Highly Hydrophobic Polytetrafluoroethylene Particle Immobilization via Polydopamine Anchor Layer on Nitric Oxide Releasing Polymer for Biomedical Applications

Abstract

Biomedical surface-associated infections and thrombus formation are two major clinical issues that challenge a medical device’s fate in the body and patient safety. Single platform multifunctional surfaces are need of the hour to address both these indwelling medical device-related problems. In this work, bio-inspired approaches are employed to fabricate a polymer composite with a versatile surface that can reduce bacterial infections and platelet adhesion in vitro. In the first bio-inspired approach, the functionality of nitric oxide (NO) produced by endothelial cell lining of the blood vessel is mimicked by incorporating S-nitroso-Nacetylpenicillamine (SNAP) within CarboSil-2080A™ (CarboSil) polymer composite matrix. The second approach involves utilizing mussel adhesive chemistry, polydopamine (PDA), to immobilize polytetrafluoroethylene (PTFE) particles on the polymer composite surface. The PTFE coating facilitates a decrease in wettability by making the polymer composite surface highly hydrophobic (contact angle ca. 120°). The fabricated polymer composite’s surface, CarboSil SNAP-PTFE, had a cobblestone-like structured appearance as characterized through scanning electron microscopy (SEM). Water contact angle (WCA) and surface tension measurements indicate no significant coating losses after 24 h under physiological conditions. NO surface flux was measured and analyzed for 5 days using a chemiluminescence-based nitric oxide analyzer and was found to be within the physiological range. CarboSil SNAP-PTFE reduced adhered bacteria (99.3 ± 0.5 % for Gram-positive S. aureus and 99.1 ± 0.4 % for Gram-negative E. coli) in a 24 h in vitro study. SEM study also showed the absence of biofilm formation on CarboSil SNAP-PTFE polymer composites, while present on CarboSil in 24 h exposure to S. aureus. Platelet adhesion was reduced by 83.3 ± 4.5%. Overall, this study’s results suggest that for NO-releasing CarboSil, a combination with PTFE coating can drastically reduce infection and platelet adhesion.

Graphical Abstract

1. Introduction

The use of medical devices has increased drastically over the years, as they have been instrumental in mitigating diseases that otherwise had limited treatment options. Despite the steady increase in medical devices, its use has often faced complications such as medical device-related infections or thrombosis [1, 2]. Medical device-related infections comprise about 40–60 % of the cases of hospital-associated infections leading to spikes among patient mortality and morbidity rates as well as the overall financial burden [3]. The incidence of device-related thrombus formation in blood-contacting implants such as left atrial appendage closure (LAAC) with Watchman device has been reported to be up to 17.6% [4]. In the case of extracorporeal life support treatments, device induced clotting has been reported to occur in 22% of cases [5].

Bacterial adhesion and proliferation have posed a critical challenge towards all biomedical surfaces in healthcare settings [6]. Microbial surface sensing is an outcome of adaptive behavior through which a bacteria can undergo morphological and physiological changes to favor attachment to the surface [7]. Near surfaces (bulk polymer and nanomaterials) planktonic bacteria interact via electrostatic or/and hydrodynamic interactions [7–10]. This process is reversible, and with time the hydrophobic regions of the bacterial cell wall start interacting with the surface via van der Waal’s forces. Besides, several conformational changes of cell surface proteins, re-orientation of bacteria, and physicochemical bond strengthening lead to an irreversible adhesion and eventually forming community-like structures called biofilms [7, 11]. Longstanding effort towards curbing surface colonization and arresting bacterial growth has resulted in mixed outcomes. Over the years, the infection problem has been aggravated due to the formation of biofilms on device surfaces and the effects of antibiotic resistance [6, 12]. Biofilms make it difficult for antibiotic therapy by forming an extra polymeric matrix, which protects the bacterial colonies from antimicrobial agents [13]. Most clinically relevant bacterial strains such as Staphylococcus aureus, Staphylococcus epidermidis, Escherichia coli, Providencia stuartii, Pseudomonas aeruginosa, etc. can form biofilms [14, 15]. For an indwelling or blood-contacting medical device, thrombosis by itself posited as a major problem apart from complementing bacterial adhesion in such scenarios or vice versa [16, 17]. Blood clots arising from devices can embolize and potentially cause stroke in a completely different part of the body [18]. The current treatment options of using blood thinners, such as heparin, in the form of systemic anticoagulation or anti-platelet therapy are not without its own set of disadvantages. Heparin induced thrombocytopenia, internal bleeding or even patients allergic to such blood thinners are some of the negative effects of such treatments [19, 20].

Most antibacterial and anti-thrombotic strategies have transitioned from a single mechanistic approach to multi-pronged systems. Primary among these strategies is a combination of a passive surface with active agents, which would result in repelling/preventing adhesion first and then killing or preventing adhesion/activation [21]. For this study, conjugation of a release-based approach and an anti-adhesion approach has been explored. The key to prevention or limiting the spread of infection or onset of thrombosis is in its early stages. For the release-based approach, a local systemic antibacterial and anti-platelet activity are obtained through NO. The bio-inspired active agent, NO, is a gasotransmitter produced from endothelial cells in the blood vessels that have gained traction in recent years as a potent antimicrobial and anti-thrombotic agent apart from other physiological functions [22]. NO is synthesized in the body when arginine is converted to citrulline catalyzed by the nitric oxide synthase (NOS) enzyme system [23]. NO has been implicated in the regulation and dispersal of biofilms and a mediator of quorum sensing [24–26]. The antimicrobial action of NO is ascribed to its ability to form peroxynitrite, other reactive nitrogen or oxygen species which can attack DNA or other intracellular targets, lipid peroxidation leading to cell surface disruption, etc. [22, 27]. In the past couple of decades, NO donor molecules have been developed and incorporated into polymers to mimic endothelial cell linings of blood vessels, which act as repositories to produce surface NO flux under physiological conditions. Work done by our group and others has shown that exogenous NO donor molecules such as Snitrosothiols (RSNO), N-diazeniumdiolates, organic nitrates, and nitrites can be extensively used to improve antibacterial activity against clinically relevant bacteria like S. aureus, E. coli, P. aeruginosa, etc., as well as anti-platelet activity [28–42]. Among all donor molecules, RSNOs have demonstrated exceptional steady release rates and biocompatibility [43–45]. However, a drawback of NO-releasing materials is that NO release has been shown to facilitate surface fouling of non-specific proteins [46]. Surface fouling can lead to an increased risk of thrombus formation or bacterial adhesion/biofilm formation [47, 48]. One way to address the issue is to modify the surface or utilize antifouling coating on the NO-releasing material. The antifouling surface coating can synergistically improve NO-releasing material by passively averting biofilms or thrombus formation [49–51].

