Nitric Oxide Release and Antibacterial Efficacy Analyses of S-Nitroso-N-Acetyl-Penicillamine Conjugated to Titanium Dioxide Nanoparticles

Abstract

Therapeutic agents can be linked to nanoparticles to fortify their selectivity and targeted delivery while impeding systemic toxicity and efficacy loss. Titanium dioxide nanoparticles (TiNPs) owe their rise in biomedical sciences to their versatile applicability, although the lack of inherent antibacterial properties limits its application and necessitates the addition of bactericidal agents along with TiNPs. Structural modifications can improve TiNP’s antibacterial impact. The antibacterial efficacy of nitric oxide (NO) against a broad spectrum of bacterial strains is well established. For the first time, S-nitroso-N-acetylpenicillamine (SNAP), an NO donor molecule, was covalently immobilized on TiNPs to form the NO-releasing TiNP–SNAP nanoparticles. The TiNPs were silanized with 3-aminopropyl triethoxysilane, and N-acetyl-d-penicillamine was grafted to them via an amide bond. The nitrosation was carried out by t-butyl nitrite to conjugate the NO-rich SNAP moiety to the surface. The total NO immobilization was measured to be 127.55 ± 4.68 nmol mg⁻¹ using the gold standard chemiluminescence NO analyzer. The NO payload can be released from the TiNP–SNAP under physiological conditions for up to 20 h. The TiNP–SNAP exhibited a concentration-dependent antimicrobial efficiency. At 5 mg mL⁻¹, more than 99.99 and 99.70% reduction in viable Gram-positive Staphylococcus aureus and Gram-negative Escherichia coli bacteria, respectively, were observed. No significant cytotoxicity was observed against 3T3 mouse fibroblast cells at all the test concentrations determined by the CCK-8 assay. TiNP–SNAP is a promising and versatile nanoparticle that can significantly impact the usage of TiNPs in a wide variety of applications, such as biomaterial coatings, tissue engineering scaffolds, or wound dressings.

1. INTRODUCTION

After the Nobel prize-winning discoveries of nitric oxide’s (NO’s) physiological role in the human body as a key gaseous signaling molecule, a colossal cascade of research sheds light on further NO biological potentials.^1–6 It is well reported that NO possesses versatile biological functions such as platelet activation and adhesion inhibition, antithrombosis, bactericidal actions, biofilm dispersion and prevention, endothelialization induction, signaling action in the immune responses, and angiogenesis promotion.⁷ This bioregulatory molecule is synthesized endogenously in the human body by a family of enzymes known as NO synthase (NOS), which catalyze NO production from l-arginine.⁸ Other than vasodilation and antithrombotic effects of NO, which are extensively studied,^9–11 the antibacterial activity of this highly reactive molecule promotes promising results against a broad spectrum of bacteria, especially antibiotic-resistant strains, that is, first, the production of peroxynitrite (OONO⁻) by reacting with superoxide (O₂⁻) which can induce lethal oxidative stress, and second, the production of S-nitrosothiols (RSNO). In the second mechanism, NO initially is oxidized to N₂O₃ (nitrous anhydride). Furthermore, it reacts with sulfhydryl groups on cysteine residues of membrane proteins to produce RSNO structures which will be lethal to bacterial cells by altering the protein functionality.^7,12,13

All the remarkable advantages of NO molecules denote promising results if an exogenous delivery system could be exploited to provide therapeutic amounts of NO to desired locations. Nevertheless, there are challenges in NO delivery systems. Being a highly unstable and reactive radical that possesses a high affinity in reacting with oxygen molecules in the surrounding environment and has a half-life of only a few seconds under physiological conditions are the challenges in using NO under physiological conditions.⁷ Hence, delivering NO to the desired sites, such as in vivo or from blood-contacting medical devices, within a prolonged time and a therapeutic capacity has been a challenging source of research. To this end, several types of NO donor molecules such as S-nitrosothiols (RSNOs), N-diazeniumdiolates (NONOates), metal NO complexes, organic nitrates, and organic nitrites have been used to store and deliver NO under physiological conditions. Within a wide range of NO donor molecules, a tremendous amount of attention has been given to RSNOs, such as S-nitrosoglutathione (GSNO), S-nitroso-N-acetyl-l-cysteine (SNAC), and S-nitroso-N-acetylpenicillamine (SNAP), owing to their biocompatibility, antibacterial and antithrombotic properties, and their relative stability.^14–18

Incorporating NO donor molecules into medical devices and biomedical substrates have been reported to increase their hemocompatibility, antibacterial efficacy, and thromboresistance.^7,19,20 There are several reports about techniques to incorporate RSNOs in polymeric biomaterials such as simple physical blending, solvent swelling, and covalently conjugating.^9,21–27 The covalent conjugation of NO donors is considered one of the most promising techniques since it regulates the undesired leaching and premature loss of NO donors, eventually increasing the biomaterial applicability.²⁸ Another approach to achieving sustained NO release is by conjugating NO donors to nanoparticles. Frost and Meyerhoff introduced covalent attachment of the SNAP moiety to fumed silica nanoparticles that can be used as a filler within a polymeric matrix to prevent leaching and loss of NO donors.²⁹ NO donor molecules have also been immobilized on diatomaceous silica earth nanoparticles,³⁰ porous and non-porous silica nanoparticles,³¹ and halloysite nanotubes.³² However, initial burst release, low payload, a short period of NO release, and imposed cytotoxicity left the quest to further investigate other nanomaterials for NO donor conjugation that can provide longer NO release and improved yields and conversion rates while maintaining biocompatibility and low cytotoxicity.

