Surface-Catalyzed Nitric Oxide Release via a Metal Organic Framework Enhances Antibacterial Surface Effects

Abstract

It has been previously demonstrated that metal nanoparticles embedded into polymeric materials doped with nitric oxide (NO) donor compounds can accelerate the release rate of NO for therapeutic applications. Despite the advantages of elevated NO surface flux for eradicating opportunistic bacteria in the initial hours of application, metal nanoparticles can often trigger a secondary biocidal effect through leaching that can lead to unfavorable cytotoxic responses from host cells. Alternatively, copper-based metal organic frameworks (MOFs) have been shown to stabilize Cu^2+/1+ via coordination while demonstrating longerterm catalytic performance compared to their salt counterparts. Herein, the practical application of MOFs in NO-releasing polymeric substrates with an embedded NO donor compound was investigated for the first time. By developing composite thermoplastic silicon polycarbonate polyurethane (TSPCU) scaffolds, the catalytic effects achievable via intrapolymeric interactions between an MOF and NO donor compound were investigated using the water-stable copper-based MOF H₃[(Cu₄Cl)₃(BTTri)₈-(H₂O)₁₂]·72H₂O (CuBTTri) and the NO donor S-nitroso-N-acetyl-penicillamine (SNAP). By creating a multifunctional triple-layered composite scaffold with CuBTTri and SNAP, the surface flux of NO from catalyzed SNAP decomposition was found tunable based on the variable weight percent CuBTTri incorporation. The tunable NO surface fluxes were found to elicit different cytotoxic responses in human cell lines, enabling application-specific tailoring. Challenging the TSPCU–NO–MOF composites against 24 h bacterial growth models, the enhanced NO release was found to elicit over 99% reduction in adhered and over 95% reduction in planktonic methicillin-resistant Staphylococcus aureus, with similar results observed for Escherichia coli. These results indicate that the combination of embedded MOFs and NO donors can be used as a highly efficacious tool for the early prevention of biofilm formation on medical devices.

Graphical Abstract

1. INTRODUCTION

Bloodstream and urinary tract infections from the surface contamination of central venous catheters, vascular access ports, urinary catheters, and other indwelling medical devices are responsible for a significant portion of nosocomial infections along the continuum of acute to long-term hospice care.¹ Skin microflora, the diverse group of microorganisms colonizing the skin surface, consists of several strains such as Staphylococcus epidermidis, Staphylococcus aureus, and Escherichia coli, which can undergo the transfer of antibiotic-resistant genes into several highly virulent strains.^2,3 This often leads to the development of drug-resistant strains, which creates an ever-evolving challenge through hospital-acquired infections (HAIs) that motivates the development of new material strategies to enhance medical device resistance to microbial contamination.

In 2002, the estimated incidence of HAIs in hospitals across the United States was 4.5%, equating to almost 1.7 million patients across medical facilities or 4.5 infections per 100 admissions.⁴ Whereas urinary tract infections were historically the most frequent HAI in developed countries, recent prevalence studies have highlighted the matched prevalence of lower respiratory tract and surgical site infections.⁵ The economic burden of HAIs has been estimated to be over $28 billion annually in the United States alone.⁶ In many of these cases, HAIs result in medical device occlusion and failure, causing further patient morbidity and mortality as well as health-care costs. In many cases, antimicrobial therapies are limited by dosing tolerances, the presence of multi-drug-resistant strains, and the development of biofilms that make last-resort antibiotics an ineffective strategy.⁷

Nitric oxide (NO) is a promising agent for the prevention of medical device surface contamination by pathogens due to its ability to be incorporated through donor compounds into medical-grade plastics via simple blending, solvent casting, or covalent immobilization to form NO-releasing (NOrel) substrates.^8-10 Because NO relies on mechanisms of oxidative and nitrosative stress alone and in tandem with derivative species (e.g., peroxynitrite, NO₂, N₂O₃, and N₂O₄),¹¹ NOrel materials can be tailored against multi-drug-resistant bacteria associated with nosocomial infection. NOrel materials have also shown promising potential to eradicate mature biofilms, which are often otherwise difficult to treat with traditional antibiotics due to poor drug permeability.¹² Despite these many advantages for medical device surface functionalization, NOrel substrates have faced several challenges, including (1) cytotoxic leachate accumulation in physiological environments, (2) poor NO surface flux longevity, and (3) incomplete consumption of the finite NO donor reservoir.¹³ For these reasons, research has focused in recent years on the incorporation of metal ions into NOrel substrates to enhance the longevity of the physiologically active NO surface flux (i.e., 0.5–4.0 × 10⁻¹⁰ mol cm⁻² min⁻¹ for endothelial cells).¹⁴

Metal ion effects from nanoparticles (e.g., Cu, Se, Zn, In, and other metal ions)^15-18 and organometallic compounds (e.g., RSe)^19,20 have been investigated for the facilitation of decomposition of S-nitrosothiols (RSNOs) to generate NO. Cu²⁺ was originally hypothesized to undergo reduction by thiolate ions (RS⁻) in solution and form Cu¹⁺ (eq 1), which can react with RSNOs to form NO and disulfide derivatives (RSSRs) (eq 2)^21,22

$${\text{Cu}}^{2+}+{\text{RS}}^{-}\to {\text{Cu}}^{1+}+0.5\text{RSSR}$$ (1)

$${\text{Cu}}^{1+}+\text{RSNO}\to {\text{Cu}}^{2+}+{\text{RS}}^{-}+\text{NO}$$ (2)

However, recent work²³ using direct ¹H NMR monitoring of RSNO decomposition has demonstrated a different route to NO generation in the absence of the precursor thiol (RSH)

$$\begin{gathered} 0.25{\text{Cu}}^{2+}+2\text{RSNO} \\ \to 2\text{NO}+0.75\text{RSSR}+0.25[\text{RSSR}-{\text{Cu}}^{2+}] \end{gathered}$$ (3)

This study also demonstrated that the overall reaction efficiency of the copper-catalyzed decomposition of RSNOs is dependent on the molar ratio of RSNO to its precursor RSH.²³

By incorporating organometallic species onto polymer surfaces, catalytic NO generation from endogenously supplied or supplemented NO donors such as S-nitrosocysteine (CysNO), S-nitrosoglutathione (GSNO), and S-nitroso-N-acetyl-penicillamine (SNAP) has shown remarkable improvements in the infection resistance and biocompatibility of medical device materials.^24-26 Similarly, embedding metal nanoparticles into polymer matrices impregnated with NO donor compounds has shown to stabilize NO surface flux from the materials, sustaining physiologically relevant release rates for over 7 days while also possessing synergistic antibacterial properties through metal ion interactions.^15,16,20 However, these materials have several drawbacks, mainly concerning the potential cytotoxic effect elicited by metal ion leachates. Therefore, it is ideal to develop polymeric materials that present ions such as Cu^2+/1+ in solid supports such as coordination complexes that are suitable for RSNO decomposition with minimal potential for cytotoxic leaching.²⁷

