Active release of an antimicrobial and antiplatelet agent from a non-fouling surface modification

Abstract

Two major challenges faced by medical devices are thrombus formation and infection. In this work, surface tethered nitric oxide (NO) releasing molecules are presented as a solution to combat infection and thrombosis. These materials possess a robust NO release capacity lasting ca. 1 month while simultaneously improving the non-fouling nature of the material by preventing platelet, protein, and/or bacteria adhesion. Nitric oxide’s potent bactericidal function has been implemented by a facile surface covalent attachment method to fabricate a triple action surface - surface immobilized S-nitroso-N-acetylpenicillamine (SIM-S). Comparison of NO loading amongst the various branching configurations is shown through the NO release kinetics over time and the cumulative NO release. Biological characterization is performed using in vitro fibrinogen and Staphylococcus aureus assays. The material with the highest NO release, SIM-S2, is also able to reduce protein adhesion by 65.8 ± 8.9% when compared to unmodified silicone. SIM-S2 demonstrates a 99.99% (i.e. ~4 log) reduction for Staphylococcus aureus over 24 h. The various functionalized surfaces significantly reduce platelet adhesion in vitro, for both NO releasing and non-NO releasing surfaces (up to 89.1 ± 0.9%), demonstrating the non-fouling nature of the surface immobilized functionalities. The ability of the SIM-S surfaces to retain antifouling properties despite gradual depletion of the bactericidal source, NO, demonstrates its potential use in long term medical implants.

Graphical Abstract

Introduction

Fouling of materials used in medical devices leads to increased risk of infection and device failure, and is the result of the adherence of proteins, bacteria, or thrombus formation. These complications can lead to large increases in healthcare costs and mortality.^1-2 While systemic heparin is often administered to aid in the prevention of thrombus formation, its prolonged use can result in morbidity and mortality while providing no activity to prevent infection.³ In addition to this, the increasing use of antibiotics has led to the development of resistant strains of bacteria, which can be attributed to the high dosages required to be effective against established biofilms.⁴ The degree of bacteria or platelet adhesion is highly influenced by the ability to prevent protein adhesion to the materials surface.^5-7 A number of strategies have been used to develop adhesion resistant materials,⁸ such as increasing the hydrophilicity,^{5, 9} super hydrophobic or patterned surfaces,^10-13 liquid-infused materials,^14-16 or grafting of polymer brushes.^17-24 While these materials are suitable for decreasing the adhesion of protein and bacteria through “passive” mechanisms, they provide no “active” mechanism to prevent platelet activation and adhesion or any bactericidal activity towards bacteria that have adhered, which ultimately leads to proliferation and biofilm formation.

Nitric oxide (NO) is an endogenous, gaseous, free radical that is produced naturally by macrophages and by endothelial cells lining the vascular walls, and is involved in various biological processes, such as preventing platelet activation and adhesion, while also being a potent, broad spectrum bactericidal agent.²⁵ To take advantage of these properties, NO donors (e.g. S-nitrosothiols or diazeniumdiolates) have been developed to allow for the storage and localized delivery of NO, and are particularly advantageous for polymeric materials typically used for medical devices, such as polyurethanes, silicones, or polyvinyl chloride.^26-27 The addition of these donors at various levels also provides a simple method for controlling the level of NO that is delivered from the materials.²⁸ Materials releasing NO have been shown to significantly reduce thrombus formation in both extracorporeal circuits and vascular catheter models, and have been shown to provide significant reductions in viable bacteria during long term catheterization.^{2, 29-30}

In this work, a novel method to prepare a material surface with the ability to not only offer an improved steric ability to prevent platelet and protein adhesion through a passive mechanism, but also utilize the active biocidal mechanism of NO. In addition to combining these mechanisms, the antifouling capability of the material is retained after the entire NO payload has been released from the surface, making it attractive for long-term applications. Specifically, this is achieved by the immobilization of the NO donor precursor N-acetyl-D-penicillamine (NAP) to various amine functionalized silicone surfaces (Figure 1A).

