Biotemplated Synthesis and Characterization of Mesoporous Nitric Oxide-Releasing Diatomaceous Earth Silica Particles

Abstract

Diatomaceous earth (DE), a nanoporous silica material composed of fossilized unicellular marine algae, possesses unique mechanical, molecular transport, optical, and photonic properties exploited across an array of biomedical applications. The utility of DE in these applications stands to be enhanced through the incorporation of nitric oxide (NO) technology shown to modulate essential physiological processes. In this work, the preparation and characterization of a biotemplated diatomaceous earth-based nitric oxide delivery scaffold are described for the first time. Three aminosilanes [(3-aminopropyl)triethoxysilane (APTES), N-(6-aminohexyl)aminomethyltriethoxysilane (AHAMTES), and 3-aminopropyldimethylethoxysilane (APDMES)] were evaluated for their ability to maximize NO loading via the covalent attachment of N-acetyl-d-penicillamine (NAP) to diatomaceous earth. The use of APTES cross-linker resulted in maximal NAP tethering to the DE surface, and NAP-DE was converted to NO-releasing S-nitroso-N-acetyl-penicillamine (SNAP)-DE by nitrosation. The total NO loading of SNAP-DE was determined by chemiluminescence to be 0.0372 ± 0.00791 μmol/mg. Retention of diatomaceous earth’s unique mesoporous morphology throughout the derivatization was confirmed by scanning electron microscopy. SNAP-DE exhibited 92.95% killing efficiency against Gram-positive bacteria Staphylococcus aureus as compared to the control. The WST-8-based cytotoxicity testing showed no negative impact on mouse fibroblast cells, demonstrating the biocompatible potential of SNAP-DE. The development of NO releasing diatomaceous earth presents a unique means of delivering tunable levels of NO to materials across the fields of polymer chemistry, tissue engineering, drug delivery, and wound healing.

1. INTRODUCTION

Nitric oxide (NO), a gaseous diatomic free radical endogenously produced via the sequential enzymatic oxidation of l-arginine, plays an essential and far-reaching role in human physiology.^1,2 In the 30 years since it was first identified as endothelium-derived relaxing factor, NO has been shown to modulate essential biological processes including smooth muscle relaxation, cell proliferation, vasodilation, neurotransmission, cell signaling, the inhibition of platelet adhesion and aggregation, and immune system regulation.^1,3–5 Nitric oxide’s efficacy in these diverse roles stems from its high membrane diffusivity and excellent reactivity with a variety of chemical species including oxygen, superoxide anions, oxyhemoglobin, thiols, pyrimidine bases, lipids, and metallic complexes.^1,6,7 In recent decades, “donor” molecules that release NO at or above physiological levels have been incorporated into biomaterials to artificially induce therapeutic effects consistent with those of endogenous NO.^4–6,8–10 These donor molecules act as NO delivery vehicles, making targeted NO administration feasible by eliminating the spatial and temporal issues surrounding the molecule’s short physiological half-life and diffusion distance (1–3 μs and 100–200 μm, respectively).^6,7,9,11,12

S-Nitrosothiols (RSNOs) in particular have emerged as one of the most popular classes of nitric oxide donors.^1,9,10 Synthesized via an acidified nitrosation reaction between thiols (RSH) and nitrite, nitrogen oxides, or alkyl nitrites, RSNOs undergo chemical, photolytic, and thermal decomposition to release NO.^9,13 S-nitroso-N-acetyl-penicillamine (SNAP), the nitrosated form of N-acetylpenicillamine (NAP), an amino acid derivative that has been used to treat cystinuria at doses of 2–4 g/day for 155 days with minimal side effects, is one of the most prevalent RSNO molecules due to its relatively high molecular stability and nontoxic origins.^1,9,10,14

To date, most research in the field of NO technology has focused on the development of antithrombogenic and antimicrobial biomaterials such as catheters, extracorporeal circuitry, biosensors, and biomedical device coatings.^12,15–20 In addition, NO has also been incorporated into food packaging, acne medications, wound healing materials, and toothpastes.^21–24 The addition of NO donor molecules to these wide-ranging materials is accomplished by either physical (blending or swelling) or chemical means (covalent attachment).^{13,15,17,18,20,24,25} Although easier to produce, blended and swollen NO releasing materials often suffer from molecular leaching and diminished release times.^12,13 The covalent attachment of NO donors to polymer backbones and release scaffolds protects against these limitations, increasing the stability, safety, and application range of materials.^12,13,26