To develop a passive antifouling surface, either the surface is rendered hydrophobic or hydrophilic to resist bacterial or platelet adhesion [21, 50, 51]. PTFE, widely used as a medical device coating, is a polymer of choice for many indwelling device applications due to its chemical inertness, low coefficient of friction, biocompatibility, etc. [52–54]. PTFE exhibits high hydrophobicity due to its low surface energy [55]. Structural changes of the boundary layer of water (in aqueous environments) in the vicinity of such hydrophobic layers are observed, which are currently understood through the concept of nanobubble formation on hydrophobic surfaces[56]. Therefore, three key players can potentially be involved in resisting biomolecule adhesion to highly hydrophobic surfaces such as that of PTFE – properties of the biomolecules, hydrophobic surface itself, and the air-water interface (also called plastron). Plastrons can be a function of surface roughness, including its geometry [57, 58]. The air-water interface is potentially considered a barrier in bacterial adhesion by limiting the contact area with the surface [59]. Therefore, we wanted to find out if increasing hydrophobicity of the polymer composite surface and NO release can reduce bacterial and platelet adhesion. Due to its inert chemical nature, it is challenging to functionalize PTFE [55]. Therefore, the use of mussel-inspired adhesive chemistry can potentially address the challenge of immobilizing PTFE on a surface without harsh chemical treatments. The underlying cause of efficient adhesive properties bio-inspired from mussels is the secretion of mussel foot proteins (Mfp-5) in the mussel’s adhesive thread [60]. The foot proteins containing 3,4-dihydroxy-L-phenylalanine (DOPA) and lysine amino acids impart excellent adhesive properties [61]. Earlier studies found that commercially available dopamine-containing catechol and lysine amino groups can be key factors in achieving extraordinary adhesion properties [62]. In the past, PDA coatings have been used to attach inorganic and organic substrates (including PTFE) [60, 63].

In the current study, a facile biomaterial design has been explored to address device-related infection and thrombosis problems through NO release supplemented by a highly hydrophobic surface to limit the adhesion of bacteria or platelets. The coatings have been characterized through WCA measurements and surface Fourier transform infrared spectroscopy (FTIR). The stability of the coating was also evaluated via FTIR, WCA, and surface tension. The surface topography was evaluated through SEM. Biological characterizations were conducted to evaluate its capabilities to resist bacteria and platelet adhesion without compromising the safety of mammalian cells in vitro.

2. Materials and Methods

2.1. Materials

N-Acetyl-D-penicillamine (NAP) (≥ 99.0 % purity), tetrahydrofuran (THF, ≥ 98.0 %), sodium nitrite (NaNO_2, ≥ 99.0 %), ethylenediaminetetraacetic acid (EDTA, ≥ 99 %) and hydrochloric acid were purchased from Sigma Aldrich (St. Louis, MO 63103). CarboSil-2080A™ (hereon as CarboSil) was purchased from DSM (Berkeley, CA). Teflon™ PTFE emulsion, DISP 30 LX, was purchased from Fuel Cell Earth (Woburn, MA); Phosphatebuffered saline (PBS), pH 7.4, containing 138 mM NaCl, 2.7 mM KCl, and 0.01 M sodium phosphate. Dulbecco’s modified Eagle’s medium (DMEM) and trypsin-EDTA were purchased from Corning (Manassas, VA 20109). The Cell Counting Kit-8 (CCK-8) was purchased from Sigma Aldrich (St. Louis, MO 63103). Penicillin-Streptomycin (Pen-Strep) and fetal bovine serum (FBS) was obtained from Gibco-Life Technologies (Grand Island, NY 14072). The bacterial strains S. aureus (ATCC 6538) and E. coli (ATCC 25922) were purchased from American Type Culture Collection (ATCC). LB broth was obtained from Fisher Bioreagents (Fair Lawn, NJ). LB Agar was purchased from Difco Laboratories Inc (Detroit, MI). The lactic dehydrogenase (LDH) kit was purchased from Roche Life Sciences (Indianapolis, IN). Hexamethyldisilazane (HMDS, ≥ 99 % purity) was obtained from Sigma Aldrich (St. Louis, MO 63103)

2.2. SNAP synthesis

SNAP was synthesized from NAP by modification of a previously established protocol.[64] The equimolar ratio of NaNO₂ and NAP was added to a mixture of de-ionized (DI) water and methanol containing 1 M HCl and 1 M H₂SO₄. The mixture was stirred in a reaction vessel in the absence of light (to avoid activation of NO release with light as a stimulant) for 15 min and then cooled in an ice bath for 4 h to obtain precipitated SNAP crystals. The SNAP crystals appear green in color. After precipitation of the SNAP crystals, the precipitate was collected, and vacuum dried overnight in the dark to remove any trace solvent present with the crystals. The purity determined through nitric oxide analyzers was found to be ≥ 99%.

2.3. Polymer composite fabrication

The CarboSil and CarboSil-SNAP composites were prepared via solvent evaporation method. 70 mg mL⁻¹ CarboSil was dissolved in THF. For SNAP-CarboSil, SNAP was added to the solution after the complete dissolution of CarboSil in THF. The solutions were cast into Teflon molds (diameter = 2.45 cm) and dried overnight in the dark to prevent light-induced release of NO from SNAP. The composites formed were further vacuum dried using a desiccator to remove residual traces of THF in the composites. A PDA solution was prepared by adding dopamine hydrochloride (5 mg ml⁻¹) to Tris buffer (pH 8.5). The composites prepared were soaked in the PDA solution for 24 h under shaking conditions at RT. After 24 h, the composites were rinsed rigorously with DI water to get rid of the excess and unbound PDA, air-dried, and kept in a desiccator overnight. An aqueous fluoropolymer colloidal dispersion (Chemours Teflon™ DISP 30) was prepared at 30, 40, 50, and 60 wt % with DI water. The PDA-coated composites were dip-coated for 10s in the aqueous fluoropolymer solution, air-dried, washed with copious amounts of water to eliminate any surfactants remaining on the PTFE particles and then put onto a hotplate at 120°C for 2 min to remove any interfacial water. After which, it was put in a desiccator overnight in the dark at RT.