Metal organic frameworks (MOFs) have been studied extensively as catalysts for NO generation from RSNOs from both endogenous blood sources and supplemented RSNO donor pools.²⁸ Traditionally, MOFs have been considered in biomedical applications for NO storage via its adsorption onto framework pores and coordination to open metal sites (i.e., chemisorption), as well as by postsynthetic modification of MOFs to develop covalent NO donor moieties (i.e., diazeniumdiolates).^29-31 However, the relative difficulty of controlling the observed NO release rates in these processes, especially with regard to the catalytic potential of metal ion complexes to decompose RSNOs, has prompted research interest in NO generation via cleavage of external NO donors in the form of low-molecular-weight RSNOs. Early efforts considered Cu(II)1,3,5-benzenetricarboxylate (CuBTC) for NO generation but faced difficulty with CuBTC crystals decomposing to constituent ligands and metal ions under physiological conditions and during fabrication steps due to its poor water stability.³² In contrast, the water and organic solvent-stable MOF Cu(II)1,3,5-benzene-Tris-triazole (CuBTTri) has gained traction in recent years as a versatile platform for biomimetic catalysis of both endogenous and supplemented NO donors with stability on a versatile array of hydrophilic polymeric scaffolds, including chitosan,^33,34 polyurethane,³⁵ and polyvinyl alcohol.³⁶ Stabilization of Cu^2+/1+ ions via coordination in MOFs such as CuBTTri has several distinct advantages for NO release from endogenous donors such as GSNO. While Cu²⁺ ions are readily reduced to Cu¹⁺ in the presence of free RSH, the stoichiometric ratio of RSHs with respect to Cu²⁺ becomes an important parameter for optimization, as large ratios can lead to RSH complexing with Cu²⁺, resulting in catalyst quenching.²³ In contrast, stoichiometric quantities of RSH have been shown to not deactivate CuBTTri and may instead contribute to potential redox cycling and subsequent formation of lower-coordinate Cu¹⁺ sites on the surface of the MOF, conferring surface site catalysis of RSNOs.^23,37 The water stability of CuBTTri in addition to the conservation of its catalytic activity establishes CuBTTri as an attractive substrate for NO generation on medical device surfaces.

Despite the many advantages of incorporating CuBTTri into polymers for NO generation from external NO donor compounds, its application for antibacterial medical device surfaces has been insofar limited by (1) the reduced catalytic efficiency of MOFs upon polymeric impregnation, (2) the reliance on maximal catalytic efficiency to confer physiological effect from endogenous NO donor reservoirs (e.g., CysNO and GSNO), and (3) the application-specific need to supplement RSNOs in solution.^35,38 Considering prior mentioned disadvantages observed with traditional NO-releasing materials and those from metal nanoparticle impregnation, the combination of NO-releasing substrates with an MOF interface establishes precedence for a sustained, intrinsically NO-releasing material with negligible cytotoxic effect or secondary antibacterial effect from metal ion leaching.

Herein, we demonstrate for the first time the direct application of a Cu-based MOF in a multilayer scaffold with an embedded NO donor for sustained NO release to achieve an antibacterial surface. Synthesized CuBTTri was characterized by powder X-ray diffraction (pXRD), Fourier transform infrared spectroscopy (FTIR), and scanning electron microscopy (SEM). The NO donor SNAP was impregnated into a base layer of thermoplastic silicon polyurethane polycarbonate (TSPCU) via simple blending, and then the film was topcoated with a layer of CuBTTri suspended in TSPCU and an additional TSPCU sealant layer (Scheme 1). These film composites were evaluated for time-dependent NO surface flux based on the weight percentage of CuBTTri incorporated and evaluated for changes in mechanical properties, surface wettability, and leaching. Composites were evaluated for cytotoxic effect in a 24 h leachate viability assay with human fibroblast and endothelial cells. Selecting formulations based on cytotoxicity thresholds, candidates were further evaluated in a set of 24 h bacterial growth studies for both planktonic and adherent growth of methicillin-resistant S. aureus (MRSA) and E. coli.

Scheme 1. Fabrication of Composite TSPCU Films Containing SNAP and CuBTTri^a.

^a TSPCU is first dissolved in anhydrous tetrahydrofuran (THF) with SNAP and cast in a mold overnight. A dip-coating solution of TSPCU and CuBTTri of variable weight percentage is prepared. The TSPCU–NO film is then dip-coated with the TSPCU–CuBTTri solution once to form a thin catalytic TSPCU–MOF layer. Finally, to control leaching and water permeation into the two active layers, a third sealant layer of just TSPCU is dip-coated onto the composite to afford TSPCU–NO–MOF.

2. EXPERIMENTAL METHODOLOGY

2.1. Materials.

CarboSil-2080A thermoplastic silicon polycarbonate polyurethane (TSPCU) was purchased from DSM Biomedical (Exton, PA). Hydrochloric acid, ethylenediaminetetraacetic acid (EDTA), methanol, N-acetyl-penicillamine (NAP), sulfuric acid, and tetrahydrofuran (THF) were acquired from Sigma-Aldrich (St. Louis, MO). Copper(II) chloride dihydrate, copper(I) iodide, and 1,3,5-triethynylbenzene were acquired from Alfa Aesar (Tewksbury, MA). Copper(II) chloride dihydrate was obtained from Acros Organics (Fair Lawn, NJ). Deionized water (18.2 MΩ) was prepared using an in-house distillation unit from Mettler Toledo (Columbus, OH) and used in all aqueous solution preparations. Phosphate-buffered saline (PBS) at pH 7.4 and supplemented with 138 mM sodium chloride, 2.7 mM potassium chloride, 10 mM sodium phosphate, and 100 μM EDTA was used in all nitric oxide analyzer (NOA) experiments and where else noted. Phosphate-buffered saline without calcium or magnesium (DPBS) was purchased from Corning (Manassas, VA) and was used for all in vitro studies. Nitrogen and oxygen gas cylinders were acquired from Airgas (Kennesaw, GA).

Methicillin-resistant S. aureus (MRSA) (ATCC BAA-41) and E. coli (ATCC 25922) were acquired from American Type Culture Collection (Manassas, VA). Luria-Bertani (LB) broth and agar, as well as Mueller–Hinton broth (MHB) and agar, were purchased from Sigma-Aldrich (St. Louis, MO). All other bacteria-related supplies were purchased from VWR (Atlanta, GA).

Human BJ fibroblast (ATCC CRL-2522) cells were subcultured from stocks originally obtained from American Type Culture Collection (Manassas, VA). Gibco human umbilical vein endothelial cells (HUVECs) were obtained from Thermo Fisher Scientific (Waltham, MA). EGM-2 endothelial cell growth basal media with EMG-2 supplements was purchased from Lonza (Greenwood, SC). Minimum essential medium (MEM), fetal bovine serum (FBS), and all other cell culture-related supplies were purchased from VWR (Atlanta, GA).