Figure 1:

A) Preparation of surface immobilized S-nitroso-N-acetyl-d-penicillamine (II. SIM-S1, III. SIM-S2, IV. SIM-S4) Structure I is a product of functionalization of PDMS surface with hydroxyl groups by submerging it in 50:50 ratio of 13 N HCl:30 wt.% H₂O₂ in H₂O and treatment with APTMES for amine functionalization. Structure II is a product of nitrosation of thiol groups with tert-butyl nitrite. Structure III and IV are synthesized after branching of primary amine via reaction with methyl acrylate and amine functionalization of branched site using ethylene diamine. B) FTIR spectra for different samples. Amine-1, Amine-2 and Amine-4 correspond to amine functionalized surfaces with unbranched and branched surfaces. 3500-2500 represents unreacted –COOH groups present after amine-functionalization for SIM-N4 and Amine-2. 2950 represents alkyl groups present in abundance in SR and aminated surfaces of SR. Double peaks of 1650,1550 represent primary amine groups in Amine-1, Amine-2 and Amine-4. 1650 represents saturated amide groups in SIM-N1, SIM-N2, SIM-N4, SIM-S1, SIM-S2 and SIM-S4. 1550 in SIM-S2 represents nitroso group of the NO-donor attached.

Materials and Methods

Materials:

N-Acetyl-D-penicillamine (NAP), sodium chloride, potassium chloride, sodium phosphate dibasic, potassium phosphate monobasic, phosphate buffered saline (PBS, pH 7.4 at 25°C), ethylenediaminetetraacetic acid (EDTA), tetrahydrofuran (THF), tert-butyl nitrite, and sulfuric acid were purchased from Sigma-Aldrich (St. Louis, MO). Aminopropyl trimethoxy silane (APTMES) was purchased from Gelest. All silicone substrates were fabricated with polydimethylsiloxane Sylgard 184 (Dow Corning). Methanol, hydrochloric acid, and sulfuric acid were obtained from Fisher Scientific (Pittsburgh, PA). Trypsin-EDTA and Dulbecco’s modification of Eagle’s medium (DMEM) were obtained from Corning (Manassas, VA 20109). The bacterial Staphylococcus aureus (ATCC 5538) strain was obtained from American Type Culture Collection (ATCC). Luria Agar (LA), Miller and Luria broth (LB), Lennox were purchased from Fischer BioReagents (Fair Lawn, NJ). All aqueous solutions were prepared with 18.2 MΩ deionized water using a Milli-Q filter (Millipore Corp., Billerica, MA).

Synthesis of SIM material:

Silicone films were first fabricated by mixing Sylgard 184 base to curing agent (ratio of 10:1). The solution was cast into Teflon molds and placed under vacuum for degassing. The casted solution was then placed in an oven (80°C, 90 min) for curing. To create a hydroxyl group functionalized surface, the silicone films were submerged in a mixture of 13 N HCl : 30 wt.% H₂O₂ (50:50) in H₂O under mild agitation (15 min). The surfaces were then rinsed with DI H₂O and dried under vacuum. The amine functionalization was then achieved by submerging the hydroxyl-functionalized surfaces in 5 wt.% APTMES in extra dry acetone for 2 h. Films were then rinsed with extra dry acetone to remove any non-covalently attached silane from the surface, and vacuum dried for 24 h. Branching of the immobilized moieties was achieved through incubation of the amine functionalized surface in 2:1 (v/v) methanol:methyl acrylate (24 h) followed by 2:1 (v/v) methanol: ethylenediamine (24h) as shown in Figure 1A. Samples were rinsed twice with methanol (20 mL) between incubating solutions. Amine-functionalized surfaces were then submerged in 10 mg mL⁻¹ NAP-thiolactone in toluene for 24 h, allowing for the ring opening reaction of thiolactone to bind to free amines.^31-32 The samples were then air-dried for 5 h to completely remove any residual solvent. Nitrosation of the immobilized NAP was achieved by incubation in neat tert-Butyl nitrite for 2 h. The resultant SIMS samples were stored at −20°C for further experiments.

Contact angle and FTIR analysis:

Surface properties and proof of attachment of nitric oxide donors to silicone surfaces was analyzed using contact angle measurements and FTIR. Static contact angle was measured using a DSA 100 drop shape analysis system (KRUÜSS) with a computer-controlled liquid dispensing system (Krüss). A 3 μL droplet of water was placed on various silicone films, and the average of left and right contact angles were measured via the Krüss software. Infrared spectroscopy studies of the samples were done using a Thermo-Nicolet model 6700 spectrometer with a grazing angle attenuated total reflectance accessory at 64 scans with a 6 cm⁻¹ resolution.