2.4. FTIR

Surface modification of the composites by chemical treatment was confirmed with FTIR measurements using Thermo Nicolet Avatar 360 in 650–4000 cm⁻¹ range with 128 scans. Composites were vacuum dried overnight and directly placed on to the attenuated total reflection (ATR) mount on the FTIR instrument. With the use of ATR, FTIR measurement gives the chemical environment of the surface rather than the interior of the sample, which helps the proper identification of surface analysis of the composite samples. Not less than three FTIR spectra were taken for each sample to confirm the equal distribution of modification over the composite surface.

2.5. SEM

The surface morphology and roughness of the polymer composites were examined by SEM (FEI Inspect F FEG-SEM). Dried composite samples were mounted on a metal stub with double-sided carbon tape and sputter-coated with 10 nm gold–palladium using a Leica EM ACE200 sputter coater. An accelerating voltage 5 kV was used for the experiment.

Microscopic images of S. aureus on the polymer composites were obtained after exposing the composites to S. aureus for 24 h. Bacteria were collected in the mid-exponential growth phase and were diluted with PBS buffer to attain 10⁸ CFU mL⁻¹. Subsequently, the polymer composites were incubated at 37 °C in 2 mL bacteria solution in a 24-well plate under shaking condition (150 rpm). After 24 h, the polymer composites were washed with PBS buffer to get rid of any planktonic or loosely attached bacteria. The composites were fixed with 3 % glutaraldehyde in 0.1 M PBS solution for 16 h. The samples were dehydrated with graded ethanol for 20 min (50, 60, 70, 80,90, and 100 vol. %). Finally, the samples were soaked and washed in HMDS and left to air-dry overnight in the fume hood away from light. Each sample was mounted and sputter-coated with gold-palladium, and random spots were chosen for imaging. The experiment was repeated three times using different passages of bacteria. Representative images have been shown.

2.6. Static water contact angle

The surface contact angle of the control composites and the coated composites was analyzed by placing 5 μL of water from a computer-controlled liquid dispensing system onto the sample using a Krüss DSA100 Drop Shape Analyzer (sessile drop method with DI water). The droplet of water was placed on various points of the composites, and the average of left and right contact angles was measured via the Krüss software.

2.7. Surface tension

The coating stability was characterized by measuring the surface tension (mN m⁻¹) of PBS buffer used for soaking the PTFE coated polymer samples. Surface tension was measured by the pendant drop method using Ossila Contact Angle Goniometer (Ossila, Sheffield). Initially, the surface tension of PTFE aqueous dispersion was measured at varying weight percentages to demonstrate a change in surface tension. CarboSil-PTFE and CarboSil SNAP-PTFE polymer composites were soaked in 2 mL of PBS buffer for 24 h at 37 °C. The composites were removed after 24 h, and the surface tension was measured with PBS buffer, without any polymer composites, as control.

2.8. NO release kinetics

The release profile of NO from the SNAP incorporated composites was recorded in real-time using Sievers chemiluminescence NO analyzers (NOA 280i, GE Analytical, Boulder, CO, USA). The reaction vessel was kept in the dark to prevent the catalysis of NO by light. The samples were soaked in PBS (pH 7.4 with 100 μM of EDTA), and NO release was measured at 37°C to simulate physiological conditions. EDTA acts as a chelating agent to prevent the catalysis of NO production from SNAP in the presence of any metal ions in the PBS. A nitrogen bubbler was placed in the solution containing the sample at a flow rate of 200 mL min⁻¹ to carry any NO being emitted to the NOA.

2.8. SNAP diffusion

The diffusion of NO donor, SNAP, from the PTFE coated, and uncoated polymer composites were measured by recording the absorbance of SNAP in PBS (with EDTA) solution at various time intervals. The SNAP containing composites were soaked in 2 mL of 0.01M PBS (pH 7.4 with 100 μM EDTA to prevent catalysis of NO release by metal ions with EDTA) at 37 °C. Absorbance measurements were taken at an optical density of 340 nm to match the UV-Vis absorbance maxima spectra for SNAP using a UV-Vis spectrophotometer (Thermo-Scientific Genesys 10S UV-Vis). Wavelength 340 nm is the characteristic absorbance maxima of the S-NO group of SNAP. A calibration curve of SNAP in PBS with EDTA was used to interpolate the absorbance quantifications recorded and determine the concentrations of SNAP diffused into PBS buffer.

2.9. Bacterial adhesion test

The ability of the coated composites to resist adhesion, inhibit growth and promote killing of the adhered bacteria on the polymer surface was tested following guidelines based on American Society for Testing and Materials (ASTM) E2180 protocol with the commonly found nosocomial pathogen, Gram-positive S. aureus (ATCC 6538) and Gram-negative E. coli (ATCC 25922). A single colony of bacteria was isolated from a previously cultured LB agar plate and incubated in LB broth (37 °C, 150 rpm, 12–14 h). The optical density of the culture was measured at a wavelength of 600 nm using a UV–vis spectrophotometer (Thermo Scientific Genesys 10S UV–vis) to ensure a mid-log phase growth and then adjusted to ~10⁸ colony forming units (CFU) mL⁻¹. The polymer composites: CarboSil, CarboSil-SNAP, CarboSil-PTFE, and CarboSil-SNAP-PTFE were then incubated in the bacterial suspension in 24-well plates at 37 °C for 24 h (150 rpm). After incubation, the polymer composites were removed from the bacterial suspension and rinsed with sterile PBS to remove unbound bacteria. The polymer composites were then sonicated using an Omni-TH homogenizer (Omni, Kennesaw, GA) at 25 000 rpm for 60 s to collect adhered bacteria in sterile PBS. To ensure proper homogeneity of the collected bacteria, the samples were vortexed for 60 s each. The collected solutions containing the resultant bacteria were serially diluted and plated on LB agar plates and incubated at 37 °C. After 24 h, the total CFUs on the plate were counted. Bacterial adhesion was calculated and analyzed as CFU cm^−2, according to the following equation.