2.2. Synthesis of S-Nitroso-N-acetyl-penicillamine (SNAP).

Synthesis of the NO donor SNAP was based on previously reported procedures followed with minor deviation.^39,40 N-Acetyl-penicillamine (23 mmol, 5 g) was dissolved in methanol (77% v/v, 130 mL) and covered from light in an ice bath. Following solubilization, concentrated HCl (20 mL) and sulfuric acid (5 mL) were slowly aliquoted into the mixture. Meanwhile, sodium nitrite (58 mmol, 4 g) was dissolved in deionized water (50 mL) and then added to the mixture. A siphon was placed inside the reaction vessel with a low airflow rate for the facilitation of solvent evaporation and SNAP crystallization. After 6 h, the solid product was collected through vacuum filtration and was washed vigorously with chilled water to remove unreacted nitrite. The crystalline product was dried under vacuum overnight and stored at –20 °C with protection from light. The final SNAP purity was verified against ¹H NMR and nitric oxide loading determined using chemiluminescence-based nitric oxide release analysis described herein. Only batches with greater than 95% product purity (i.e., moles of NO per mole of SNAP × 100%) were used in experiments.

2.3. Synthesis of Cu(II)1,3,5-Benzene-tris-triazole (CuBTTri).

CuBTTri was synthesized using methods adapted from Demessence et al.⁴¹ In brief, the ligand 1,3,5-tris(1H-1,2,3-triazol-5-yl)benzene (H₃BTTri) was first prepared by combining copper(I) iodide (505 mg, 2.60 mmol) and 1,3,5-triethynylbenzene (2.65 g, 17.6 mmol) in dimethylformamide (DMF) (90 mL) and methanol (10 mL) under anaerobic atmosphere. Trimethylsilylazide (9.16 g, 79.5 mmol) was then added, and the reaction was allowed to stir for 36 h at 100 °C. The resulting solution was filtered, with the filtrate then concentrated to approximately 10 mL. Water (30 mL) was added to the filtrate, which resulted in a pale green precipitate, H₃BTTri. The precipitate was washed with water and diethyl ether and then dried under vacuum.

To synthesize CuBTTri (H₃[(Cu₄Cl)₃(BTTri)₈]), H₃BTTri (225 mg, 0.937 mmol) was first dissolved in DMF (40 mL) via sonication. Copper(II) chloride dihydrate (283 mg, 2.25 mmol) was then added to the H₃BTTri solution and heated to 100 °C for 72 h. The purple precipitate, CuBTTri, was collected via filtration and washed with DMF and water. CuBTTri was then added to water (50 mL) and heated to 100 °C for 24 h. The final product was collected via filtration and washed with water three more times. CuBTTri was used in its hydrated form, H₃[(Cu₄Cl)₃(BTTri)₈-(H₂O)₁₂]·72H₂O.

2.4. Fabrication of SNAP and CuBTTri–SNAP Films.

The synthesis of SNAP and CuBTTri–SNAP films was completed through solvent evaporation followed by two dip-coating steps (Scheme 1). TSPCU (2000 mg) was dissolved in THF (40 mL) on a stirring plate while being heated to 65 °C for 2 h. Once the solution has been cooled to room temperature, 10 wt % SNAP (222 mg) was added when applicable and mixed until dissolved. The polymer solution (36 mL) was then dispensed into a square Teflon mold (3 in × 3 in) and allowed to dry overnight with protection from light. The resulting film was cut into approximately 10 mm wide strips. TSPCU and TSPCU–NO control composites were then dip-coated first in TSPCU in THF (50 mg/mL) with 15 min of drying, followed by a second dip coat in the TSPCU solution with drying overnight. The rest of the film strips were first dip-coated in a solution of TSPCU in THF (50 mg/mL) with the respective CuBTTri [1 3 5] wt % (i.e., the mass of CuBTTri per mass of TSPCU × 100%). After a 15 min drying period, a second dip coat layer consisting of TSPCU in THF (50 mg/mL) was applied and dried overnight. The dip-coated strips were then hole-punched with an 8 mm biopsy punch to produce the final films used for testing. Films were protected from light during the drying processes and stored at −20 °C after fabrication.

2.5. Proton Nuclear Magnetic Resonance (¹H NMR) Spectroscopy.

¹H NMR spectra of the as-prepared SNAP were collected using a Varian Mercury 300 equipped with a 5 mm broad-band roomtemperature probe with a 1H/19F channel and a proton signal-to-noise ratio of 161. Samples were prepared in deuterated water. Spectra were collected from a total of 128 scans and processed using MestreNova and NMRbox software.⁴²

2.6. Attenuated Total Reflectance-Fourier Transform Infrared (ATR-FTIR) Spectroscopy and Powder X-ray Diffraction (pXRD) Studies.

ATR-FTIR spectra of the as-prepared CuBTTri were collected using a Nicolet iS50 FTIR spectrometer with an ATR accessory equipped with a Ge diamond crystal. Spectra were collected from 4000 to 650 cm⁻¹ with a total of 32 scans and a resolution of 4 cm⁻¹. pXRD patterns were collected with a Bruker D8 Discover DaVinci Powder X-ray Diffractometer with a Cu Kα radiation at a rate of 0.2 seconds/step with a step size of 0.02°.

2.7. Scanning Electron Microscopy (SEM) and Energy-Dispersive X-ray Spectroscopy (EDS).

SEM imaging and corresponding EDS analyses of TSPCU films were performed using a Thermo Fisher Scientific Teneo Field Emission SEM. Specimens were mounted onto aluminum specimen mounts using double-sided black carbon adhesive. Mounts were sputter-coated with a 10 nm Au–Pd powder via a Leica EM ACE200 sputter coater (Buffalo Grove, IL). Surface imaging of films was done with an accelerating voltage of 20 kV.

2.8. Tensile Testing.

Ultimate tensile strength (UTS) was determined for the TSPCU composite formulations using a Mark-10 Force and Tensile Measurement System (Copiague, NY). In brief, TSPCU composite films were die cut following an ASTM D-1708 template and subjected to increased load until breaking. The load at break was normalized to the gauge area of the specimens to calculate UTS. The final data are shown as mean ± standard deviation (SD) (N = 3 independently prepared films per sample type).

2.9. Static Contact Angle Measurement.

Static contact angles of deionized water on the TSPCU films were investigated using an Ossila Contact Angle Goniometer (Sheffield, U.K.) on 1 cm² film specimens. Five independently prepared films of each sample type were evaluated for static contact angles using 5 μL droplets of deionized water. Video imaging of the substrates at a frame rate of 20 frames per second was performed to evaluate the drop shape during the first 10 s of surface contact. Using a combination of droplet edge detection and a polynomial fitting model, the contact angle in each video frame was measured with an average calculated using Ossila Contact Angle (v3.0.3.0) software. Final averages are reported as the mean ± SD (N = 5 films evaluated per sample type).