The propitious usage of nanoparticles in medicine has been on a rise for the past decades as it can be rewarding under many circumstances. Not only do they possess beneficial applications solely due to their unique properties, but also linking a therapeutic agent to nanoparticles can enhance their selectivity and targeted delivery, induce multifunctionality, with moderate side effects, amplify therapeutic efficacy, and impede systemic cytotoxicity.^33,34 Among all the nanoparticles, titanium dioxide (TiO₂) nanoparticles (TiNPs) have been given prominent attention for the past decades owing to their unique properties such as chemical stability, a hydroxyl-rich surface, biocompatibility, non-toxicity to mammalian cells, hypoallergenic and photodynamic properties, a high refractive index, and a relevantly low cost.³⁵ TiO₂ nanoparticles (TiNPs) have been used extensively in anticancer therapy,^36,37 photo-activated antibacterial surface coatings,^38–40 biosensors,⁴¹ cosmetics,⁴² food and drug colorants,³⁷ water treatment technologies,^43,44 anticorrosion applications,⁴⁵ and white paint production.^46,47 Furthermore, the antibacterial capacity of TiNPs under UV irradiation techniques has been extensively studied and utilized for years, although there are concerns for the utilization of UV irradiation as it can induce adverse effects such as cell death and host tissue cell necrosis.^37,42 Therefore, extensive research is being done to improve TiNP antibacterial efficacy by embedding other elements to the TiNP structure to obtain higher bactericidal efficacy and facilitate or even obviate light activation of the particles.^35,37,48

In this study, TiNP is used as a novel NO delivery substrate for the first time with a chemical immobilization of a NO donor molecule to the nanoparticles. The hydroxyl-rich surface of TiNP provided the opportunity for silane functionalization of the nanoparticles using an amino silane agent, (3-aminopropyl)triethoxysilane (APTES), and further, free amine functional groups enabled covalent conjugation of N-acetyl-d-penicillamine (NAP) thiolactone as the precursor of the NO donor molecule SNAP. Finally, nanoparticles equipped with thiol functionality underwent nitrosation reaction using tert-butyl nitrite to generate SNAP conjugated TiNPs (TiNP–SNAP). Fourier transform infrared spectroscopy (FTIR) and nuclear magnetic resonance (NMR) were used to evaluate the quality of surface modifications. Ninhydrin and Ellman’s assays were utilized to measure the quantity of amine and thiol functional groups embedded on the nanoparticles, and chemiluminescence was used to determine the NO payload and release behavior. Furthermore, in vitro evaluations were conducted to evaluate the cytocompatibility of the TiNP–SNAP nanoparticles and their antimicrobial efficacy against Gram-positive Staphylococcus aureus and Gram-negative Escherichia coli bacteria strains.

2. MATERIALS AND METHODS

2.1. Materials.

TiNP (nanopowder 99.5%, 10–30 nm anatase and rutile mixture) was purchased from SkySpring Nanomaterials Inc. (Houston, USA). NAP, APTES, cyclam, potassium bromide (KBr), anhydrous toluene, l-cysteine hydrochloride monohydrate (≥98.0% purity), ninhydrin reagent (2% solution), Ellman’s reagent [5,50-dithiobis(2-nitrobenzoic acid), DTNB] (≥98.0% purity), glycine hydrochloride (≥99.0% purity), Luria–Bertani (LB) broth and LB broth with agar media powders, ethylenediaminetetraacetic acid (EDTA), sterile phosphate-buffered saline (PBS) powder 0.01 M, pH 7.4, containing 138 mM NaCl, and 2.7 mM KCl, were purchased from Sigma-Aldrich (St. Louis, MO). Absolute ethanol (≥99.8% purity), anhydrous pyridine, acetic anhydride, chloroform, Acros Organics copper(II) chloride dihydrate (≥99% purity), hexane, hydrochloric acid (technical grade), isopropanol, acetone, and tert-butyl nitrite were purchased from Fisher Scientific (Hampton, NH). Sodium acetate (NaAc) was obtained from VWR International (West Chester, PA). The bacteria strains S. aureus [American Type Culture Collection (ATCC 6538)] and E. coli (ATCC 25922) and the 3T3 mouse fibroblast cells (ATCC 1658) for cell compatibility experiments were purchased from ATCC. Dulbecco’s modified Eagle’s medium (DMEM) and trypsin–EDTA were purchased from Corning (Manassas, VA). The antibiotic mixture of penicillin–streptomycin and fetal bovine serum (FBS) was obtained from Gibco-Life Technologies (Grand Island, NY). Colorimetric assay cell counting kit-8 (CCK-8) with WST-8 monosodium salt was obtained from Sigma-Aldrich (St. Louis, MO). Deionized water (DI water) was used to prepare all aqueous solutions. PBS 0.01 M containing 100 μM EDTA was used in the material characterizations and NO analysis experiments unless noted otherwise.

2.2. TiNP Surface Silanization.

The surface silanization of nanoparticles was carried out following the previously reported protocol with slight modifications.⁴⁵ Briefly, TiNP (3 g) was dispersed in an aqueous hydrochloric acid solution (pH 2) using an ultrasonic water bath for 1 h at room temperature (RT). Further, the nanoparticles were washed with DI water, followed by centrifugation (Allegra X-30 Series Benchtop Centrifuge, Beckman Coulter, Brea, CA) at 12,000 rpm for 10 min (three times), and two times with isopropanol and one time with acetone to remove impurities, such as chloride ions, and absorbed moisture. Finally, the particles were centrifuged and dried overnight in an oven at 120 °C. The purified TiNPs were then dispersed in dry toluene (100 mL) using a sonication bath for 30 min in a nitrogen atmosphere. For surface silanization, 3 mL of APTES was added to the TiNP dispersion and refluxed at 120 °C for 12 h under a nitrogen atmosphere with constant stirring to obtain TiNP–APTES. The particles were centrifuged at 12,000 rpm for 10 min, washed with dry toluene several times to remove any unreacted APTES, and dried overnight in the oven at 100 °C.