Nitric Oxide Release Characteristics:

Nitric oxide release from the films containing SNAP was measured using a Sievers Chemiluminescence Nitric Oxide Analyzer (NOA) 280i (Boulder, CO). The Sievers chemiluminescence Nitric Oxide analyzer is considered as the gold standard for detecting nitric oxide and is widely used due to its ability to limit interfering species, such as nitrates and nitrites, as they are not transferred from the sample vessel to the reaction cell. Films were then placed in the sample vessel immersed in PBS (pH 7.4, 37°C) containing 100 μM EDTA. Nitric oxide was continuously purged from the buffer and swept from the headspace using nitrogen sweep gas and bubbler into the chemiluminescence detection chamber.

Samples were analyzed for NO release and stored in the same conditions as found physiologically for medical implants/devices (shielded from light and at 37 °C). For each measurement, NO release was allowed to plateau so that burst effect of NO release was not included in the average flux for each time point measurement (approximately 0.5 h for each measurement). The detection limit of the NOA for measurement of gas-phase NO is ~0.5 part per billion by volume (~1 picomole).

Thiol quantification by Ellman's assay:

The covalent attachment of NAP-thiolactone can be directly related to the amount of free sulfhydryl groups on the surface of the PDMS. Measurement of these functional groups was done using Ellman's assay, which reacts 5,5'-dithio-bis-(2-nitrobenzoic acid) (DTNB) with free sulfhydryl groups to form conjugated disulfide and 2-nitro-5-thiobenzoic acid (TNB). While in solution, TNB's extinction coefficient has been recorded to be 14,150M-1 at 412 nm. Surface functionalized films were first cut to a recorded surface area before being placed in a solution containing 50 μL of DNTB stock solution (50 mM sodium acetate and 2 mM DTNB in DI water), 100 μL of PBS, and 850 μL of DI water. The samples were then thoroughly mixed and allowed to incubate at room temperature for 5 minutes. Optical absorbance was then recorded at 412 nm using a UV-Vis spectrophotometer. A standard calibration curve using acetyl cysteine was made to correlate thiol concentration with absorbance.

Protein Repulsion Quantification:

Levels of protein adhesion were quantified for the various materials using a modified version of a previously reported method.⁷ FITC-human fibrinogen (13 mg/mL, Molecular Innovations) was diluted to achieve 2 mg mL⁻¹ in PBS (pH 7.4). Silicone disks were incubated at 37°C for 30 min in a 96-well plate, followed by the addition of the stock protein solution to achieve a concentration of 2 mg mL⁻¹.⁷ Following 2h of incubation, infinite dilution of the well contents was carried out to wash away the bulk and any loosely bound protein from the materials. The fluorescence of each well (n=8) was then measured using a 96-well plate reader (Biotek Cytation 5), and the amount of protein adsorbed was determined via a calibration curve. The excitation and emission wavelength for FITC are 495 and 519 nm.

Bacterial Adhesion Assay:

The ability of the samples to 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 E2180 protocol with the commonly found nosocomial pathogen, Gram-positive S. aureus (ATCC 6538). A single colony of bacteria was isolated from a previously cultured LB-agar plate and incubated in LB Broth (37°C,150 rpm, 14-16h). The optical density of the culture was measured at a wavelength of 600 nm using a UV-vis spectrophotometer (Thermoscientific Genesys 10S UV-Vis) to ensure the presence of ~10⁸ CFU mL⁻¹. The overnight culture was then centrifuged at 2500 rpm for 7 min to obtain the bacterial pellet. The bacterial pellet obtained was resuspended in sterile PBS. The polymer samples (SR control, SIM-N1, SIM-S1, SIM-N2 and SIM-S2) were then incubated in the bacterial suspension (37°C, 24h, 140 rpm). After incubation, samples were removed from the bacterial suspension and rinsed with sterile PBS to remove any unbound bacteria. They were then sonicated for 1 min each using an Omni Tip homogenizer for 1 min to collect adhered bacteria in sterile PBS. To ensure proper homogenization of the collected bacteria, the samples were vortexed for 45s each. The solutions were serially diluted, plated on LB agar medium and incubated at 37°C. After 24h, the total CFUs for serially diluted and plated bacterial solutions were counted.