2.10. SNAP Leaching.

Determination of the leached SNAP content from the TSPCU films was determined following a previously reported method with minor deviation.⁴⁰ TSPCU composite films fabricated with or without CuBTTri (Scheme 1) were soaked in 1 mL of PBS supplemented with EDTA (100 μM) at 37 °C and protected from light over a 24 h time-course experiment. At time points of 0.5, 2, 4, 12, and 24 h, the buffer was siphoned off for analysis and replaced with fresh buffer. The absorbance of the analyte was measured at a wavelength of 340 nm, corresponding to the wavelength of maximum absorbance for SNAP from its S-nitrosothiol bond, by use of a Genesys 10S UV–vis spectrophotometer (Thermo Fisher Scientific). Using a standard curve prepared from serial dilutions of SNAP in PBS with EDTA, the concentration of SNAP in each analyte (C_SNAP,analyte) was calculated against the leachate volume (V_analyte) and initial quantity of SNAP in each film sample (m_SNAP,loaded, approximately 2 mg) to determine the remaining SNAP, as shown in eq 4

$$\begin{gathered} (\%\text{SNAP remaining}{)}_t \\ =\frac{m_{\text{SNAP,loaded}}-[(C_{\text{SNAP,analyte}})\times V_{\text{analyte}}{]}_t}{m_{\text{SNAP,loaded}}}\times 100\% \end{gathered}$$ (4)

The percent SNAP remaining (mol/mol × 100%) and cumulative SNAP leached (nmol SNAP/mg film) are reported as a mean ± SD at each time point (N = 5 films per sample type).

2.11. MOF Leaching.

2.11.1. Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES).

In this work, ICP-OES was implemented to detect any CuBTTri leaching and subsequent decomposition leading to Cu ion accumulation following incubation under physiological conditions. In each study, 8 mm films of TSPCU–NO–MOF and TSPCU–MOF were incubated in DPBS (1 mL) for 24 h at 37 °C to mimic physiological conditions wherein leaching may occur. Following 24 h incubation, the samples were diluted 1:5 in 2% nitric acid for introduction to the ICP-OES system. Cu content was calculated against a standard curve. The total mass of Cu leached was then determined against the total mass of CuBTTri in each film, which was calculated from the average difference in the mass of films after the first dip coat with TSPCU–MOF. The final average mass of Cu leached per mg of CuBTTri incorporated into each film was calculated and reported as the mean ± SD of independently prepared film samples (N = 3).

2.12. Chemiluminescence-Based Detection of NO Release.

NO release from TSPCU films was investigated over a 7 days longevity study via chemiluminescence-based detection from a Zysense Nitric Oxide Analyzer (NOA) Model 280i. In a representative study, an 8 mm circular punchout of a composite specimen approximately 0.33 mm in thickness was prepared to achieve a final surface area of exactly 1.01 cm². Sample chambers were pretreated using 1× PBS with EDTA to neutralize any residual metal ions in the chamber. Films were immersed in 2 mL of PBS (calcium and magnesium free) in an amber glass sample chamber heated in a circulating water bath to 37 °C. Setting the sample chamber pressure to 6 torrs with an oxygen pressure of 6.0 psi, a dry nitrogen purge stream (200 cm³/min) was passed through the solution phase inside the chamber, instantaneously sweeping NO out of the chamber into the reaction chamber of the NOA. Wherein, NO reacted with ozone simultaneously fed in from a separate stream to form nitrogen dioxide in its excited state. The relaxation of the excited state results in photoemission that is internally detected and quantified by an internal photomultiplier tube. By correlating this emission to a standard inlet stream of 45 PPM NO, a calibration constant (mol NO/PPB × s) was determined and used to calculate the instantaneous NO surface flux (mol NO/cm² × s). The final results are reported as the mean ± SD of each composite formulation (N = 5).

2.13. In Vitro Evaluation of Mammalian Cell Cytocompatibility.

Experiments for the cytocompatibility testing of TSPCU films following ISO 10993-5 standards for the biological evaluation of medical devices.⁴³

2.13.1. Mammalian Cell Culture.

BJ cells were revived from cryopreservation stocks and cultured in MEM media supplemented with 10% FBS and 1% P/S under a 5% CO₂-humidified atmosphere at 37 °C. Similarly, HUVECs were cultured from cryopreserved stocks in EGM-2 basal media supplemented with growth factors for endothelial cell growth following the manufacturer’s recommended protocol. Cells were subcultured for up to 10 passages between experiments. Once cell monolayers developed >70% confluency, cells were detached using 0.05% trypsin supplemented with 5 mM EDTA. Cells were collected via centrifugation, stained with trypan blue dye, and counted using an EVE cell counting system from NanoEnTek (Waltham, MA). For subsequent cytotoxicity testing, cells were seeded in 96-well plates at a density of 10 000 cells/well and grown for 24 h to achieve >70% monolayer confluency before testing.

2.13.2. Film Leachate Cytotoxicity Testing (24 h).

Circular punchouts of TSPCU films of 8 mm diameter were prepared for each TSPCU formulation, with caution to normalize the sample surface area across each formulation. The film specimens were subjected to UV sterilization for 15 min on both faces of the film. Afterward, each film was placed individually in 1 mL of supplemented media, sterilecapped, and incubated for 24 h at 37 °C under a humidified 5% CO₂ atmosphere. Control media containing no leachate was also incubated. Seeded 96-well plates were grown for 24 h. Afterward, the media in wells was replaced with the incubated leachate media for the different TSPCU formulations as well as media controls. Media from a given TSPCU film was plated across eight replicates, with a total of three films tested per cell passage across three independent passages of each cell line. Following 24 h incubation in leachate-containing media, the media in each plate was replaced with clean supplemented media containing 10% CCK-8 reagent. Cells were incubated with CCK-8 for 3 h based on previously developed standard curves for each cell line and read for absorbances at optical densities (ODs) of 450 and 650 nm. The final relative cell viability of each test condition was calculated according to eq 5. Final values are reported as mean ± SD (N = 3 treatment sets across independent passages). Standard curves based on 24 h exposure of cell monolayers to different concentrations of SNAP were also developed for reference (N = 3 treatment sets across independent passages)

$$\begin{gathered} \text{relative cell viability} \\ =\frac{({\text{OD}}_{450}\text{read}-{\text{OD}}_{650}\text{read}{)}_{\text{leachate treatment group}}}{({\text{OD}}_{450}\text{read}-{\text{OD}}_{650}\text{read}{)}_{\text{untreated group}}}\times 100\% \end{gathered}$$ (5)

2.14. In Vitro Evaluation of Bacterial Inhibition.

The antibacterial activities of the TSPCU composites were evaluated against Gram-positive methicillin-resistant S. aureus (MRSA) and Gram-negative E. coli compared to those of unmodified materials.