2.3. SNAP Functionalization of TiNP–APTES.

2.3.1. NAP Thiolactone Synthesis.

NAP–thiolactone was synthesized by following the reported procedure.⁴⁹ Briefly, 5 g of NAP was dissolved in 20 mL of pyridine in a round-bottom flask and chilled in an ice bath for 30 min. Ice-cold acetic anhydride (10 mL) was slowly added to the reaction mixture and stirred for 30 min. Afterward, the reaction mixture was removed from the ice bath and stirred at RT for 20 h. Following the completion of the reaction, pyridine was removed from the reaction mixture using a rotary evaporator (IKA RV 10 auto pro V–C) to obtain an orange crude residue. The crude residue was dissolved in 20 mL of chloroform, washed, and rinsed three times with 20 mL of 1 M HCl. The organic layer was dried over anhydrous sodium sulfate and filtered. The chloroform was removed using the rotary evaporator to obtain a crystalline solid, which was further washed with hexane to remove the colored impurity and dried to obtain light-yellow-colored pure NAP–thiolactone.

2.3.2. NAP Attachment to TiNP–APTES.

NAP attachment to dried TiNP–APTES was done with slight modifications of a protocol published by Frost et al⁵⁰ The silanized TiNP–APTES (1 g) was homogeneously dispersed in dry toluene (15 mL) using a sonication bath for 30 min, followed by the addition of NAP–thiolactone (500 mg). A tethering reaction took place within 24 h under vigorous stirring in the dark at RT. TiNP–NAP was obtained by centrifuging the mixture at 12,000 rpm for 10 min, and the obtained powder was washed with dry toluene three times before the drying step. The obtained TiNP–NAP after washing steps was left in a desiccator for 24 h at RT to dry.

2.3.3. Nitrosation and SNAP Functionalization.

The synthesized TiNP–SNAP, as the final product, was obtained using an established SNAP conjugation reaction protocol (nitrosation), with minor changes.⁵⁰ Briefly, tert-butyl nitrite first was chelated of any copper contamination by mixing an equal volume of it with 20 mM cyclam solution with vigorous shaking three times. Each time, the t-butyl nitrite layer was isolated from the mixture. Finally, the chelated and pure t-butyl nitrite was refrigerated at 4 °C in an amber vial to be protected against the light for further usage. Previously modified TiNP–NAP (500 mg) was homogeneously dispersed in 7.5 mL of dry toluene using 30 min of sonication. Cleaned t-butyl nitrite (500 μL) was added to the mixture, and the reaction mixture was stirred at RT for 1 h to obtain TiNP–SNAP. The product was washed with dry toluene several times and isolated from the mixture using centrifugation at 12,000 rpm for 10 min and kept in a desiccator for 30 h at RT to dry and kept at −20 °C for further usage.

2.4. Primary Amine Quantification.

To measure the surface silanization efficiency and quantify the number of amine groups introduced to the surface of TiNP, a ninhydrin assay was implemented. Briefly, TiNP–APTES was dispersed in 0.05% glacial acetic acid (0.2 mg/mL) and was mixed with ninhydrin reagent (1 mL) in an amber vial. The mixture was kept in a boiling water bath for 10 min. Thereafter, the mixture was left to cool down to RT, and 95% ethanol (5 mL) was added to the mixture to make the final volume of 8 mL. The mixtures were centrifuged at 12,000 rpm for 10 min to separate the nanoparticles from the reaction solution. The absorbance value of the reaction solutions at 570 nm was determined using a UV–vis spectrophotometer (Cary 60, Agilent Technologies, Santa Clara, CA). To quantify the amine content using the UV–vis absorbance values, a calibration curve of known amine content was made using glycine hydrochloride with a molar absorptivity of 7.36 × 10³ M⁻¹ cm⁻¹.

2.5. Thiol Content Quantification.

The number of thiol groups on the surface of TiNP–NAP was measured using Ellman’s assay.³⁰ Briefly, DTNB (5,50-dithio-bis-[2-nitrobenzoic acid]) solution containing 2 mM DTNB and 50 mM sodium acetate (NaAc) was prepared and the resulting 2 mM DTNB and 50 mM NaAc solution (50 μL) mixed with PBS (100 μL), DI water (800 μL), and the sample (50 μL). The absorption values at a wavelength of 412 nm were recorded using UV–vis spectroscopy. l-Cysteine with a molar absorptivity of 14.16 × 10³ M⁻¹ cm⁻¹ was used to generate a calibration curve of known thiol concentrations and was used to determine the thiol content of unknown samples.

2.6. FTIR and NMR Analysis.

The surface modification and further successful functionalization of TiNPs were analyzed by FTIR using a PerkinElmer Spectrum 3 FTIR spectrometer. The TiNP powders before and after modification were mixed with potassium bromide (KBr) with a ratio of 1:100 and formed into a disk for FTIR analysis. Each spectrum was collected over a wavenumber range of 500–4000 cm⁻¹ with a scan rate of 16 and a resolution of 8 cm⁻¹. A Bruker NEO 600 MHz spectrometer was used for solid-state cross-polarization magic angle spinning carbon-13 NMR (CP/MAS ¹³C NMR) analysis. Due to a low carbon count, the CP/MAS ¹³C NMR spectra for TiNP–APTES particles were obtained from the summation of 8 spectra at 256 scans at a spinning speed of 10 kHz. However, for TiNP–NAP, a well-resolved ¹³C NMR spectrum was observed from the single spectra at 256 scans at a spinning speed of 10 kHz.

2.7. Dynamic Light Scattering and Zeta Potential Measurements.

The zeta potential of the particles was estimated using a Zetasizer nano-ZS ZEN3600 (Malvern Instruments Ltd. USA) at different surface modification stages to evaluate their colloidal stability. Moreover, dynamic light scattering (DLS) analysis was also employed to evaluate the average particle size at different modification stages. Following the previously established protocol,⁴⁶ with slight modifications, TiNPs, TiNP–APTES, TiNP–NAP, and TiNP–SNAP were dispersed in DI water (0.05 mg mL⁻¹) using a sonication bath for 30 min. Before running the instrument, each suspension was vortexed for 2 min and left to settle for 1 min.