$$\mathit{\text{Reduction}}\mathit{\text{efficiency}}\left(\%\right)=\frac{C_{\mathit{\text{control}}}-C_{\mathit{\text{sample}}}}{C_{\mathit{\text{control}}}}\times 100$$ (eqn. 1)

Where, C = CFU cm⁻²

2.10. Platelet adhesion test

Samples were exposed to blood plasma with a known quantity of platelets to assess antiplatelet efficacy. All protocols pertaining to the use of whole blood and platelets were approved by the Institutional Animal Care and Use Committee. Freshly drawn porcine blood (Lampire Biological) with 3.9% sodium citrate at a ratio of 9:1 (blood: citrate) was used. The anticoagulated blood was centrifuged at 300 rcf for 12min using a Beckman Coulter Allegra X-30R Centrifuge. The platelet-rich plasma (PRP) portion was carefully drawn with a pipet, so the buffy coat is not disturbed. The remaining samples were then spun again at 4000 rcf for 20 min to collect platelet-poor plasma (PPP). Total platelet counts of both the PRP and PPP fractions were determined using a hemocytometer (Fisher). The PRP and PPP were combined in a ratio to give a final platelet concentration 2×10⁸ platelets mL⁻¹. Calcium chloride (CaCl₂) was added to the final platelet solution to a final concentration of 2.5 mM to reverse the anticoagulant (Na-citrate) [65], and after that, samples were placed in blood tubes and exposed to approximatively 4 mL of the calcified PRP. The tubes were then incubated at 37°C for 90min with mild rocking (25 rpm) on a VWR blood tube rocker. Following the incubation, the tubes were infinitely diluted with phosphate buffer solution. The degree of platelet adhesion was determined using the lactate dehydrogenase (LDH) released when the adherent platelets were lysed with a Triton-PBS buffer (2% v/v Triton-X-100 in PBS) using a Roche Cytotoxicity Detection Kit (LDH). A calibration curve was constructed using known dilutions of the final PRP solution, and the platelet adhesion of the various sponge types was interpolated from the calibration curve.

2.11. Cytotoxicity assay

The cytotoxicity of the extracts obtained from the polymer composites was investigated on NIH 3T3 mouse fibroblast cell lines (ATCC 1658) per ISO 10993 standards. The cytotoxicity was determined using Cell Counting Kit-8 (CCK-8), which utilizes a highly water-soluble tetrazolium salt WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)- 5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt]. WST-8 is reduced by dehydrogenases in cells to produce water-soluble formazan dye (an orange-colored product). The amount of formazan dye generated by dehydrogenases in cells is directly proportional to the cells, and its absorbance can be measured at 460 nm.

Mouse fibroblast cells were cultured in a humidified atmosphere with 5% CO₂ at 37 °C in complete DMEM medium containing 10% fetal bovine serum (FBS) and 1% penicillinstreptomycin. The cells were removed from the T-flask using 0.25 % trypsin and 5 mM EDTA, once 80–90% confluency was attained. The cells were seeded into the wells of a 96-well plate at a concentration of 2 × 10⁴ cells mL⁻¹ and incubated for 24 h in a humidified atmosphere with 5% CO₂ at 37 °C. The extracts from each sample (CarboSil, CarboSil SNAP, CarboSil-PTFE, and CarboSil SNAP-PTFE) were obtained by soaking the samples in complete DMEM for 24 h at 37 °C amber vials away from light. After 24 h, the media from the wells containing fibroblast cells were eluted and replaced with the complete DMEM with the extracts and incubated for another 24 h. To each of the wells containing fibroblast cells, 10 μL of the CCK-8 solution was added and incubated for 3 h. No extracts were present in the control wells. Absorbance values were measured at 450 nm.

2.12. Statistical analysis

All data are represented as mean ± standard deviation (SD) with n=3 unless stated otherwise. Each result is an average of at least three parallel experiments. The statistical significance was assessed by student’s t-test analysis - * (p < 0.05), ** (p < 0.01), *** (p < 0.001). Statistical analysis and graphs were prepared using GraphPad Prism version 8.0.

3. Results and discussions

3.1. Fabrication and physical characterization

To avoid harsh chemical treatment and circumvent the chemically inert nature of PTFE, the fabrication of the biomaterial is bio-inspired from the adhesive nature of a) high catechol content (due to the presence of 3,4-dihydroxy-L-phenylalanine (DOPA)) and amines (lysine) in mussel adhesive proteins. A facile dip-coating method was used with the self-polymerization ability of dopamine hydrochloride in Tris buffer (pH 8.5) to form PDA. The presence of both DOPA and amine in PDA imparts binding ability with a wide range of adhesive capabilities [60]. Several studies have shown that PDA can effectively bind to even inert polymer surfaces like PTFE [60, 63]. CarboSil is a biomedical grade polymer (FDA approved), a thermoplastic silicone- polycarbonate-urethane containing polycarbonate-urethane with a mix of poly (dimethylsiloxane) and polycarbonate as well as a hard segment of methylene diphenyl isocyanate (MDI), polymeric platform and characterized for biological responses. Based on previously conducted X-ray diffraction studies, 10 wt% SNAP (NO donor molecule) was used to make the composites that result in the crystallization of SNAP within the CarboSil and as a result, imparts NO-releasing capabilities to the polymer composite [43, 44].