2.14.1. Bacterial Viability Study (24 h).

In vitro antibacterial assays were conducted to quantify the number of viable planktonic and adhered bacteria after 24 h of exposure using a prior protocol.⁴⁴ Briefly, each bacterial strain was inoculated from frozen stock and incubated at 37 °C overnight at 150 rpm. MRSA was inoculated into MHB broth, and E. coli was inoculated into LB broth. Each culture was then centrifuged for 7 min at 4400 rpm and washed with sterile PBS (pH 7.4). The cultures were centrifuged again and resuspended in PBS supplemented with media. The MRSA culture was resuspended in PBS supplemented with 250 mg/mL MHB media. The E. coli culture was resuspended in PBS supplemented with 1 g/L LB media. The optical density of the resuspended cultures was taken at 600 nm via UV–vis to determine the concentration of log-phase bacterial cells in the suspension. The cultures were then diluted to roughly 10⁸ colony forming units (CFUs) per mL. Materials were brought into the biosafety cabinet and sterilized by UV light. Each material was placed into a well of a 24-well plate, and then 1 mL of 10⁸ CFUs/mL diluted bacterial suspension was pipetted into each well. The plate was incubated for 24 h at 37 °C at 150 rpm. To quantify the number of adhered bacteria on the materials, each material was removed from the well plate and gently washed with 1 mL of sterile PBS before being placed into a 15 mL centrifuge tube containing 1 mL of sterile PBS. The materials were each homogenized for 60 s at 25 000 rpm to remove any adhered bacteria from the surface. One milliliter of PBS containing the removed bacteria was then serially diluted and plated. To quantify planktonic bacteria, the solutions within each well of the 24-well plate were serially diluted and plated. MRSA dilutions were plated on MHB agar plates, while E. coli dilutions were plated on LB agar plates. The plates were incubated overnight at 37 °C, and the CFUs on each were counted to quantify the viable bacteria. Results are reported at CFUs per cm² (for viable adhered bacteria) and CFUs per mL (for viable planktonic bacteria) and analyzed to gauge material efficacy to inhibit bacterial adhesion and growth, respectively, under physiological conditions (eq 6). Results are reported as mean ± SD (N = 4 per film type)

$$\begin{gathered} \text{reduction efficacy} \\ =\frac{{\text{count}}_{\text{control}}({\text{CFU cm}}^{-2})-{\text{count}}_{\text{treated}}({\text{CFU cm}}^{-2})}{{\text{count}}_{\text{control}}({\text{CFU cm}}^{-2})} \\ \times 100\% \end{gathered}$$ (6)

2.15. Statistical Analysis.

Data are reported as mean ± standard deviation (SD) if not otherwise explicitly stated. All statistical analyses were carried out using Prism 9.1 (Graphpad Software, San Diego, CA). Statistical comparisons of treatment groups against control groups in biological studies were determined using ordinary one-way analysis of variance (ANOVA), with corrections for multiple comparisons between means of the sample groups using Tukey’s method. Time-course-dependent studies (i.e., NO release studies and leaching) were analyzed for statistical significance using two-way ANOVA with Tukey’s method for correction of multiple comparisons. Values of P < 0.05 were deemed significant.

3. RESULTS AND DISCUSSION

3.1. Materials Characterization.

Synthesized SNAP was characterized by ¹H NMR for confirmation of nitrosation and purity of the final product (Figure S1) via identification of chemical shifts consistent with previous reports.⁴⁵ The prepared CuBTTri (Figure 1A) was characterized via ATR-FTIR and pXRD studies for chemical and structural confirmation based on our prior work (Figure 1B,C).³⁵ ATR-FTIR confirmed the complete exchange of DMF with water based on the absence of characteristic DMF peaks in the spectra (Figure 1B). The diffraction pattern collected by pXRD was found to be consistent with those from previous reports,^35,37,41 confirming the crystallinity of the synthesized CuBTTri based on the presence of sharp, distinct peaks (Figure 1C).

Figure 1.

Characterization of CuBTTri and TSPCU–NO–MOF films after incorporation of the synthesized CuBTTri. (A) Idealized subunit structure of CuBTTri with H atoms omitted for clarity. (B) FTIR of CuBTTri demonstrating the exchange of DMF from the MOF. (C) pXRD pattern of prepared CuBTTri, demonstrating crystallization consistent with prior reports. The overlay shows CuBTTri powder in its hydrated form. (D) SEM imaging of TSPCU films showing surface morphology from SNAP and CuBTTri incorporation. Overlays show MOF aggregates embedded within the composite films.

3.1.1. Scanning Electron Microscopy (SEM) and Energy-Dispersive X-ray Spectroscopy (EDS).

Further SEM analysis of the prepared TSPCU film formulations showed surface homogeneity across all four sample types, with only a slight increase in surface imperfections on TSPCU–NO films relative to TSPCU control films (Figure 1D), potentially caused by remnant surface-bound SNAP crystals during the initial casting step (Scheme 1). CuBTTri crystal aggregates were shown in both the TSPCU–MOF and TSPCU–NO–MOF films, demonstrating the maintenance of crystalline structure after incorporation into the polymeric scaffold (Figure 1D). Cross-sectional SEM imaging of films (Figure S2) demonstrated the successful lamination of the TPSCU top layers and supported the presence of MOF crystalline structures enveloped within the films. Furthermore, EDS mapping of the TSPCU–NO and TSPCU–NO–MOF (5%) cross sections demonstrated the crystallization of SNAP in the inner TSPCU layer from sulfur mapping (Figure S2E,K), as corroborated by prior studies.⁴⁰ The heterogeneous state of sulfur corresponds to the crystallization of SNAP within the films, as previous work with SNAP–TPSCU materials has demonstrated that the low water uptake of the polymer leads to slow solubilization of the crystallized SNAP within the film nearest to the polymer/solution interface.⁴⁰ Copper mapping showed the highest signal intensity (Figure S4L) in thin stretches corresponding to the first dip-coated layer in TSPCU–NO–MOF (5%). Prior studies have also shown increased percentages of MOF resulting in more crystalline coverage of polymer surfaces (Figure 1D). The difficulty of detecting copper via EDS analysis may be explained by the low atomic ratio of Cu with respect to other components of CuBTTri and the limitations of surface analysis with EDS. SEM imaging of the top surfaces of the composite corroborates the dispersion of CuBTTri within the intermediary layer (Figure 1D); however, the overtaking surface coverage may affect the mechanical properties of the composite as well as NO catalytic rates.^33,36

3.1.2. Tensile Testing.

Ultimate tensile strength (UTS) was evaluated for all wt % of CuBTTri with SNAP (TSPCU–NO–MOF) and without (TSPCU control). Control samples were consistently about 5 N/mm² stronger than the SNAP-blended samples for 0, 1, and 3 wt % CuBTTri films (P < 0.01). This relationship was observed with both trendlines in Figure S3 running parallel to each other until the final data point at 5 wt % CuBTTri. UTS for the two sample classes converged at a 5 wt % CuBTTri incorporation, implying that 5 wt % was both the weakest control sample type and the strongest SNAP sample type. Similar results in UTS reduction have been observed before in other SNAP-doped thermoplastic polymers, further demonstrating the importance of controlling the weight percent of SNAP and MOF in the composites to maintain the baseline mechanical properties of the polymer.⁴⁶ The results confirmed that the addition of SNAP and MOF, separately or together, does not significantly impact the mechanical integrity of the samples (i.e., less than 20% change in UTS at their breakpoint).