A second antimicrobial test was done, in addition to the one mentioned above, to evaluate the robustness of the NO-releasing materials when compared to antifouling materials. In this method, all steps were followed according to the process mentioned above except for the addition of an initial 24 h incubation in fibrinogen (2 mg/mL) from human serum. This incubation step was added before the bacterial incubation step due to the sequence of exposure seen in physiological conditions.³³ The results are shown on Figure S3.

The efficiency of the sample to inhibit bacterial attachment was calculated according to the following formula shown in Equation 1.

$$Bacteriainhibition(\%)=\frac{(CFUc{m}^{-2}oncontrolsample-CFUc{m}^{-2}ontestsample)}{(CFUc{m}^{-2}oncontrolsample)}\times 100$$ (Equation 1)

Platelet Adhesion Assay:

Freshly drawn citrated (3.8%, 9:1 citrate: whole blood) porcine blood was purchased from Lampire Biologicals. The anticoagulated blood was centrifuged (1100 rpm, 12 min) using the Eppendorf Centrifuge 5702. The platelet rich plasma (PRP) portion was collected carefully with a pipet as to not disturb the buffy coat. The remaining samples were then centrifuged (4000 rpm, 20 min) to retrieve platelet poor plasma (PPP). Total platelet counts in both 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 ca. 2 x 10⁸ mL⁻¹. Calcium chloride (CaCl₂) was added to the final platelet solution to achieve a final concentration of 2.5 mM.⁷ Disks of each respective surface were placed in a 5 mL blood tube. Approximately 4 mL of the calcified PRP was added to each tube and incubated (37°C, 90 min) with mild rocking (25 rpm). Following the incubation, the tubes were infinitely diluted with normal saline. The degree of platelet adhesion was determined using the lactate dehydrogenase (LDH) released when the adherent platelets were lysed with a Triton-PBS buffer using a Roche Cytotoxicity Detection Kit (LDH). The silicone disks were then incubated in 1 mL of Triton-PBS buffer. After 25 min, 100 μL of the buffer was transferred to a 96-well plate and combined with 100μL of the LDH reagent buffer per the supplier specifications. The absorbance of each well (duplicates of n=6,492 nm and 690 nm) was further duplicated) was then measured using a 96-well plate reader (Biotek Cytation 5), and the number of platelets adhered was determined using the calibration curve.

Results and Discussion

Functionalization and characterization of silicone surfaces for non-fouling coatings with active release of NO

A detailed schematic of the reactions used for branching of the initial alkyl-amine spacer is included in Supplement Figure S1. Briefly, the increasing grafting density of free amines was done through branching free amines using sequential reactions of 1:2 methyl acrylate: methanol and 1:2 ethylene diamine:methanol for 24 h. The surfaces were then incubated for 24 h in 10 mg/mL NAP-thiolactone (dissolved in toluene). Following immobilization, the free thiols of the grafted donor are nitrosated to its NO-rich form S-nitroso-N-acetyl-D-penicillamine (SNAP) using tert-butyl nitrite. To ensure covalent bonding of the surface modifications, FTIR measurements were carried out (Figure 1B). Since nitrogen atoms overlap in terms of FTIR peaks, appearance and disappearance of amide and primary amines were observed as the reaction steps were completed. This was followed by measurement of water contact angle (Table 1) to evaluate any significant differences in hydrophilicity of the functionalized surface. It is interesting to note here that hydrophilicity of the surfaces increased with increased NO-release (will be discussed in the next section). This could be attributed to lower availability of amine functionalized surfaces as the reaction is more complete.

Table 1:

Contact angle measurements compared between all NAP-thiolactone and nitroso group functionalized surfaces.