An adherent thin film of PDA can be achieved by simple immersion [60]. The dopamine undergoes a series of reactions starting with the formation of dopamine quinone through oxidation, eventually forming isomers of PDA, where the hydroxy functional groups are attached to the CarboSil surface. As there is no concrete consensus on the reaction pathway for the formation of PDA, proposed pathways of covalent polymerization and physical aggregation by self-assembly are a possibility [62]. Studies conducted by Leng et al. on PDA films suggest that (dihydroxy indole)₂/pyrrole carboxylic acid trimer complex are the primary building blocks of PDA [66]. Earlier, d’Ischia and co-workers had demonstrated that the main building blocks of PDA are – uncyclized catecholamines/quinones, cyclized units of dihydroxy indole and pyrrole carboxylic acid moieties [67]. With the PDA coating acting as an anchoring layer, PTFE coating was introduced through dip-coating. The PTFE solution used is an aqueous dispersion (Chemours PTFE DISP 30) containing PTFE particles (~ average particle size 220 nm) at a concentration of 60 wt %. The objective of the PTFE layer was to introduce a thin coating of PTFE to impart a robust highly hydrophobic surface. Therefore, the PTFE colloidal dispersion was diluted to 30, 40, 50 wt. % in addition to the as-is concentration of 60%. The 40 wt. % colloidal dispersion was chosen for further characterization as higher percentages showed signs of flaking while 30 wt% did not provide the desired hydrophobicity (data not shown). The choice of 40 wt. % PTFE particle dispersion was also, in part, inspired by the study conducted by Beckford et al. on PTFE thin films enabled by PDA layer on stainless steel where they demonstrate excellent wear resistance of the PDA/PTFE thin films [68]. The final samples fabricated were CarboSil, CarboSil-SNAP, CarboSil PTFE, and CarboSil-SNAP PTFE. The thickness of the coating was measured through SEM image analysis using ImageJ software. The PTFE coating on CarboSil-PTFE had a thickness of 5 ± 2 μm and on CarboSil SNAP-PTFE, 6 ± 4 μm. A schematic depiction outlining the material system and the biointerface is shown in scheme 1.

Scheme 1-.

Schematic diagram depicting the biomaterial system and the biointerface.

The initial analysis of the surface, confirming the PTFE coating of the polymer composites was conducted using FTIR. As demonstrated in Figure 1a, the characteristic peaks for symmetrical and asymmetrical stretching of CF₂ are present in PTFE coating (1145 and 1203 cm⁻¹, respectively). CF₂ wagging at 769 cm⁻¹, CF₃ vibrations at 1016 cm⁻¹, and the amorphous nature of the PTFE can be determined through its peak at 791 cm⁻¹. This infers that the PTFE has been successfully coated onto the polymer composites. Figure 1b depicts the confirmation of the presence of the PTFE coating after CarboSil-PTFE, and CarboSil SNAP-PTFE is soaked in 0.01 M PBS buffer for 24 h under shaking conditions at 37 °C. Although FTIR analysis does not provide information about subtle changes of a coating loss, it can confirm that the PTFE coating is not entirely depleted or peeled off under robust physiological conditions.

Figure 1-.

a) FTIR spectra of CarboSil SNAP-PTFE (blue), CarboSil-PTFE (red), CarboSil SNAP (green), CarboSil (black) after preparation of samples. b) detection of PTFE coating on CarboSil-PTFE and CarboSil SNAP-PTFE after soaking the polymers composites in PBS for 5 days at 37 °C under shaking condition (120 rpm)

The surface morphology of the polymer composites was evaluated through SEM. Figure 2 shows the surfaces of CarboSil, CarboSil SNAP, CarboSil-PTFE, and CarboSil SNAP-PTFE. The success of the PTFE coating on the CarboSil and CarboSil SNAP polymer composites could also be verified by observing the surface morphology. SEM analysis revealed a change in the smoothness of the CarboSil SNAP surface compared to the pristine CarboSil polymer composite surface. The surface roughness observed may be due to the interaction of SNAP and the polymer solution during the solvent evaporation stage. Previous work by Wo et al. have established the formation of orthorhombic SNAP crystals within CarboSil when the concentration of SNAP exceeds 3–4 wt % [43]. Interestingly, the PTFE coated polymer composites had a homogenous cobblestone-like structured appearance. The structured cobblestone-like arrangement of the PTFE particles (ca. 200 nm) can be the cause of increased hydrophobicity of CarboSil SNAP-PTFE and CarboSil-PTFE polymer composites. It is highly likely that air pockets are entrapped within the cobblestone-like structured surface, and the resulting interaction causes water repellence, as evident from the WCA measurements discussed below.

Figure 2-.

Representative SEM micrographs depicting the surface morphology of a) CarboSil b) CarboSil SNAP c) CarboSil-PTFE and d) CarboSil SNAP-PTFE (Scale bar: 10 μm)

Static WCA measurements or surface wettability provide an idea of the surface property of the PTFE coated polymer composites. Generally, water contact angles above 90° are considered hydrophobic. As expected, the WCA measurements also confirmed the PTFE coating on the CarboSil substrate (Figure 3a). Pristine CarboSil surface had WCA 102.2° ± 1.5° while CarboSil SNAP surface measured 101.0° ± 4.1°. PDA coatings on substrates have been shown to impart a hydrophilic character to the surfaces [60, 69]. The hydrophilicity is mainly attributed to hydroxyl, carboxylic acid, and amine functional groups in PDA [70]. A hydrophilic shift in the surface wettability was observed after the PDA coating for both CarboSil (CarboSil-PDA, 58.0° ± 4.3°) and CarboSil SNAP (CarboSil SNAP-PDA, 53.1° ± 7.0°). Compared to CarboSil control, CarboSil-PTFE had a contact angle of 121° ± 5° (p < 0.05) and for CarboSil SNAP-PTFE, 120.6° ± 2.2° (p < 0.05). The difference in contact angles between CarboSil-PTFE and CarboSil SNAP-PTFE was not statistically significant.

Figure 3-.