2.8. NO Loading Capacity and Release Kinetics of TiNP–SNAP Nanoparticles.

A Sievers chemiluminescence NO analyzer (NOA) 280i (Zysense, Weddington, NC, USA) was utilized to analyze the real-time NO release behavior in terms of NO loading capacity and release kinetics under physiological conditions. Samples were prepared on the day of the experiment according to the prior reported protocol with slight changes.^30,32 The synthesized TiNP–SNAP (1 mg) was homogeneously dispersed in 1 mL of 0.01 M PBS solution (pH 7.4) by 30 min of sonication (n = 3). The suspension further was transferred into a dialysis tubing with a molecular weight cutoff of 3.5–5 kDa secured at both ends to ease the operation process and facilitate the transfer of the nanoparticles to the sample cell of the instrument. Once the dialysis bags were submerged in PBS within the sample cell of NOA, 0.25 M copper(II) chloride (1 mL) and 0.5 M ascorbic acid (1 mL) were injected into the NOA cell, which led to catalytic decomposition of the S–NO bond on the particles and release of the total NO content of the covalently bound TiNP–SNAP. To obtain the total amount of loaded NO, the trapezoidal rule was used to calculate the area under the release curve.³²

The synthesized TiNP–SNAP (1 mg) was homogeneously dispersed in 1 mL of 0.01 M PBS solution (pH 7.4) with 100 μM EDTA by 30 min of sonication (n = 3), followed by transfer to the dialysis bags as described above. To evaluate the NO release kinetics under physiological conditions, the dialysis bags were submerged in 3 mL of 0.01 M PBS solution (pH 7.4) with 100 μM EDTA inside an amber NOA sample cell. The sample cell was incubated in a 37 °C water bath to maintain a physiological temperature and was supplied with pure nitrogen gas (N₂) at a constant flow of 200 mL min⁻¹. The cell pressure of the NOA cell was set to 9.8 Torr with an oxygen supply pressure of 6.6 psi. The NO gas emitted from TiNP–SNAP was purged by N₂ and continuously swept by vacuum to the chemiluminescence reaction chamber where the concentration of NO molecule was measured within 1 s time intervals. To evaluate the release kinetics over 20 h, data were collected for each sample at several time points (0, 2, 4, 6, 12, 14, 16, and 20 h) until the release profile plateaued at a constant value. Samples were incubated in an incubator at 37 °C in 3 mL of 0.01 M PBS (pH 7.4) incorporated with 100 μM EDTA inside an amber vial between measurements.

2.9. In Vitro Antibacterial Efficacy of TiNP–SNAP over 24 h.

To assess the antibacterial efficacy of the modified TiNP–SNAP, S. aureus and E. coli were used as the representative Gram-positive and Gram-negative bacteria, respectively. An established 24 h direct contact assay³² was used to identify the antibacterial behavior of TiNP–SNAP in comparison to control TiNPs dispersed in PBS (pH 7.4) with different concentrations (0.01, 0.1, and 5 mg mL⁻¹). An isolated colony of S. aureus or E. coli was inoculated in LB broth in a shaker incubator at 150 rpm at 37 °C. After 16 h, inoculums containing bacteria in their log phase of growth were centrifuged at 3500 rpm for 7 min, and the supernatants were decanted. The bacteria pellets were washed in sterilized PBS (pH 7.4). The obtained bacteria suspensions were centrifuged at 3500 rpm for 7 min again, and the pellets were resuspended in sterilized PBS using a vortex mixer. Sterile PBS was used to dilute the bacteria suspensions to obtain an approximate concentration of 10⁷ colony forming units (CFU) in 1 mL of PBS. The bacterial solution (100 μL) was mixed with 100 μL of different concentrations of TiNP–SNAP suspended in PBS (n = 4) in a 96-well plate. Bacteria-containing wells without nanoparticles were used as negative controls. The plates were incubated at 37 °C in a shaker incubator at 150 rpm for 24 h while protected from exposure to ambient light. After incubation, 50 μL of the solution in each well was plated on an LB agar plate using a spiral plater (EDDY JET 2 W, IUL Instruments, Cincinnati, OH, USA) and then incubated at 37 °C for 24 h prior to counting the CFUs. An automated colony counter unit (SphereFlash IUL Instruments, Cincinnati, OH, USA) was used to determine the concentration of CFUs (CFU mL⁻¹) on each plate and consequently calculate the viable bacteria reduction of different test groups. The % reduction in bacterial viability was calculated according to eq 1 below (where C = CFU mL⁻¹)

$$\text{bacteria}\text{viability}\text{reduction}(\%)=\frac{C_{\text{control}}-C_{\text{sample}}}{C_{\text{control}}}\times 100$$ (1)

2.10. In Vitro Cell Viability Studies.

Leachates from the particles (concentrations of 0.01–5 mg mL⁻¹) were prepared following a previously reported protocol.³⁰ Briefly, 1 mL of each particle suspension in DI water was transferred into a 5 cm dialysis bag (molecular weight cutoff = 3.5–5 kDa), secured at both the ends, and soaked in 10 mL of DMEM supplemented with 10% FBS and 1% penicillin–streptomycin for 24 h at 37 °C. After 24 h, the dialysis bags were removed, and leachates were used for further assessment. The cytotoxicity evaluation of TiNP–SNAP and control TiNPs was conducted as per the ISO 10993–5 in vitro cytotoxicity test. For this, NIH-3T3 mouse fibroblast cells at 5000 cells mL⁻¹ were seeded into a cell culture-treated 96-well plate and incubated in a humidified incubator at 37 °C, 5% CO₂ for 24 h. Leachates obtained from TiNP–SNAP and controls were exposed to cells in a well plate (10 μL) and incubated for another 24 h in an incubator at 37 °C. The cell viability test was performed using CCK-8 following the manufacturer’s protocol (Sigma, OH). CCK-8 dye (10 μL) was added to each well and incubated for 1 h, after which the absorbance of the cells was measured at a 450 nm wavelength using a microplate reader (Cytation 5 imaging multi-mode reader, BioTek). Data obtained from the study are presented as relative cell viability of TiNP–SNAP samples compared to the control cells in DMEM that received no treatment and is calculated using eq 2 below, where A = Absorbance

$$\text{Relative}\text{cell}\text{viability}=\frac{A_{\text{TiNP}}-A_{\text{TiNP}-\text{SNAP}}}{A_{\text{cells}\text{in}\text{DMEM}}}$$ (2)