Material	Static Water Contact Angle (°)
SR	106.77 ± 3.36
SIM-N1	94.36 ± 3.36
SIM-S1	101.56 ± 4.48
SIM-N2	64.95 ± 11.87
SIM-S2	53.78 ± 5.23
SIM-N4	93.09 ± 2.91
SIM-S4	90.90 ± 7.97

As seen in the design strategy, the NO-load and release capacity of the materials was varied by branching of the initial alkyl spacer to increase the number of free amines. This variation in NO-load and release capacity was measured by using a chemiluminescence nitric oxide analyzer (NOA). The NOA is the gold standard for measurement of NO flux from materials and is a very efficient and sensitive instrument which can analyze NO release down to 1/10^th of ppb.³⁴ Samples are analyzed for NO release by shielding them from light and incubated in a phosphate buffered saline at 37 °C. These parameters mimic the physiological conditions of indwelling medical devices and have been used as a guideline in previous research.^{5, 35-36} One of the theoretical expectations was to see increasing NO-load and release measurements with an increase in branching. However, as seen in Figure 2, NO release measurements were significantly higher for SIM-S2 (cumulative release: 4. 34 ± 0.04 μmol cm⁻²) when compared to SIM-S4 (cumulative release: 2.33 ± 0.03 μmol cm⁻²) over the 25-d period. Tabulated values for instantaneous and cumulative NO release are in Supplementary Tables S1 and Table S2, respectively, and were comparable to total free thiol content as measured via Ellman’s assay (Figure S2).³⁷ There could be two possible explanations for this: steric hindrance in the case of higher branching and hence NAP thiolactone was not able to completely bind to the amine groups, and/or more branching increases the probability of chain interactions within the polymer during reactions with ethylene diamine, decreasing the total free amines available for binding with NAP thiolactone. Therefore, the conclusion from this study was that increased branching does not necessarily increase the total NO-load or release. Further, this enables the designed surface to release NO up to 25 d at effective flux levels (ability to reduce platelet activation and bacterial adhesion significantly).^{5, 35-36} This increasing branching method is a novel technique to increase NO release characteristics much like the function of metal ions when added to NO releasing polymers.³⁸ However, this material proves to be more advantageous as it also imparts antifouling characteristics to the material as seen in the following studies conducted. To ensure NO release was only from the surface functionalization and not from the bulk material due to possible swelling of the diamine group during the reaction period, control measurements were done on samples using the same reaction scheme without immobilization of the aminosilane. The results are not shown in figures or tables since no NO release was observed without the immobilization of the aminosilane. This proved that the aminosilanes immobilization was a necessary step for NO release and that NO release was not due to any swelling of the material from the solvents/reactions used for immobilization of the NO donor.

Figure 2:

A) Comparison of day by day NO release measurements between SIM-S1, SIM-S2, and SIM-S4. (n=3). B) Comparison of cumulative NO release from SIM-S1, SIM-S2, and SIM-S4. (n=3)

Assessment of the Non-fouling nature of various SIM surfaces against protein, bacteria, and platelet rich plasma

One common method for assessing the fouling of materials in vitro is to examine the ability of the material to resist non-specific protein adhesion, or if intended for blood contacting applications more specifically, fibrinogen (Fg). The adsorption of Fg to the material surface greatly aids in the ability for activated platelets or bacteria to bind to the surface, leading to higher risks of thrombus formation or infection.⁶ While the orientation of Fg adsorption has been shown to determine the degree of platelet adhesion, limiting protein adhesion regardless of orientation is generally considered to be an improvement in the hemocompatibility of a material.⁷ Developing NO-releasing materials that can reduce protein adsorption could provide drastic improvements in the overall hemocompatibility and antibacterial nature of these materials. To examine if the surface immobilized NO donors (both nitrosated and non-nitrosated) can provide a decrease in protein adhesion observed on NO-releasing materials, 2 h exposure to FITC-labeled fibrinogen (2 mg mL⁻¹) was conducted at 37°C (Figure 3A, Table 2). While minimal changes in contact angle were observed, increasing the branched nature of the surface grafted NAP groups decreased the degree of Fg adsorption, and is hypothesized to result from increases in steric hindrance.³⁴ However, altering the chemistry of the linkages to the amine functionalized surface could greatly increase the non-fouling ability of these materials. Overall, reductions in protein adsorption were observed to reach 65.8 ± 8.9% for SIM-S2 when compared to the unmodified SR. It is also interesting to note that the release of NO from the surface had no significant effect on the amount of adsorbed Fg. Previous reports have suggested that increasing NO release levels from other NO donors lead to increased fibrinogen adsorption, where additions of 1 and 9.2 wt.% N-diazeniumdiolated dibutyl-hexanediamine were added to polyvinylchloride films.³⁹ The NO release levels of 10 x10⁻¹⁰ mol cm⁻² and 15 x10⁻¹⁰ mol cm⁻² were shown to increase protein adsorption compared to PVC (226 ± 99% and 2334 ± 496%, respectively). However, extensive studies regarding the interaction of proteins and various NO donors have yet to be studied, specifically the differences between S-nitrosothiols and diazeniumdiolates, as well as a broader examination of the levels of NO release. However, the immobilized SIM-S surfaces demonstrate the effectiveness of the surface immobilized NO donors as providing active NO release while decreasing protein adsorption.

