore, the literature uses BSA as a standard molecule to assess surface biofouling.9, 49-51 Figure 2a illustrates the change in resonant frequency with respect to time for an untreated gold surface when exposed to a BSA solution. This figure depicts a 30 min PBS baseline before a 3 h BSA exposure followed by a PBS wash. The PBS wash was allowed to continue until the signal plateaued, thus ensuring the removal of all weakly adsorbed BSA molecules. of -147.11 Hz was observed, which translates to 80.77 ng of BSA fouling onto the untreated gold surface. Figure 2b reports BSA adsorption onto a cysteine SAM-coated surface. This figure depicts a 45 min PBS baseline followed by a 3 h BSA exposure. The PBS wash was allowed to continue for the same duration of time as what had transpired during the BSA-on-gold experiment. A resulting of -8.02 Hz was observed which translates to 4.40 ng of BSA fouling. Compared with the untreated gold surface, the cysteine coated surface was able to reduce BSA fouling by 94.55%. The QCM results agree with the SEM images (Figures 6E and 6F), where spherical BSA clusters were observed throughout the untreated gold surface but these clusters were very rarely observed on the cysteine surface. To further demonstrate the anti-fouling property of zwitterionic cysteine, a 24 h exposure to BSA was conducted under the same experimental conditions and 8.03 ng of BSA fouling was observed on the cysteine SAM (QCM figure not shown).


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 34

Figure 2. BSA adsorption onto an (a) untreated gold surface and (b) a cysteine SAM-coated surface for 3 h. The total amount of BSA fouling was 80.77 ng and 4.40 ng respectively. Compared with the untreated gold surface, the cysteine SAM-coated surface reduced BSA fouling by 92.78%. Fg is a key protein involved in the clotting cascade and is widely viewed as a significant factor determining the longevity of implantable devices.52-53 Figures 3a and 3b illustrate the amount of Fg adsorbed onto an untreated gold and a cysteine SAM-coated surface. For the gold surface, a 10 min PBS baseline was followed by 3 h of Fg exposure. To remove weakly bound Fg, the postexposure PBS wash was allowed to continue until the signal plateaued. A of -180.59 Hz was observed, which translates to 99.10 ng of fibrinogen fouling onto the untreated gold surface. Considering the impressive performance of cysteine in resisting BSA fouling, a lengthened protein exposure was conducted for the cysteine surface. Fg was exposed to the cysteine SAMcoated surface for 14 h. The PBS wash was allowed to continue for the same duration of time as what had transpired during the Fg-on-gold experiment. To improve visualization of these data, only the initial 10 h of the exposure is shown. In addition, the x-axis has been rescaled for the PBS baseline and the PBS wash portions to preserve the data trend. A of -16.56 Hz was


14

Page 15 of 34 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

observed, which translates to 9.09 ng of fibrinogen fouling onto the cysteine SAM. Compared with the untreated gold surface, the cysteine SAM-coated surface was able to reduce Fg fouling by 90.82%. The QCM results agree with the SEM images (Figures 6I to 6L), which reveals large clusters of Fg on the gold surface, whereas the clusters are absent on the cysteine surface. The adsorption of albumin onto both surfaces is initially rapid and, as coverage increases, adsorption begins to slow as proteins compete for fewer available sites. This single-step adsorption profile is expected since albumin is globular, and thus, adsorption is orientationindependent.47 In contrast, the adsorption processes for Fg appears to occur in a multistep fashion (Figure 3B), where the protein initially adsorbs rapidly in a flat orientation saturating the surface with its long axis parallel to the sample surface to maximize interactions. After some time, Fg begins to rearrange into a perpendicular orientation that exposes additional adsorption sites on the surface and allows for greater packing. This second adsorption step is more gradual when compared with the initial step, as expected, since the rearrangement of surface proteins is a slow process. A schematic depiction of Fg’s multistep adsorption process is shown as the inserts in Figure 3B. The rearrangement of adsorbed Fg and its multistep adsorption profile have been reported in other QCM studies54-55 and are believed to be driven by entropic interactions between the long axes of Fg.