2.11. Statistical Analysis.

All the data obtained in this study are reported as mean ± standard deviation (SD) with a sample size of 3 (n = 3) unless stated otherwise. To evaluate the statistically significant differences between values, a standard two-tailed t-test was used and is presented in the form of p-values.

3. RESULTS AND DISCUSSION

3.1. Chemical Surface Modifications.

The NO-releasing SNAP functionality was immobilized on the surface of the TiO₂ nanoparticles using thiolactone chemistry (Figure 1). First, APTES was chosen as the silanization agent due to its suitable reactivity and conformational properties that provide higher yields in surface modification reactions.³⁰ The inherent reactivity of APTES is due to its three ethoxy groups readily available to be anchored with the hydroxyl groups of the TiNPs via condensation reaction liberating ethanol as a byproduct. Thorough toluene washing steps removed the unreacted amino silanes and provided nanoparticles for further modification steps. The available amine-rich surface was further reacted with NAP–thiolactone, which is a self-protected derivative of penicillamine. The ring-opening reaction of NAP–thiolactone provided free thiol groups on the surface of the nanoparticles that were converted to the S-nitrosothiol SNAP functionality in the presence of the tert-butyl nitrite nitrosating agent.

Figure 1.

Synthesis route of TiNP–APTES, TiNP–NAP, and TiNP–SNAP. Acid-treated TiNPs were reacted with APTES using a reflux system under a nitrogen gas atmosphere at 120 °C for 12 h (a), followed by NAP thiolactone reaction at RT with vigorous stirring for 24 h (b). Finally, nitrosation reaction with t-butyl nitrite at RT for 1 h led to the production of SNAP-immobilized to TiNPs (c). Toluene was used as the solvent for all the reactions.

3.2. Primary Amine Quantification.

The silanization process was verified by quantifying the primary amine content on the TiNPs by performing the ninhydrin assay. Both qualitative and quantitative measurements of primary amine groups have been reported using a ninhydrin reagent.^32,51 In a mildly acidic environment and elevated temperatures (100 °C), within 10 min, the hydroxyl group in the ninhydrin reagent structure undergoes a nucleophilic displacement by the amine content on the surface of the nanoparticles, leading to the generation of a purple-colored (Ruhemann’s purple) complex, which exhibits absorbance at 570 nm.⁵² Generation of the free dye complex is advantageous since it can be separated from the particle mixture that might interfere with the UV–vis spectroscopy results. The results from the ninhydrin assay revealed that 0.24 ± 0.01 μmol mg⁻¹ (n = 3) of primary amine groups are conjugated to TiNPs after APTES reflux reaction at an elevated temperature of 120 °C and several washing steps to remove excess unreacted APTES from the mixture.

3.3. Thiol Content Quantification.

Free sulfhydryl groups (thiol groups) can be quantified through a well-reported spectrophotometric Ellman’s assay.⁵³ Ellman’s assay relies on free dye formation as the product of a chemical reaction between 5,5′-dithio-bis-(2-nitrobenzoic acid), known as DTNB or Ellman’s reagent, and thiol groups. The disulfide and dianion mixture of 2-nitro-5-thiobenzoic acid (TNB²⁻) in water is a yellow-colored dye that exhibits an absorbance at 412 nm in UV–vis spectroscopy. The absorbance values are directly proportional to the amount of thiol groups presented in the reaction.⁵⁴ Meanwhile, NAP–thiolactone, as the precursor of the NO donor moiety, has a protected sulfur bond, and the –SH is exposed only through a ring-opening conjugation reaction.^28,55 Therefore, after the NAP attachment to TiNP–APTES, Ellman’s assay can adequately estimate the reaction yield. The results revealed the quantity of the thiol group on the surface of TiNP–NAP to be 0.20 ± 0.01 μmol mg⁻¹, which results in a conversion rate of 83.33% of amines to thiol functionalities (Table 1). A wide range of conversion rates (~2–80%) is reported previously in the literature for covalent attachment of NAP–thiolactone to silanized _surfaces.^{30,32,50,56–58}

Table 1.

Ninhydrin and Ellman’s Assays Demonstrated the Content of Free Amine and Thiol Groups, Respectively, on the Surface of the Modified Nanoparticles

amine content (μmol/mg)	thiol content (μmol/mg)	conversion rate (%)
0.24 ± 0.01	0.2 ± 0.01	83.33