3.1.3. Static Contact Angle Measurement.

Once the SNAP-doped TSPCU base layers had been double dip-coated and dried, the surface wettability of the composites was tested to study the effects of SNAP and CuBTTri incorporation. TSPCU films displayed a slight hydrophilic contact angle of 89.30 ± 1.17°. Surface wettability was not significantly affected by the introduction of SNAP on the TSPCU–NO films (89.54 ± 3.69°). SNAP impregnation of polymer films has been shown to affect surface wettability, often by the presence of microcrystalline structures on the surface of films.⁴⁴ However, the introduction of two TSPCU dip-coated layers onto the SNAP-doped base layer was shown to negate this effect, with no statistically significant difference in surface wettability found between TSPCU controls and TSPCU–NO (Figure S4).

The contact angle of TSPCU–MOF (3%) (100.50 ± 6.64°) was not significantly affected by the addition of SNAP, as shown in the TSPCU–NO–MOF (3%) films (97.76 ± 1.20°) (Figure S4). TSPCU–MOF (3%) showed a significant increase in hydrophobicity (P < 0.0001) when compared to TSPCU and TSPCU–NO. Likewise, TSPCU–NO–MOF (3%) was significantly more hydrophobic (P < 0.001) than both TSPCU and TSPCU–NO. The introduction of MOF into the materials brings heterocyclic groups to the surface of the polymer.⁴⁷ Heterocyclic hydrocarbons are known to affect the surface hydrophobicity of polymers, which supports observed results based on the dip-coating strategy with CuBTTri.

3.2. SNAP and Copper Ion Leaching.

In developing biomaterials that intrinsically prevent bacterial contamination via the active surface release of NO, it is beneficial that these surfaces do not elicit secondary biocidal effects from elution of the incorporated SNAP and CuBTTri. Leached SNAP may elicit a secondary biocidal effect through the systemic accumulation of its decomposition products (Figure 2A), while CuBTTri may elicit secondary biocidal effects through its deterioration and subsequent metal ion accumulation. In the biomaterials design space, keeping material leaching below thresholds to prevent secondary biocidal action or induction of a cytotoxic effect is critical for the feasibility of translation. To this end, SNAP leaching from the TSPCU formulations was first assessed in a 24 h study under physiological conditions. Based on UV–vis spectroscopy (Figure S5A,B), the total SNAP content leached over 24 h was determined with respect to the total amount of SNAP loaded as well as the mass of the composites (Figure S5C,D). Similar to our prior reported findings, less than 10% by moles of the total loaded SNAP was lost in each formulation after 24 h, with no statistically significant difference found at any time point between formulations.²⁰ In parallel, the cumulative SNAP leaching with respect to the mass of the composite was found similar to previously reported findings, with no statistically significant difference found between formulations at each time point.⁴⁰ These findings confirm that the presence of CuBTTri does not affect the diffusion of SNAP from the polymer matrix into the surrounding environment and further clarifies that the CuBTTri facilitated intrapolymeric decomposition of SNAP and subsequent diffusion of NO from the polymer as the main source of NO release.

Figure 2.

Overview of physiological NO release. (A) NO is produced by the degradation of SNAP in the presence of heat, light, and metal ions. Herein, exterior coordination sites on CuBTTri are thought to catalyze the homolytic cleavage of the S-nitrosothiol bond. (B) Proposed mechanism for elevated NO release in composite TSPCU films with CuBTTri based on solubilization of crystallized SNAP in the polymer proximal to the polymer/solution interface, leading to intrapolymeric diffusion and interaction between SNAP and CuBTTri that facilitates NO formation and release. (C) Representative initial NO release profiles of TSPCU–NO–MOF films under physiological conditions. (D) Average surface fluxes of NO from TSPCU–NO–MOF film formulations after an initial 4 h under physiological conditions. Data presented as mean ± SD (N = 3). (E) Average surface fluxes of NO from TSPCU–NO–MOF films over a 7 days incubation study under physiological conditions. Data presented as mean ± SD (N = 3). Statistical significance presented as * (P ≤ 0.05), *** (P ≤ 0.001), and **** (P ≤ 0.0001).

Knowing Cu^2+/1+ as a potent biocidal agent, the TSPCU composites were further investigated for Cu^2+/1+ leaching to evaluate if secondary biocidal effects from these ions were feasible. Based on ICP-AES studies, less than 38 ppb Cu was generated in solution when incubated for 24 h under physiological conditions mirroring the NOA studies. This corresponded to a less than 0.1% leaching of Cu with respect to the mass of CuBTTri incorporated in both TSPCU–MOF (3%) and TSPCU–NO–MOF (3%) composites. These results further corroborate the SNAP leachate findings, supporting NO released from the polymer as the main biocidal agent.

3.3. NO Release under Physiological Conditions.

Reduced copper has been shown to act as a catalyst in addition to heat and light toward the decomposition of S-nitrosothiol bonds, leading to the formation of NO (Figure 2A). NO-releasing copper nanoparticle composite films have been investigated as coating materials on medical-grade polyvinyl chloride for application in extracorporeal circulation for the prevention of bacterial infection.^15,16 Using SNAP as the NO donor, previous TSPCU film formulations have been able to achieve up to 12 × 10⁻¹⁰ mol cm⁻² min⁻¹ in the initial release, whereas GSNO formulations have achieved around 7 × 10⁻¹⁰ mol cm⁻² min⁻¹ in the initial 4 h of release.^15,16 In the present study, the incorporation of SNAP and CuBTTri into sandwiched composite layers (Figure 2B) enabled tailoring of the NO release profiles from TSPCU films based on the weight percent CuBTTri incorporation. Representative NO release profiles during an initial 1 h of incubation under physiological conditions are shown in Figure 2C. During the 1 h studies, the maximum instantaneous NO surface flux for each formulation was found to be dependent on weight percent CuBTTri incorporation, with 5 wt % producing the highest instantaneous flux at an average of 88 ± 15 × 10⁻¹⁰ mol cm⁻² min⁻¹ (Figure S6A). This trend was mirrored in calculations of the total NO released during the initial 1 h incubation, with each film formulation producing nanomolar quantities of NO normalized to the surface area (Figure S6B). In each case, higher CuBTTri incorporation yielded greater cumulative NO release, providing further evidence of CuBTTri mass fraction-dependent catalytic activity within the composites.