(a) WCA measurements of CarboSil, CarboSil-SNAP, CarboSil-PTFE, and CarboSil SNAP-PTFE. (b) WCA measurement of CarboSil-PTFE and CarboSil SNAP-PTFE do determine loss of hydrophobicity after soaking in PBS buffer under shaking condition (rpm 120) at 37 °C for 24 h. (c) change in surface tension (mN m⁻¹) with increasing weight percentages of PTFE nanoparticle dispersion. (d) the surface tension of the PBS buffer where CarboSil-PTFE and CarboSil SNAP-PTFE were immersed under shaking condition (rpm 120) at 37 °C for 24 h. (e) schematic illustration depicting cobblestone-like ordered structure provided by PTFE nanoparticles and its wetting property. Data represent the mean ± standard deviation (n = 3). Statistical significance is indicated by an asterisk. ** (p <0.01), *** (p <0.001) ns = not significant

The polymer composites after PTFE coating (CarboSil-PTFE and CarboSil SNAP-PTFE) were soaked in PBS and kept at 37 °C for 24 h under shaking conditions (120 rpm). WCA was measured after air-drying the polymer composites to determine significant changes in the surface wettability when immersed in 0.01 M PBS buffer under shaking conditions and physiological temperature. Although FTIR results revealed the presence of the PTFE coating after 24 h, it is limited in determining coating losses. However, significant changes in contact angle can indirectly indicate coating losses. Figure 3b shows that after 24 h, the polymer composites did not have a significant change in contact angle. Due to a hydrophilic PDA layer underneath the PTFE coating, any coating loss would expose the PDA layer resulting in significant changes to contact angles. In this case, the coatings have remained intact in their hydrophobicity.

Conversely, the surface tension of the 0.01 M PBS buffer used for soaking the PTFE coated polymer composites (CarboSil-PTFE and CarboSil SNAP-PTFE) were measured and compared after 24 h. Figure 3c reveals a marked decrease in surface tension with increasing weight percentage (in the lower ranges) of PTFE dispersion in PBS. After 24 h incubation at 37 °C under shaking conditions (rpm 120), no significant observable difference is recorded for either CarboSil-PTFE or CarboSil SNAP-PTFE (Figure 3d).

The transition in the surface wettability observed after PTFE coating indicates a highly hydrophobic state of the surface. Changes in hydrophobicity on the surface can dictate interfacial interactions of the polymer composite and surrounding aqueous or physiological environment. Interfacial nanobubbles are likely to form at the nano- or sub-micron levels upon immersion [71, 72]. Drawing inferences from the SEM images of PTFE-coated polymer composites, the hydrophobicity can be subject to surface roughness as well as the ordered structure (Figure 3e). Work done by Cassie and Baxter has shown that air or gas can be entrapped within the cavities of the rough surface resulting in an air-liquid interface [73]. Peng et al. has demonstrated increasing PTFE nanoparticle concentration to 75 wt. % could provide surface roughness at the nano-scale to achieve superhydrophobicity [74].

3.2. NO release kinetics and SNAP diffusion

As an active agent with a versatile physiological effect, NO release profile is an important parameter in the design and fabrication of PTFE coated CarboSil SNAP. NO release was measured in real-time via a chemiluminescence based analyzer. A representative bar graph depicting the NO release profile for 120 h has been shown in Figure 4a. The samples were tested in 0.01 M PBS with EDTA at 37 °C (n=3). The nitrogen bubbler and sweep gas had a combined flow rate of 200 mL min⁻¹. The initial NO flux of CarboSil- SNAP samples (2.3 ± 0.6 × 10⁻¹⁰ mol cm⁻² min⁻¹) were significantly higher than CarboSil SNAP-PTFE (1.0 ± 0.3 × 10⁻¹⁰ mol cm⁻² min⁻¹). The lower release can be attributed to the presence of PDA and PTFE layers in CarboSil SNAP-PTFE samples, which prevented the excessive release of NO by preventing interaction with water molecules. Over 5 days of measurement, CarboSil SNAP-PTFE had a steady NO surface flux rate compared to CarboSil SNAP. Due to its higher release flux on day 1, the NO donor, SNAP, reservoir got exhausted faster than the samples with PTFE layers. On day 5, a significant difference in NO flux was observed between both samples with CarboSil-SNAP-PTFE having significantly higher flux (p < 0.05). The NO surface flux of CarboSil SNAP and CarboSil SNAP PTFE were 0.2 ± 0.1 × 10⁻¹⁰ mol cm⁻² min⁻¹ and 0.5 ± 0.1 × 10⁻¹⁰ mol cm⁻² min⁻¹, respectively.

Figure 4-.

a) NO release of CarboSil SNAP and CarboSil SNAP-PTFE on specific days measured in PBS with EDTA at 37 °C (n=3). b) Cumulative SNAP diffused from CarboSil SNAP-PTFE with CarboSil SNAP as control. Measured in PBS with EDTA at 37 °C (n=3). * (p < 0.05), *** (p < 0.001)

Diffusion of the donor molecule can significantly shorten longevity or release lifetime of NO from the polymer composites. Therefore, SNAP diffusion from PTFE coated and uncoated surfaces were quantified to ascertain differences in cumulative diffusion with progressing time, affecting longevity or release profiles. In a span of 24 h, SNAP leaching per cm² of the polymer composites observed for CarboSil SNAP was higher than CarboSil SNAP-PTFE (Figure 4b). Further, 5.1 ± 0.4% of total SNAP loaded diffused out of CarboSil SNAP while 2.3 ± 0.3 % diffused out of CarboSil SNAP-PTFE. This was expected as PTFE forms a physical barrier that limits the mobility of SNAP molecules outwards from the polymer composites. The PTFE coating can also reduce the interaction of the SNAP molecules with water molecules. CarboSil is a hydrophobic polymer [50]. The hydrophobic properties of CarboSil and the PTFE coating, therefore, can play a limiting role in water availability and diffusion of SNAP into the PBS buffer - more so for CarboSil SNAP-PTFE due to its increased hydrophobicity compared to CarboSil.