3.4. FTIR and NMR Analysis of the Functionalized TiO₂ Nanoparticles.

FTIR analysis was performed to evaluate the surface functionalization and chemical modifications of the TiNPs at each modification stage for the unmodified and surface-modified TiNPs (Figure 2). A broad peak in the range of 3450–3200 cm⁻¹ along with a peak around 1630 cm⁻¹ was observed for TiNP, which are attributed to the stretching vibration of hydroxyl groups chemically bound to the surface of the TiNP and deformation vibration of adsorbed water present on the TiNP surface, respectively. The broad peak for the Ti–O–Ti bond stretch was also observed between 850 and 600 cm⁻¹. Several new vibrational modes related to the organic moieties were observed after the surface functionalization of TiNP with APTES. The characteristic –CH₂ stretch and bending peak for the methylene group (–CH₂) of the APTES backbone were observed at 2929 and 1319 cm⁻¹, respectively. The relatively broad peaks at 3413 cm⁻¹ and a sharp peak at 1623 cm⁻¹ appeared due to –NH₂ stretch and bending, respectively. The other two characteristic absorption peaks for Si–O–Si (asymmetric) and Si–O–Si (symmetric) observed at 1118 and 1026 cm⁻¹, respectively, further confirmed the successful surface modification of TiNP with APTES, leading to an amine-rich titania surface.⁴⁵ The NAP–thiolactone reaction with TiNP–APTES was also evident with the two characteristic peak appearances in the FTIR spectra at 1756 and 1657 cm⁻¹ corresponding to –C=O stretch for the two different amide bonds. The broad –NH₂ stretch band observed at 3413 cm⁻¹ for TiNP–APTES was also shifted to 3246 cm⁻¹ after reacting with NAP–thiolactone. The peak at 1554 and 1292 cm⁻¹ can be assigned for the –N–H bending and –C–O stretch (ester), respectively. Further, the appearance of a new peak at 650 cm⁻¹ corresponding to the –SNO bending confirmed the successful transformation of TiNP–NAP to TiNP–SNAP via nitrosation reaction.^30,32

Figure 2.

Representative FTIR spectra of TiNP, TiNP–APTES, TiNP–NAP, and TiNP–SNAP.

The surface modification of the TiNPs was further characterized by CP/MAS ¹³C NMR to confirm the modification of TiNPs by APTES, followed by the subsequent attachment of NAP (Figure 3). In Figure 3a, three characteristic NMR signals were observed at 10.26, 22.84, and 42.99 ppm for the three propyl carbon atoms of the APTES moiety (marked as 1, 2, and 3). However, no NMR signal for the ethoxy carbon atoms of the APTES moiety was observed, which confirmed the attachment of APTES to the TiNP with the liberation of the ethoxy unit as ethanol.³² The ring-opening reaction between the amine moiety of APTES and NAP–thiolactone increased the carbon content attached to TiNPs, as evident by the NMR signal intensity for TiNP–NAP (Figure 3b). All characteristic carbon peaks were observed for the attached NAP moiety, which further corroborates the covalent linkage of NAP–thiolactone to TiNP–APTES.

Figure 3.

Representative ¹³C CP/MAS NMR spectra of (a) TiNP–APTES and (b) TiNP–NAP. (a) Three characteristic NMR signals were observed at 10.26, 22.84, and 42.99 ppm for the three propyl carbon atoms of the APTES moiety (marked as 1, 2, and 3). (b) Two characteristic carbonyl carbon atom peaks for the NAP thiolactone unit were observed at 171 and 179 ppm along with other characteristic carbon peaks.

3.5. DLS and Zeta Potential Analysis.

Nanoparticle modifications are usually accompanied by alterations in surface electrical charges and consequently impose changes to the particles’ colloidal stability. The TiNPs are known to possess a hydroxyl-rich surface that leads to a negative surface charge (zeta potential = −41.5 ± 1.75 mV).⁵⁹ Immediately after silanization, the modified TiNP–APTES surface became amine-rich (NH₂) and subsequently, an increase in the zeta potential (−5.1 ± 0.99 mV) was observed.⁴⁵ Further, the ring-opening reaction between amine and NAP–thiolactone made the TiNP–NAP surface thiol-rich (–SH), where an attenuation in zeta potential (−32.3 ± 2.8 mV) was observed.⁶⁰ Furthermore, after the conversion of free thiols to nitrosothiols, the reduction of thiol content also mitigated the surface charge and therefore an increase in zeta potential value (−10.7 ± 2.95 mV) was observed (Figure 4). Zeta potential measurements were conducted in DI water. These results are in agreement with previously reported zeta potentials for nanoparticle surface modifications.^45,61 The change in zeta potential value after each surface modification also plays a significant role in the dispersion behavior and the stability of the modified NPs. The increase in the zeta potential value of NPs increases the probability of aggregation, and therefore, an increase in hydrodynamic diameter of NPs is obvious. The DLS analysis for TiNPs, TiNP–APTES, TiNP–NAP, and TiNP–SNAP as measured in DI water revealed the average particle sizes to be 86.5 ± 2.8, 111.3 ± 5.9, 99.6 ± 13.3, and 126.6 ± 11.9 nm, respectively (Figure S1 in the Supporting Information document). The observed changes in hydrodynamic diameters for nanoparticles were in correlation with their respective zeta potential values. The zeta potential and DLS analysis further corroborate the successful surface modification of TiNP at each step.

Figure 4.

DLS (black bars) and zeta potential (red line) measurements of TiNPs at different stages of surface modification (n = 3).

3.6. Total NO Loading.

To evaluate the final reaction efficiency (i.e., the nitrosation of TiNP–NAP using t-butyl nitrite), chemiluminescence was utilized as the gold standard to evaluate the NO release.^62,63 After dispersing TiNP–SNAP in 1 mL of PBS, 1 mL of 0.25 M copper(II) chloride and 1 mL of 0.5 M ascorbic acid were injected into the NOA sample cell, which led to the catalytic release of NO from the –SNO moiety on the surface of the TiNP–SNAP nanoparticles.^50,64