An extended study to 4 h of incubation further confirmed these observations, with each formulation producing a statistically significant increase in the average NO surface flux with each step increase in CuBTTri weight percentage incorporation (Figure 2D). Being able to achieve a higher NO surface flux is beneficial in an antimicrobial context, as low-dose treatments with NO have been shown to enhance the resistance of many pathogens to further NO treatment, but high burst release profiles such as NO–MOF (5%) in Figure 2C have also been associated with enhanced cytotoxic effect.⁴⁶

Prior work with SNAP-blended TSPCU composites has demonstrated their shelf-stability with greater than 88% retention of the initial SNAP loading after 8 months of dry storage at 37 °C with protection from light.⁴⁰ These studies and our corroborating work with NO-releasing polymers have shown that SNAP retains remarkable stability in low water uptake polymers provided the materials are protected from heat, light, and moisture.⁴⁶ The presence of CuBTTri is not expected to substantially affect SNAP retention during dry storage, as the driving force for NO release is hypothesized to be the dissolution of crystalline SNAP within the polymer from moisture permeation, leading to further homolytic cleavage of the S-nitrosothiol bond in SNAP and the diffusion of NO from the polymer.^40,48 In this sense, CuBTTri facilitates the decomposition of the dissolved SNAP, providing an elevated threshold for NO release during the initial placement of the films under physiological conditions. However, prior work with copper-catalyzed NO release has only considered release profiles with up to 24 h of incubation under physiological conditions, warranting further investigation of the release kinetics over longer periods.^15,16

3.3.1. Long-Term NO Release Studies (7 days).

Realizing the importance of long-term, sustained NO surface flux from the TSPCU composites for medical device coating application necessitated the further study of the films over a 7 days study under physiological conditions. As shown in Figure 2E, all formulations sustained NO release greater than the physiologically estimated minimum for endothelial cells (i.e., 0.5 × 10⁻¹⁰ mol cm⁻² min⁻¹).¹⁴ Whereas other previously reported Cu nanoparticle-based composites have demonstrated NO release for only up to 24 h, our studies herein demonstrate the sustained NO release over the 7 days study, with release after 7 days on par with our prior work with selenium composites.^15,16,20

Whereas the 5 wt % CuBTTri formulation showed the highest NO surface flux on the first day of the study (i.e., 4 h average), all CuBTTri films stabilized to a surface flux of approximately 3–4 × 10⁻¹⁰ mol cm⁻² min⁻¹ after 24 h, with no statistically significant difference in the NO surface flux (P > 0.05) found between formulations after 24 h incubation (Figure 2E). These results signify that CuBTTri incorporation elicits its main catalytic effect during the initial 24 h incubation period, with NO flux eventually becoming limited as some function or a combination thereof of intrapolymeric substrate transport limitation, CuBTTri catalytic exhaustion, or substrate consumption. Given that prior studies demonstrated NO release from TSPCU–SNAP blends for over 20 days,⁴⁰ the authors reason the stabilization of NO release after 3 days to be the consequence of the heterogeneous matrix formation within the TSPCU–SNAP layer. Prior studies have demonstrated that a heterogeneous matrix of crystalline SNAP forms during the solvent evaporation process, resulting in a portion of the SNAP crystallizing inside the polymer matrix rather than just dissolving in it.^40,49 This cited work substantiating the formation of crystalline SNAP further suggested that the solubilized SNAP within the polymer matrix is susceptible to degradation as a result of increased availability from moisture permeation under incubation at physiological conditions.⁴⁰ Based on these prior findings, we hypothesize that the drop in NO release by 3 days is the result of the consumption of dissolved SNAP in the polymer, leaving most of the SNAP reservoir intact in the heterogeneous crystalline state, which is subject to decomposition over a time course dependent on water permeability, incubation temperature, and diffusion of metal ion catalysts within the composite. Regardless, the initial elevation in NO release via catalytic effects with CuBTTri promotes the potential for TSPCU–NO films for eradicating bacterial infection via higher, initial NO release, provided these surface fluxes do not elicit cytotoxic effects.

3.4. In Vitro Cytocompatibility Evaluation.

Ensuring that the TSPCU composites do not lead to the induction of a cytotoxic effect is a critical benchmark to pass before further consideration as a medical device coating. To this end, TSPCU composites of SNAP and CuBTTri were evaluated in a 24 h leachate toxicity test followed ISO 109993-5 standards for the biological evaluation of medical devices. Two human-derived cell lines were used in these studies—BJ fibroblast and HUVEC. As shown in Figure 3A, cytotoxic responses (i.e., below 80% relative viability of test group relative to control group) were observed only for the TSPCU–NO–MOF (5%) formulation with BJ fibroblasts. This same trend was observed with HUVEC (Figure 3B), implicating that the TSPCU–NO–MOF (5%) must pass a critical threshold for the induction of cytotoxicity.

Figure 3.

Cytocompatibility evaluation of TSPCU–NO–MOF films over a 24 h leachate study challenged against (A) human fibroblasts (BJ) and (B) vascular endothelial cells (HUVEC). Data presented as mean ± SD (N = 3) from independent passages. Statistical significance presented as * (P ≤ 0.05), ** (P ≤ 0.01), *** (P ≤ 0.001), and **** (P ≤ 0.0001).

Previous studies have demonstrated the biocompatibility of CuBTTri-doped PVC films with over 22 days of coincubation.³² In the same light, our studies have demonstrated no appreciable difference in SNAP leaching in each TSPCU formulation dependent on CuBTTri incorporation (Figure S5). Our SNAP leaching studies demonstrated less than 2.5% (wt SNAP/wt TSPCU composite) leaching, corresponding to SNAP accumulation of less than 250 μM after 24 h incubation under physiological conditions. Challenging both cell lines against SNAP, we found that similar quantities did not induce any major cytotoxic response (Figure S7).