Figure 3:

A) Adsorption of fibrinogen to modified SR surfaces over a 2 h period. Values are expressed as mean ± standard error. Measurements were conducted using n=8 per group. B) SIM-S2 was able to reduce bacteria adhesion by ~4 log when compared to control samples. C) Comparison of adsorbed platelets per surface area between SR, SIM-N1, SIM-S1, SIM-N2 and SIM-S2.

Table 2:

Ability of various surface modified SR substrates to reduce nonspecific protein adsorption over 2h.

	SR	SIM-N1	SIM-S1	SIM-N2	SIM-S2
Fg Adsorption (μg cm−2)	72.4 ± 16.4	74.5 ± 13.7	51.0 ± 15.5	33.0 ± 11.0	24.7 ± 3.2
Reduction (%)	-	-	29.6 ± 26.7	54.4 ± 18.3	65.8 ± 8.9
p value vs control	-	NS	0.024	1.94 ×10−4	6.59 ×10−5
p value vs SIM-S2	6.59 ×10−5	1.43 ×10−5	0.027	NS	-

Bacterial adhesion, which ultimately results in biofilm formation, is a predominant issue for implanted devices aided by the moist and microbiome sustaining milieu. Coupled with fouling proteins, implants can become hosts to several pathogens that ultimately leads to medical device failure, infection (including bloodstream infection), and sometimes death.⁸ Antimicrobial efficacy of the designed non-fouling antimicrobial coating SIM-S materials was compared to the SR control samples to confirm their superior bacterial repulsion properties. The samples were incubated in bacterial solutions containing ~10⁸ CFU mL⁻¹ of S. aureus, which is one of the most commonly found nosocomial infection bacteria.^{35, 40} These infections are most commonly associated with catheters, stents, and prosthetic devices among other implants. As mentioned in the Introduction Section, our hypothesis was that the immobilized structure was expected to repel proteins and bacteria while simultaneously releasing NO to actively kill bacteria, thus enhancing the biocompatibility of the material even after all the NO load was depleted. The antimicrobial efficacy of the SIM-S surfaces was clearly observed after 24 h of incubation, the crucial time for initiation of bacterial adhesion on the surface that leads to infection. The CFU cm⁻² of viable S. aureus adhered to each sample was determined by plate counting (Figure 3B). SIM-S2 showed the highest bactericidal efficiency with a reduction of 99.99 ±0.002 % (Table 3) when compared to the SR samples, where a growth of ~10⁸ CFU cm⁻² was observed. This reduction is higher as compared to samples with only NAP thiolactone functionalization (SIM-N1= 82.14 ±22.20 % and SIM-N2= 96.86 ±0.50 %) and SIM-S1 (85.71 ±24.74 %). It can also be concluded from the results that NAP thiolactone functionalized surfaces alone only reduces bacteria adhesion because it cannot kill bacteria as it does not have any bactericidal property. However, the presence of protein is also known to drastically increase the bacterial adhesion, particularly those associated with medical devices. To further demonstrate the robustness of the antimicrobial activity of NO-releasing SIM-S surfaces and compare to only non-fouling SIM-N surfaces, in the presence of protein attachment similar to the methods described by Smith et. al (details in Supplementary section, Figure S3).⁴¹ These surfaces functionalized with NO-releasing moieties make effective bactericidal-releasing coatings which significantly enhance the antimicrobial efficacy. From the protein and bacterial adhesion assays, the varying effects of the modifiable NO-release kinetics from SR surface were observed which in turn can help significantly reduce dire clinical consequences of a medical implantation.

Table 3:

Ability of various surfaces to decrease bacterial adhesion over 24 h.