The real-time NO release data collected from NOA were used to evaluate the total amount of NO immobilized on the TiNP–SNAP nanoparticles. The results revealed that TiNP–SNAP was capable of storing NO up to 127.55 ± 4.68 nmol mg⁻¹ (n = 3), which is higher than that of previously reported diatomaceous earth silica particles (31.2 nmol mg⁻¹). These differences can be attributed to the larger particle size of diatomaceous earth silica particles that provided a less overall surface area for reaction and SNAP conjugation.³⁰ The conversion rate from TiNP–NAP to TiNP–SNAP was 63.77 ± 2% (n = 3), which revealed improvement of TiNPs in delivering NO compared to previously reported halloysite nanoparticles with approximately 50% conversion rate.³² The higher conversion rate of TiNP–SNAP may relate to nanoparticle morphology and colloidal stability of NAP-attached particles, which provided a higher surface area and subsequently increased the reaction success rate. The lack of complete conversion can be related to thermal decomposition of the –SNO group during the nitrosation reaction plus steric hindrance imposed by tertiary thiol functionality and the existence of disulfides in the TiNP–NAP structure. Moreover, steric hindrance imposed by tertiary thiol functionality, and finally, the existence of disulfides in the TiNP–NAP structure can be other impediments in nitrosation reaction.^32,58

3.7. In Vitro NO Release Kinetics under Physiological Conditions.

The development of antibiotic-resistant bacteria strains has been a definitive challenge in medicine.³³ The development of nanoparticles with either innate or imparted antibacterial properties that can be delivered to the site of infections has become one of the most promising solutions to reduce the need for conventional antibiotics and the emergence of mutated resistant strains. In addition, utilizing endogenous substances to augment the bactericidal efficacy of nanoparticles while minimizing the bacterial resistance and the required nanoparticle concentration has been a subject of interest recently. Although NO has been a target for such studies, the lack of efficient delivery from nanomaterials has limited its practical usage.³⁴ In this study, for the first time, TiNP–SNAP is synthesized to deliver NO throughout 20 h in an effective antibacterial range (Figure 5). The TiNP–SNAP antibacterial efficacy relies on the NO delivery behavior of the particles under physiological conditions. The major mechanism of most NO-donating materials that deliver NO is via thermal decomposition of the S–NO bond to disulfide species and a free NO molecule.⁷ The resulting inter- and intra-nanoparticle disulfide formation and S–S crosslinking can lead to aggregation of the particles as reported in prior studies,^56,65 which was also evident by the results obtained from DLS studies. In this study, it is observed that within the first 2 h of particle exposure to 37 °C and the buffer environment, higher NO release values of around 0.4–0.8 × 10⁻¹⁰ mol min⁻¹ mg⁻¹ are observed. However, a more steady and gradual release behavior over 20 h with a cumulative NO concentration of 240.05 ± 74.94 × 10⁻¹⁰ mol min⁻¹ mg⁻¹ was observed. The release behavior of NO from TiNP–SNAP observed here indicates an improvement to the previously reported NO-releasing nanoparticles like SNAP conjugated diatomaceous earth silica with a burst release within the first 2 h of NOA studies or SNAP conjugated halloysite nanoparticles, which had a significantly higher concentration of NO release (0.05 μmol mg⁻¹) but exhibited a much shorter release time of 8 h.^{30,32,56–58,66}

Figure 5.

In vitro, NO release kinetics of the TiNP–SNAP obtained from a 1 mg mL⁻¹ colloidal suspension of the nanoparticles in PBS containing 100 μM EDTA at 37 °C within the amber NOA sample cell. Data represent as the mean ± SD (n = 3).

3.8. TiNP–SNAP 24 h Antibacterial Efficacy Evaluation.

Two of the most common pathogens abundant in medical device infections are S. aureus and E. coli. Although several antibiotics are being used to combat infections, the growing threat of antibiotic-resistant strains and the emergence of mutated S. aureus and E. coli strains force the urge to develop better ways to deal with bacterial infections.⁶⁷ Through the past decades, nanoparticles have been investigated for various antimicrobial applications as they can be beneficial in different ways such as generating a broader spectrum of bactericidal mechanisms, enabling delivery of therapeutic agents in a targeted and controlled manner, expanding the versatility of new drugs, and facilitating the use of non-invasive mediators such as light.⁶⁸ However, the rise of bacterial resistance to new elements such as silver nanoparticles escalated challenges in using metallic nanoparticles.⁶⁹ It has been proven that NO exhibits non-specific mechanisms of bactericidal activity. Production of reactive oxygen species (ROS), damaging the bacteria cell membrane, cleaving the DNA structure, and disturbing the protein functionality of the cell membrane are proposed as antibacterial mechanisms of NO.⁷ However, TiNP is known to have no antibacterial activity, unless excited in the presence of light irradiation, although this can lead to systemic cell apoptosis or genotoxicity.^40,42 Enhancing TiNPs with the NO donor moiety conjugation significantly reduced the viable CFUs from both S. aureus and E. coli bacterial cultures with no need for external stimulations like light irradiation (Figure 6). A concentration-dependent activity was observed in the bactericidal efficacy of TiNP–SNAP (Table 2). More than 99.99 ± 0.01% reduction in viable S. aureus was achieved when it was exposed to a concentration of 5 mg mL⁻¹ TiNP–SNAP. However, the antibacterial efficacy is slightly diminished against Gram-negative E. coli, and 99.70 ± 0.01% reduction in viable CFUs was obtained by using the same concentration of 5 mg mL⁻¹ TiNP–SNAP. Although the difference between the two bacteria strains is not significant, the slight reduction in antibacterial efficacy against Gram-negative bacteria can be related to a specific DNA repair mechanism when exposed to NO that abates DNA damage induced by NO.⁷⁰ In addition, flavohemoglobin (Hmp) enzymes synthesized in Gram-negative bacteria strains can defend against nitrosative stress caused by NO.⁷¹ The observations about NO dose-dependent antibacterial efficacy and less susceptibility of E. coli have been reported in prior studies.⁷¹

Figure 6.

Bactericidal efficacy analysis of TiNP–SNAP against (a) Gram-positive S. aureus and (b) Gram-negative E. coli. The concentration of bacteria (control) without exposure to nanoparticles is presented by solid black bars. Striped black and solid red bars are used to show the concentration of viable bacteria after 24 h exposure to TiNP and TiNP–SNAP, respectively. Data are presented as the mean values ±SD (n = 4), where ** = p ≤ 0.01 and *** = p ≤ 0.001 for TiNP vs TiNP–SNAP.