Prior studies have also demonstrated stoichiometrically 10-fold lower quantities of NAP and NAP disulfide leaching from 10 wt % SNAP-loaded TSPCU composites fabricated in a manner similar to the TSPCU–NO films prepared herein, further discounting the role of leaching toward the induction of cytotoxic effect.⁴⁰ Based on our NOA results, an almost 2× increase in the stabilized NO surface flux was obtained from TSPCU–NO–MOF (5%) compared to the 3 wt % formulations (Figure 2D), corresponding to more than 2× cumulative NO released over the initial hour of incubation (Figure S6B). Based on these results, we postulate the cytotoxic effects observed with TSPCU–NO–MOF (5%) to possibly be the result of the rapid accumulation of NO in solution leading to peroxynitrite formation (ONOO⁻), which is known to induce cytotoxic effects in mammalian cells.⁵⁰ In addition, serum-containing media may act as a sink for the released NO, possibly through nitrosation of serum proteins during the 24 h incubation process, leading to delayed release and the formation of other reactive nitrogen species such as nitrite $({\text{NO}}_2^{-})$ and nitrous acid (HNO₂).⁵¹ With the elevated NO release observed in TSPCU–NO–MOF (5%) with respect to the 1 and 3 wt % formulations, a proportionate induction of cytotoxic effect is expected.

Upon further investigation, it was also observed that TSPCU–MOF (3 and 5%) composites elicited a statistically significant decrease (P < 0.001) in the viability of BJ fibroblasts with respect to TSPCU alone, suggesting some potential for CuBTTri leaching triggering cytotoxic effects. However, this same trend was not mirrored with HUVEC. Given the cytotoxic response triggered by TSPCU–NO–MOF (5%), further biological studies considered the 3 wt % formulations. As shown in Figure S8, exposure to leachates from TSPCU–NO–MOF (3%) leads to no appreciable change in the cellular morphologies of either BJ fibroblasts or HUVEC compared to controls, further supporting the use of this formulation for subsequent studies.

3.5. In Vitro Evaluation of Bacterial Inhibition.

Bacterial surface contamination is the culprit behind many hospital-acquired infections (HAIs). In the United States alone, estimates place more than 1.7 million patients as being affected annually by a HAI.⁴ Often, these infections lead to medical device failure, requiring the removal of the device and leading to increased patient suffering and health-care-related costs. NO is a potential alternative to traditional biocidal agents and demonstrates an ability to prevent medical device infections via incorporation of donor compounds into medical-grade polymers via blending, solvent casting, or covalent immobilization methods.^8-10 Previous research has demonstrated the ability of NOrel materials to prevent the adhesion of bacteria, kill planktonic bacteria, and eradicate biofilms by several different mechanisms including DNA cleavage, lipid peroxidation, and nitrosative and oxidative stress.^20,52-56 Furthermore, MOFs can be integrated as a part of the polymer to stabilize Cu^2+/1+ ions via coordination.³² In this case, water-stable CuBTTri serves to enhance NO release to aid bacterial killing efficiency.

To assess the antibacterial efficacy of the synthesized materials, in vitro antibacterial assays (Figure 4) were conducted to quantify the number of viable adhered bacteria (Tables S1 and S2) and planktonic bacteria (Tables S3 and S4) after 24 h of exposure. Isolated strains of MRSA and E. coli were suspended in media-subsidized PBS and exposed to the fabricated composites for 24 h. Bacteria were provided with optimum supplemented media to keep the bacteria out of survival mode and allow for active adherence to the material surface.

Figure 4.

Antibacterial studies of TSPCU films challenged against clinically relevant pathogens. Adhered studies of (A) MRSA and (B) E. coli from 24 h incubation of films, normalized to the exposed surface area. Data are shown as mean ± SD (N = 4). Planktonic viability studies of (C) MRSA and (D) E. coli from 24 h incubation with films, normalized per volume of media. Data are shown as mean ± SD (N = 4). Statistical significance presented as ns (not significant), * (P ≤ 0.05), ** (P ≤ 0.01), *** (P ≤ 0.001), and **** (P ≤ 0.0001).

As shown in Figure 4, SNAP-containing materials (TSPCU–NO and TSPCU–NO–MOF) significantly reduced the viability of adhered and planktonic bacteria compared to TSPCU control. When comparing SNAP-containing materials with and without CuBTTri, the presence of CuBTTri helped to promote the stable SNAP release, which is essential for antibacterial efficiency. When compared to the TSPCU unmodified material, the TSPCU–NO material demonstrated a 1.88- and 0.98-log reduction, while the TSPCU–NO–MOF material demonstrated a 2.74- and 1.23-log reduction in adhered MRSA and E. coli, respectively. Similarly, when compared to the enumerated 10⁸ inoculums, the TSPCU–NO material demonstrated a 1.03- and 0.54-log reduction, while the TSPCU–NO–MOF material demonstrated a 1.33- and 0.90-log reduction in planktonic MRSA and E. coli, respectively. These findings showcased how the combination of SNAP and the CuBTTri MOF leads to the lowest number of both viable adhered and planktonic bacteria under physiological conditions, indicating that TSPCU–NO–MOF has the best likelihood of reducing the presence of bacterial biofilm growth on a polymeric medical device surface. The combination of low weight percentages of MOF and the nitric oxide donor SNAP is a novel method for increasing antibacterial efficacy without cytotoxic effects.

4. CONCLUSIONS

The incorporation of MOFs into NO-releasing substrates is a strategic design choice for developing an initial, elevated NO surface flux for enhanced antimicrobial activity without increased susceptibility for biocidal agent leaching. The CuBTTri-dependent nature of the surface properties of TSPCU composites shown herein demonstrates the versatility of the platform and its ability to be tailored to a variety of biomedical applications, such as urinary catheters, blood-contacting surfaces, and other indwelling medical devices. Incorporation of CuBTTri into TSPCU composites was shown to afford surfaces with a stabilized distribution of MOF aggregates, enabling enhanced NO release during the first 24 h of incubation under physiological conditions—an often-critical threshold for early infection prevention. Manipulating the CuBTTri weight percent incorporation was found to not significantly affect SNAP leaching, nor was CuBTTri found to produce any significant quantities of Cu leaching. These results pose MOF-based composites to be an attractive option over prior metal nanoparticle composite work, demonstrating the stability of the Cu catalytic centers over time.

In vitro testing of TSPCU–NO–MOF composites demonstrated many attractive properties for biomedical applications. First, CuBTTri enabled the controlled NO release profiles during the initial 24 h of incubation under physiological conditions, which correlated to dose-dependent responses in cytotoxicity in 24 h studies of exposure with BJ fibroblasts and HUVECs. Further evaluation under a 24 h bacterial viability study demonstrated that the cytocompatible TSPCU–NO–MOF (3%) formulation was able to elicit over 99 and 95% reductions in adhered and planktonic MRSAs, respectively. Similar results were also found for E. coli, demonstrating the value of enhanced NO surface release during the initial exposure time of composites such as TSPCU–NO–MOF for application in medical devices. Reducing early-stage bacterial microadherence to medical device surfaces is critical to the prevention of biofilm formation. In all, the ease of fabrication and simplicity of surface design presented herein with TSPCU–NO–MOF films makes the functionalized layering material design an attractive option for a variety of medical device applications and warrants future study with other MOFs and further surface modifications.