A popular material of choice for covalently formed NO releasing scaffolds is silicon dioxide due to its chemical inertness, tunable particle size, affordability, and abundance.^26–31 A handful of research groups have previously produced silica-based NO scaffolds using a variety of methodologies.^26–31 For instance, Zhang et al. modified fumed silica with amine-containing silylation reagents to create nonporous diazeniumdiolated silica particles (0.2–0.3 μm) that enhanced thromboresistance when embedded in extracorporeal circuit tubing.²⁷ In a similar fashion, Frost et al. employed silylation agents to tether RSNOs to nonporous fumed silica particles (7–10 nm) and analyzed their release kinetics under various conditions.²⁶ Shin et al. synthesized nonporous diazeniumdiolated NO releasing silica particles (20–500 nm) de novo via the co-condensation of two silicon alkoxide precursors.^28,29 Hetrick et al. later tested these particles for antibiofilm and bactericidal efficacy.^32,33 Most recently, Soto et al. created diazeniumdiolated porous silica particles (30–1100 nm) using a modified alkoxysilane co-condensation synthesis.³⁰

Previously synthesized scaffolds have struggled to strike an ideal balance between NO release kinetics, morphology, particle size, and ease of synthesis. Specifically, diazeniumdiolate-based silica scaffolds undergo burst release, eliminating all stored nitric oxide within hours and limiting their utility.^27–30 Additionally, virtually all previously reported NO releasing scaffolds have been nonporous and on the nanoscale.^26–31 Mesoporous, micrometer scale silica RSNO scaffolds have not yet been created and demonstrate novel physical properties and NO release kinetics. There exists perhaps no better template for such a scaffold than diatomaceous earth.

Diatomaceous earth (DE) consists of the fossilized 10–150 μm shells of diatoms—a class of unicellular marine algae possessing extraordinarily intricate and porous three-dimensional morphologies.^31,34,35 With an estimated world reserve of 800 million metric tons and countless applications across the food, cosmetic, chemical, pharmaceutical, and medical industries, diatomaceous earth is a material as ubiquitous as it is versatile.^31,34 The unique structure of diatoms, coupled with their high amorphous silicon dioxide content (at times 95%), renders diatomaceous earth a low density, high surface area, chemically inert, all-natural, abrasive absorptive.^31,34 Because of these properties, diatomaceous earth is routinely used as a filtration aid, natural detoxifier, cosmetic and personal hygiene abrasive, insecticide, drug delivery and tissue engineering scaffold, wound healing agent, and polymeric filler.^{31,34,36–39} Chemically modifying diatomaceous earth to release nitric oxide stands to enhance the material’s already manifest versatility and efficacy in these biomedical applications and others.

In this work, three primary aminosilanes (APTES, AHAMTES, and APDMES) were used to tether NAP thiolactone, a self-protected penicillamine derivative, to 10–15 μm size DE particles. The efficiencies of both surface silylation and NAP thiolactone attachment were compared between aminosilanes. Diatomaceous earth modified with APTES/NAP yielded the highest levels of silane and NAP attachment and was nitrosated for further evaluation. The chemical modification of DE and retention of particle morphology throughout the derivatization were verified by Fourier transform infrared spectroscopy (FTIR), energy-dispersive X-ray spectroscopy (EDS), and scanning electron microscopy (SEM). Nitric oxide release over a 24 h period and total NO content were determined by chemiluminescence. Lastly, the antibacterial and noncytotoxic properties of the biotemplated NO-releasing diatomaceous earth silica scaffolds were evaluated.

2. MATERIALS AND METHODS

2.1. Materials.

Fossil Shell Flour Diatomaceous Earth was purchased from Perma-Guard, Inc. (Bountiful, UT). 200 proof ethanol was obtained from Decon Laboratories, Inc. (King of Prussia, PA). Toluene and methanol were purchased from Fischer Scientific (Waltham, MA). (3-Aminopropyl)triethoxysilane (APTES), l-cysteine hydrochloride monohydrate, 4-dodecylbenzenesulfonic acid, Ellman’s Reagent (5,5′-dithiobis(2-nitrobenzoic acid), DTNB), glycine hydrochloride, 1,4,8,11-tetraazacyclotetradecane (cyclam), tert-butyl nitrite, and potassium cyanide (KCN) were purchased from Sigma-Aldrich (St. Louis, MO). N-(6-Aminohexyl)aminomethyltriethoxysilane (AHAMTES) and 3-aminopropyldimethylethoxysilane (APDMES) were purchased from Gelest, Inc. (Morrisville, PA). Sodium acetate was obtained from EMD Chemicals, Inc. (Gibbstown, NJ). The bacterial strain Staphylococcus aureus (ATCC 6538) and mouse 3T3 cells (ATCC 1658) were originally purchased from American Type Culture Collection (ATCC).