	SR	SIM-N1	SIM-S1	SIM-N2	SIM-S2
Average CFU of S. aureus cm−2	6.81 x106	1.22 x106	9.73 x105	2.14 x105	3.89 x102
Reduction (%)	-	82.14 ± 22.20	85.71 ± 24.74	96.86 ± 0.49	99.99 ± .002
p value vs control	-	0.01	0.01	0.02	0.02
p value vs SIM-S2	0.02	NS	NS	0.01	-

Platelet activation and adhesion are important considerations when determining the hemocompatibility of materials. Upon activation, platelets release several coagulation agonists, such as phospholipase A₂ (which is then converted into thromboxane A₂), which further increase platelet activation, the coagulation cascade, and thrombin generation.⁴² One key predecessor of platelet activation is the adsorption of fibrinogen to the materials surface, where changes in the protein conformation allows for binding to the Gp IIb/IIIa receptors on platelets. Therefore, reducing protein adhesion alone can act as a mechanism to reduce platelet activation. The aim of this study was to confirm that while NO-releasing SIM-S materials have been shown to significantly reduce platelet adhesion, the functionality of the modified surface is not lost after all the NO payload has been released. In fact, small molecules with free thiol groups (similar to NAP) have been shown to provide potent thrombolytic effects when administered systemically by binding to von Willebrand factor crosslinks of adhered platelets in arterial thrombi.⁴³ Both nitrosated and non-nitrosated surfaces were incubated in porcine platelet rich plasma for 90 min, where the degree platelet adhesion was then determined using a lactate dehydrogenase (LDH) assay (Figure 3C). Each variation of the surface modifications was able to provide significant reductions in platelet adhesion when compared to unmodified SR controls (Table 4). The SIM-N1 and SIM-S2 modifications provided the highest reductions, but were not statistically significant when compared to each other (p value > 0.5). However, as the brush size increased from the SIM-N1 to the SIM-N2 configuration, the ability to prevent platelet adhesion begins to decrease (p = 0.001). This may stem from the surface wetting characteristics of the methacrylate and diamine linkages, as a decrease in contact angle from 94.36 to 64.95 ± 11.87 was observed. The hypothesis is that the increase in platelet adhesion stems from the confirmation of fibrinogen adsorption, even though the overall protein adsorption was decreased.^{7, 44} While the SIM-S2 configuration did not provide significant reductions in platelet adhesion when compared to SIM-N1 or SIM-S1, the significant increase in NO release can provide increased bactericidal activity for extended durations.

Table 4:

Platelets adsorbed per surface area over a period of 90 mins.

	Control SR	SIM-N1	SIM-S1	SIM-N2	SIM-S2
Platelets cm−2 (x106)	13.5 ± 0.4	1.5 ± 0.1	2.8 ± 0.1	3.6 ± 0.2	1.6 ± 0.1
Reduction (%)	-	89.1 ± 0.9	79.1 ± 1.0	73.4 ± 1.3	87.7 ± 0.7
p value (x106)	-	0.1	1.7	0.5	0.2

Conclusion

In summary, this study presents a facile method to attach various amounts of the NO-releasing donor, SNAP, to any polymer material in order to provide both bactericidal/antiplatelet activity while adding a non-fouling nature to the material surface. Through the covalent attachments, the NO release characteristics from the modified surface were tunable by varying the branching linkers for NO donating structures. This enabled detailed analysis of the protein repelling, antithrombotic, and antibacterial properties of each of the modified surfaces that had varying degrees of NO release for different spans of time. It is a significant progress towards quantitatively controlling the release of NO and hence use this technique to further study NO-releasing properties for applications including microfluidic devices for high-throughput assays. This method will also be highly applicable for biomedical device materials that are prone to infection- and thrombosis-related failures, and can easily be coupled with existing NO-releasing polymers.

Supplementary Material

Supporting Information_Brush

Figure S1: Detailed schematic of various SIM surfaces

Figure S2: A) Quantification of free thiols on various SIM surfaces. B) Calibration curve of free thiols

Figure S3: CFU of S. aureus per cm² of sample after 24 h exposure to fibrinogen from human serum followed by 24 h of S. aureus incubation under physiological conditions (n=5)

Table S1: Day by day NO release measurements for 600 h/25 d. (n=3)

Table S2: Cumulative NO release over 600 h/25 d. (n=3)