3.3. Antibacterial activity

The in vitro adhesion of viable bacteria onto the polymer composites was examined by challenging the surfaces with clinically relevant Gram-negative E. coli and Gram-positive S. aureus. The polymer composites were exposed to ~ 10⁸ CFU mL⁻¹, which is considerably more rigorous than the practically relevant amount in the case of indwelling medical device infections. For context, central venous catheters, or other bloodstream related infections, bacteriuria can be around (15 – 10³ CFU/ catheter segment), and for urinary tract infections, it could be 10³- 10⁵ CFU/mL[75, 76]. Figure 5 reveals that the PTFE coated NO-releasing composites could reduce the viability of adhered S. aureus and E. coli by 2.17 and 2.06 log₁₀ reduction, respectively. CarboSil SNAP-PTFE had a reduction efficiency of 99.3 ± 0.5 % and 99.1 ± 0.4 % for S. aureus and E. coli, respectively, compared to control, which did not have PTFE coating and NO release capabilities (Table 1). The bio-active agent in the polymer composite, NO, is known to elicit antibacterial action through the production of free radicals (reactive nitrogen species) and in aerobic conditions, can interact with superoxides (O²⁻), and hydrogen peroxides (H₂O₂) to form reactive antimicrobial species [77]. It can form reactive intermediate derivatives of NO, such as peroxynitrite (OONO⁻), nitrogen dioxide (NO₂), dinitrogen trioxide (N₂O₃), and dinitrogen tetroxide (N₂O₄) which can attack the DNA, cause oxidative damage to intracellular targets, proteins, etc. [22]. Nitrosative stress emanating as a product of NO reactivity can also cause cell surface disruption by lipid peroxidation [22]. The current design of the biomaterial allows a passive and active approach towards making the surface antibacterial. The initial bacterial attachment is due to weak interactions like van der Waals [78]. Interactions on Teflon coated substrates with dispersions are predicted to be van der Waals forces and low surface energies, which reduces the initial attachment of bacteria to the surfaces [79, 80]. On highly hydrophobic surfaces such as that of PTFE coating, air-water interfaces, or the presence of interfacial nanobubbles can considerably lower the interaction of the bacteria with the surface [56, 59]. Therefore, in principle, the PTFE coating passively reduces the attachment of bacteria to the surface of the polymer composites, whereas NO release actively kills bacterial cells that have managed to attach/adhere to the surface.

Figure 5-.

Graph representing the reduction of viable bacteria adhered to CarboSil, CarboSil SNAP, CarboSil-PTFE and CarboSil SNAP-PTFE (in log CFU cm⁻²). Data represent mean ± standard deviation (n=3). Statistical significance is represented by asterisk, * (p < 0.05), ** (p <0.01), *** (p <0.001)

Table 1-.

Reduction efficiency of the polymer composites on the viable adhered bacteria. Data represent represent mean ± standard deviation (n=3).

Bacterial strain	Reduction Efficiency (%) compared to CarboSil
	CarboSil-PTFE	CarboSil SNAP	CarboSil SNAP-PTFE
S. aureus	67.8 ± 1.3	97.6 ± 0.6	99.3 ± 0.5
E. coli	74.0 ± 3.8	95.6 ± 1.5	99.1 ± 0.4

SEM analysis was carried out to examine morphological changes to S. aureus adhered to and biofilm formation on the surface of the polymer composites (Figure 6). Notable differences can be observed between CarboSil and other polymer composite surfaces. Bacterial adhesion is observed throughout the CarboSil surface along with biofilm formation in some areas (Figure 6a). The bacterial cells are well-defined, spherical, and have an intact smooth surface (characteristic of a coccoid) where no membrane disruption or distortion is observed. Terminal stages of bacterial cell division with roughly two hemispherical bacteria attached to a single contact point or a progressive separation during cell division are prevalently observed in some regions of the surface. S. aureus is attached to the surface surrounded by the extracellular polymeric substrates (EPS) of the biofilm structure (indicated by red arrows). The biofilm architecture in the image has a flattened structure instead of a three-dimensional gellike structure due to drying out steps during the preparation method [81]. During the dehydration process, the EPS with its surrounding fibrillar structure has possibly condensed [82]. The presence of biofilm structure elevates the life-threatening risks that are associated with infections of indwelling medical devices. Contrary to the CarboSil control, NO-releasing or/and PTFE coated polymer composites do not show signs of biofilm formation on the surface during 24 h. CarboSil SNAP surface revealed changes in the morphology of the cells adhered to the surface (Figure 6b). Primarily, most cells had a shrunken or indented appearance. Cell debris or blebs were also observed on the surface indicating dead cells or leakage of intracellular components post bacterial membrane rupture (indicated with red arrows). NO is also known to regulate as well as disperse bacterial biofilms [24, 26]. It is interesting to note, comparatively lesser number of bacteria adhered to the surface of the CarboSil SNAP surface. This is possibly due to the effect of NO on planktonic bacteria. In our previous studies, we have shown that NO-releasing polymer can decrease planktonic bacteria [28]. CarboSil-PTFE had lower adhered bacteria on the surface, and the cells appear to be morphologically intact (Figure 6c). The manufacturer of the PTFE particles uses a nonionic surfactant as a stabilizer. Although nonionic surfactants are known to elicit less toxicity compared to cationic or anionic surfactants, any effect of the surfactants on the bacterial cells would be obvious by the presence of dead cells on the surface of CarboSil-PTFE. The lack of dead cells/ cell debris suggests no interference of any residual surfactant on the PTFE coated surfaces. Therefore, reduced number of cells are solely due to properties of PTFE immobilization. The bacterial cells also seemed to aggregate at localized regions on the surface. This observation also holds for CarboSil SNAP-PTFE (Figure 6d). Furthermore, there are no signs of EPS secreted from the cells or biofilm formation, which indicates, PTFE coating can resist the formation of biofilms in 24 h compared to CarboSil surfaces. For the CarboSil SNAP-PTFE surface, in addition to less adhered cells, morphologically damaged bacterial cells were also observed. Figure 6e provides a closer look at the S. aureus cells adhered to CarboSil SNAP polymer composites with morphological changes such as cell surface distortion, collapsed, and disrupted membranes [22]. It is established that NO can affect the structural integrity of the cell membrane via lipid peroxidation. Nitrosative species produced by NO reactivity can cause DNA deamination and S-nitrosation of thiols on the cell surface and intracellular proteins, ultimately causing cell death [22]. Figure 6f shows morphological changes to S. aureus adhered to CarboSil SNAP-PTFE. Apart from the typical S. aureus morphology, two other characteristic features have been observed. First, cells in the intermediate stages of cell division can be seen. These are identified as smooth flattened septum as the undeveloped or un-expanded hemisphere of the daughter cell. The ring-like appearance are scars corresponding to the cell division site [83]. Eventually, the flattened septum will enlarge into an expanded hemisphere with the progression of the cell cycle [83]. The other characteristic morphology is cells with distorted or disrupted surface appearances, which are likely dead or dying S. aureus.