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 34

Figure 3. Fg adsorption onto (a) an untreated gold surface for 3 h and (b) a cysteine SAM-coated surface for 14 h. To improve data visualization, only the initial 10 h of exposure is shown for the cysteine surface and the x-axis has been rescaled for the PBS portions to preserve the data trend. The total amount of Fg fouling was 99.10 ng and 9.09 ng, respectively. When compared with the untreated gold surface, the cysteine SAM-coated surface reduced Fg fouling by 90.82% in spite of the increased exposure time. The foreign body response to implants depends on the response of adhesion receptors on inflammatory cells to adsorbed proteins. The presence of adsorbed proteins such as albumin, Fg, fibronectin, vitronectin and γ globulin, regulate inflammatory cell interactions and adhesion, and thus, are critical determinants of subsequent degradation and isolation cascade.56-59 To fully investigate this complex interaction, a half-diluted human blood sample (50%HBS) was applied to cysteine and gold surfaces. The dilution of HBS was necessary for QCM characterization: during experiments involving whole HBS, the QCM detector became immediately unresponsive upon exposure. We hypothesize the loss of signal was caused by the high intrinsic viscosity of undiluted human blood (typically 4-5 times greater than water). By exposing the quartz crystal to a solution with high viscosity, a sharp change in resonant frequency is expected in accordance


16

Page 17 of 34 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

with Eq. 1; the sudden change in frequency is believed to be the cause of the irreversible signal loss from the QCM detector. Figure 4a and 4b illustrates the adsorption profile of biological substances onto untreated gold and a cysteine SAM-coated surfaces. The gold surface was exposed to 50%HBS for 1 h followed by PBS wash. The was -3012.06 Hz, which translates to 1653.68 ng of adsorbed mass on the untreated gold surface. The QCM response for the untreated gold surface is very jagged and exhibits sudden rapid decreases (for example, from 30 to 35 min, and again from 47 min to 53 min), which greatly differs from the smooth, continuous adsorption profile exhibited by the homogeneous protein studies presented earlier. This observation indicates a strong case of protein-enhanced adsorption where the presence of specific human blood proteins on a surface induces further adsorption of other proteins and cells through complex biological interactions.60 Indeed, images captured by the SEM showed ruptured cells throughout the untreated gold surface (Figure 6P). Considering the impressive performance of cysteine in resisting BSA and fibrinogen fouling, an expanded exposure was again conducted for the cysteine-coated surface. It was exposed to 50%HBS for 1.5 h before the PBS wash. The was -217.36 Hz, which translates to 119.20 ng of biological fouling onto the cysteine surface. Compared with the untreated gold surface, the cysteine SAM-coated surface reduced biofouling by 92.79%. Furthermore, the QCM response here was smooth and leveled off after 300 s of exposure. Additional decrease in signal was gradual and the adsorption trend is very similar to that of the homogeneous protein studies presented earlier. This adsorption profile suggests that protein-enhanced adsorption is insignificant on a cysteine SAM-coated surface. SEM images support this observation as there were no indications of adsorbed cells on the cysteine SAM-coated surface.


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 34

Figure 4. Half-diluted human blood exposure onto (a) an untreated gold surface for 1 h and (b) a cysteine SAM-coated surface for 1.5 h. Adsorption data for the untreated gold surface is jagged and exhibits sudden decreases (from 30 min to 35 min and from 47 min to 53 min), which greatly differs from the smooth and continuous adsorption profile exhibited by homogeneous protein studies presented earlier. This observation indicates protein-enhanced adsorption, which is not observed on the cysteine SAM-coated surface. The total amount of biological fouling was 1653.28 ng and 119.20 ng for gold and cysteine surfaces, respectively. Table 1 summarizes the results of the QCM study after replication (n = 3 for BSA and Fg). Our design outperforms alternative methodologies employing cysteine significantly (42.95 ng/cm2 BSA fouling as opposed to 66 µg/cm2,61 and 5.19% BSA fouling as oppose to 55%62). The improved performance may be due to careful fabrication of a complete monolayer—without secondary and tertiary layers, which properly orients cysteine molecules to expose their zwitterionic moieties. Similarly, our results are comparable to, if not exceeds, the performance of recent reports of antifouling hydrogels,63 some zwitterionic coatings,64-66 cellulose structures,67 PEG-based polymer brushes68-69. Dense long-chained zwitterionic polymer brushes are known to achieve ultralow BSA fouling (

Low-Fouling Characteristics of Ultrathin Zwitterionic Cysteine SAMs

Recommend Documents