Table 2.

Viability Reduction of Bacterial Cells (%) after 24 h Exposure to TiNP–SNAP Compared to Control Bacteria Wells Containing No Nanoparticles^a

	bacteria cell viability reduction (%)
nanoparticle concentration (mg mL−1)	S. aureus	E. coli
0.01	94.52 ± 1.18	78.83 ± 2.46
0.1	97.66 ± 0.55	72.31 ± 1.67
5	99.99 ± 0.01	99.70 ± 0.01

^aData are presented as the mean ± SD (n = 4).

3.9. Cytotoxicity Analysis.

In order to evaluate the cytocompatibility of TiNP and TiNP–SNAP particles, the particle leachates from different concentrations (0.01–5 mg/mL) were exposed to 3T3 mouse fibroblast cells for 24 h following a previously reported method.³⁰ Using the CCK-8 assay, the cell viability was determined compared to cells in DMEM as the control. The 3T3 cells exhibited no significant change in the growth profile after the addition of leachates from TiNP–SNAP as well as TiNPs as compared to untreated control cells. Notably, all exposed cells and the untreated control exhibited >90% cell viability over 24 h (Figure 7). In the past, studies involving NO-releasing nanoparticles/nanotubes for biomedical applications have reported similar results with no potential toxicity from the NO-donating functionalities, SNAP, or its degraded byproducts such as NAP and NAP dimers.^30,32 Overall, the results obtained in this study corroborate with previously reported studies highlighting the biocompatibility of NO-releasing materials for clinical translation. Moreover, several studies have been done on the NO effect on the viability of mammalian cells. It is well demonstrated that exposure to a low concentration of NO can impart a regulatory effect on several types of mammalian cells and increase their growth and proliferation rates.²⁵ The toxicity profile and allergic reactivity of TiNP are also well established in previous reports.^37,42 Many cosmetic products are allowed to contain up to 25 wt % TiNP in them under European regulations.⁷² Although TiNPs are demonstrated to not exhibit cell cytotoxicity in their stable conditions, it is reported that in a UV-excited state, oxidative stress-induced cell apoptosis can occur in a concentration-dependent manner,⁴² and the subject of potential risks in inhaling TiNPs has been studied previously. Lee et al. reported cystic keratinizing squamous cell carcinomas occurring in a rat model after long-term exposure to a high concentration of TiNPs (250 mg m⁻³ for 2 years, 5 days a week, and 6 h each day).⁷³ Accordingly, in this study, the SNAP-immobilized TiNPs exhibit an excellent bactericidal activity without the use of any light activation method, while the TiNP–SNAP leachates exhibit no cytotoxicity. These findings can expand TiNP–SNAP applications in cosmetics, drug delivery systems, and polymeric matrix nanofillers with a low risk of leaching of the cytotoxic NO donor moiety or loss of NO due to diffusion of the donors to the surrounding environment with the added benefit of antimicrobial activity and obviated need of UV excitation.

Figure 7.

Cytocompatibility of TiNP and TiNP–SNAP leachates in media toward NIH 3T3 mouse fibroblast cells. The viability of cells was assessed using a CCK-8 cell viability kit. The cells exposed to the TiNP and TiNP–SNAP extracts depicted by striped black and solid red bars, respectively, exhibited cell viability similar to that of the untreated control cells in DMEM that were not exposed to TiNPs, depicted in solid black bars, highlighting the non-toxic nature of the particle leachate. Data are presented as mean ± SD for n = 6, normalized to the untreated cell control.

4. CONCLUSIONS

Infections and biofilm formation are among the most challenging burdens on medical devices and systems. One of the most promising solutions to this problem is the development of nanoparticles to combat bacterial infections. One of the biocompatible nanoparticles that have been a matter of interest for a variety of applications is TiNPs. TiNPs are being used in a wide range of biomedical applications such as fillers, coatings, medicine, and pharmacology, and they are proven to be bioinert and exhibit antibacterial efficacy when excited under light irradiation. However, the lack of innate antibacterial properties without an external source of irradiation hindered the application in the biomedical realm. In this study, for the first time, the NO-releasing SNAP functionality was covalently conjugated to the TiNP surface through a sequence of chemical reactions. Successful conjugation was qualitatively confirmed via FTIR and NMR analysis. The primary amine group quantity on the surface of TiNP–APTES was measured to be 0.24 ± 0.01 μmol mg⁻¹ according to ninhydrin assay results. More than 83% of primary amine groups were converted to sulfhydryl groups (~0.20 ± 0.01 μmol mg⁻¹) as detected by Ellman’s assay. The chemiluminescence NO analysis revealed 127.55 ± 4.68 nmol mg⁻¹ NO loading on TiNP–SNAP after nitrosation reaction with t-butyl nitrite with an efficiency rate of more than 63%. Zeta potential and DLS analysis results were in good correlation with chemical modifications and corroborated colloidal stability and nanostructural integrity of particles in each step of modification. The TiNP–SNAP exhibited promising concentration-dependent antibacterial results against two of the most common healthcare-associated infections, E. coli and S. aureus. More than 99.99% and 99.70% reduction in viable CFUs of S. aureus and E. coli were observed, respectively, after 24 h of exposure to a 5 mg mL⁻¹ TiNP–SNAP suspension. No cytotoxicity was observed from the leachate assay of TiNP–SNAP against mouse fibroblasts with respect to the control. The results from this study illustrated the great potential of TiNPs as a suitable nanocarrier of NO donors. Future work will include studying the photoactive properties of TiNP–SNAP as well as evaluating their photoactivity after the NO reservoir has been depleted in order to evaluate the synergy of the NO release with the photoactive antimicrobial activity of TiNP. In conclusion, the resulting TiNP–SNAP can be used in many clinical and preclinical applications such as antimicrobial delivery, wound healing, polymer fillers, coatings, and so forth.