Figure 6-.

Representative SEM images of S. aureus adhesion on the polymer composites after 24 h incubation at 37 °C. (a) S. aureus adhering to the CarboSil surface (Scale bar = 10 μm). Biofilm formation is observed on the surface (indicated by red arrows) (b) CarboSil SNAP surface (Scale bar = 10 μm), red arrows indicate dead bacterial cell or cell debris. (c) S. aureus adhered to CarboSil-PTFE polymer composite (Scale bar = 10 μm). Aggregation of live S. aureus is indicated by a red arrow. (d) bacterial adhesion on the CarboSil SNAP-PTFE surface (Scale bar = 10 μm), red arrows indicate dead bacteria. Morphological changes in S. aureus adhered to (e) CarboSil SNAP (scale bar = 1 μm) and (f) CarboSil SNAP-PTFE (scale bar = 1 μm). Yellow arrows indicate PTFE particles.

3.4. Reduction in platelet adhesion

To ensure that the PTFE coating can repel the cellular contents of blood, control CarboSil, and test samples were exposed to platelet-rich porcine plasma for 90 minutes in physiological conditions. After incubation, adhered platelet cells were lysed and quantified using an LDH Roche assay. The assay data revealed significant antiplatelet efficacy in both PTFE and NO-releasing samples (Figure 7a). CarboSil-PTFE surface was able to reduce platelet adhesion by 46.3 ± 5.7 % (p < .01) compared to CarboSil control. The reduction of platelet adhesion in PTFE samples is due to their passive ability to prevent platelet adsorption from the increased hydrophobicity. While this passive approach reduced platelet adhesion, and even higher platelet reduction of 74.5 ± 5.2 % was observed in the CarboSil SNAP polymer composite. The CarboSil SNAP composites had 52.6 ± 9.7 % less adhered platelets than CarboSil-PTFE due to NO’s bio-active ability to prevent platelet activation. NO’s antiplatelet mechanisms work by stimulating cGMP levels to prevent activation of the fibrinogen-binding IIb/IIIa glycoprotein at the surface of platelet cells [84]. By combining passive antifouling modification and a bio-active agent, the CarboSil SNAP-PTFE surface exhibited superior hemocompatibility by significantly reducing platelet adhesion by 83.3 ± 4.5% when compared to untreated controls, which included a 34.3 ± 17.7 % decrease compared to CarboSil SNAP polymer composites. By exhibiting superior antifouling efficacy against platelet cells, it can be concluded that the PTFE coating process can be used on a polymer surface along with NO-releasing capabilities to increase the hemocompatibility of NO-releasing materials.

Figure 7-.

(a) Graph representing platelet adhesion (cm⁻²) on polymer composites after 2 h exposure to porcine platelet-rich plasma (n = 8) (b) Graph representing cell viability of 3T3 mouse fibroblast cells after exposure to leachates from polymer composites (n = 5). Control did not have any leachate. Data represent the mean ± standard deviation. Statistical significance is represented by an asterisk. ** (p < 0.01), *** (p < 0.001)

3.5. In vitro cytotoxicity evaluation

For the indwelling medical device coating, the biomaterial design functionality should not elicit toxic effects or compromise the host cells in the vicinity. Therefore, the biocompatibility of the coating is a crucial factor, while still retaining its antibacterial or anti-platelet properties. Cytotoxicity, in this study, is measured via the viability of cells in the presence of extracts from all the polymer composites. The number of viable cells was measured via WST-8 dye-based CCK-8 assay. Figure 7 (b) shows the viability of cells expressed as absorbance at 450 nm relative to control (which did not have any extracts). The extracts from the polymer composites do not show significant loss of viable cells. No significant difference was observed between PTFE coated and non-coated polymer composites. It is not surprising as PTFE is known to be a chemically inert polymer and has been commercially used as medical device coatings [55].

4. Conclusion

A superhydrophobic [85, 86] or hydrophilic [50] surface modifications or coatings for NO-releasing polymers are known to resist bacterial adhesion or thromboresistivity. However, there are limited studies on surfaces that consist of a highly hydrophobic interface, especially for NO-releasing polymers. The current work provides some insight into infection and platelet resistance of such a highly hydrophobic interface achieved through PTFE particle coating in addition to nitric oxide release. The fabrication process involved a facile dip-coating of polydopamine anchor layer to immobilize colloidal dispersion of PTFE particles onto NO-releasing medical-grade polymer, CarboSil. The combination of bio-active NO donor, SNAP incorporated into the CarboSil polymer matrix, and PTFE particle coating has been shown to prevent adhesion of both bacteria and platelets synergistically. The release of NO from the composites mimicked endogenous NO release from the endothelial cell linings of blood vessels and had a surface flux of 1.0 ± 0.2 × 10⁻¹⁰ mol cm⁻² min⁻¹ on the first day. After five days of measurement, NO surface flux was calculated to be 0.5 ± 0.1 × 10⁻¹⁰ mol cm⁻² min⁻¹ which is well within the physiological range. CarboSil SNAP-PTFE was also able to successfully reduce the viability of adhered bacteria by 99.3 ± 0.5 % and 99.1 ± 0.4 % for S. aureus and E. coli, respectively after 24 h incubation at 37 °C. SEM analysis showed the absence of S. aureus biofilm formation on CarboSil SNAP-PTFE surfaces while control surfaces developed biofilms in 24 h. CarboSil SNAP-PTFE could also reduce platelet adhesion by 83.3 ± 4.5% compared to untreated controls after exposure to platelet-rich porcine plasma for 90 minutes under physiological conditions. Overall, the current study provides substantial proof of concept for NO-releasing polymer with a highly hydrophobic PTFE particle coating as an effective biomedical interface with potential medical device applications. The present study can be the basis for future studies on the tunability of PTFE coating and NO donors for long term applications.

Highlights.

• A facile bioinspired technique for immobilization of Polytetrafluoroethylene (PTFE) particles on a medical-grade polymer.

• Nitric oxide (NO) release from endothelial cell lining is mimicked to fabricate NO-releasing medical-grade polymer interface

• Combination of NO release and immobilized highly hydrophobic PTFE particles exhibited enhanced resistance to bacterial and platelet adhesion