2.2. Preparation of N-Acetyl-d-penicillamine (NAP) Thiolactone.

Self-protected NAP thiolactone was synthesized via a slightly modified protocol by Moynihan and Robert.⁴⁰ A solution of 5 g of NAP in 10 mL of pyridine and a separate mixture of 10 mL of pyridine and 10 mL of acetic anhydride were made. Both solutions were chilled in an ice bath for 1 h before being combined and continuously stirred for 24 h. Afterward, all pyridine in the solution was removed by rotary evaporation at 60 °C to leave behind a small amount of viscous, orange material. This material was dissolved in chloroform and repeatedly washed and extracted with 1 M HCl. The organic layer containing NAP thiolactone was then dried using anhydrous magnesium sulfate subsequently eliminated by filtration. Chloroform was removed under vacuum at room temperature. The collected solid product was washed with hexanes and allowed to dry overnight at room temperature before being stored at 5 °C.

2.3. SNAP-Functionalized Diatomaceous Earth Derivatization.

A schematic overview of the SNAP-functionalized diatomaceous earth derivatization is shown in Figure 1. First, purified DE was amine-functionalized via silylation with APTES, AHAMTES, or APDMES. Next, amine-functionalized particles were reacted with NAP thiolactone to covalently tether NAP to DE. Finally, NAP-DE was treated with tert-butyl nitrite under acidic conditions to form NO-releasing SNAP-DE.

Figure 1.

NO-releasing diatomaceous earth derivatization schematic featuring APTES as a representative silane.

2.3.1. Diatomaceous Earth Purification.

To remove trace organic impurities from DE, an aqueous diatomaceous earth suspension was made in a beaker and sonicated. After sonication, dark impurities settled while DE remained suspended. The suspension was decanted into a separate beaker, and the process was repeated three times or until all sediment was eliminated. The water in the purified suspension was removed by centrifugation at 3500 rpm for 3 min and dried under vacuum.

2.3.2. Surface Silylation.

Silylated diatomaceous earth (aminosilane-DE) was prepared by refluxing purified DE with one of three aminosilanes (APTES, AHAMTES, and APDMES) in toluene for 24 h in accordance with a previously reported protocol (1 g of DE:21.4 mmol of aminosilane:100 mL of toluene).²⁶ Primary amine-containing silanes (Figure 2) were selected as cross-linkers because of their ability to promote the NAP-thiolactone ring-opening required to tether NAP to aminosilane DE via an amide bond.⁴¹ After each reflux, the amine-functionalized DE products were washed four times with toluene and twice with ethanol before being dried in an oven at 80 °C overnight.

Figure 2.

(a) Traditional structural representations of silylation agents. (b) Conformational representations of silylation agents during attachment to silica surfaces. The red “×” illustrates the AHAMTES conformer’s reduced ability to participate in intramolecular catalysis.

2.3.3. NAP Attachment.

NAP-DE was prepared by stirring aminosilane-DE with NAP-thiolactone for 24 h in toluene (100 mg of silylated DE:80 mg of NAP-thiolactone:5 mL of toluene). Reaction products were washed twice with toluene and dried under vacuum at room temperature for 24 h.

2.3.4. Nitrosation.

NAP-DE was added to a solution of 10% methanol, 90% toluene along with 4-dodecylbenzenesulfonic acid (1 mL of DBSA:100 mg of APTES/NAP diatoms) and a molar excess of tert-butyl nitrite. The tert-butyl nitrite was first cleansed of any copper contaminants by vortexing with an equal volume of 20 mM cyclam. The reaction vessel was shielded from light and agitated for 2 h before its contents were dried at room temperature under vacuum for approximately 30 h.

2.4. FQ Primary Amine Quantification.

The ATTO-TAG FQ test for primary amines was conducted in accordance with a previously reported protocol.⁴² Stock solutions of 10 mM FQ and 10 mM KCN in methanol and water, respectively, were prepared. A working ATTO-TAG FQ solution was created which consisted of 10 μL of FQ stock solution, 20 μL of KCN stock solution, 190 μL of water, and 5 μL of sample. A microplate reader (Biotek, Winooski, VT) recorded fluorescence measurements at an excitation of 480 nm and emission maxima at 590 nm. Using the ATTO-TAG FQ solution, a calibration curve of known glycine hydrochloride concentrations was created and the amine content of aminosilane-DE was determined.

2.5. Ellman’s Test for Free Sulfhydryls.

Ellman’s Reagent, 5,5′-dithiobis(2-nitrobenzoic acid), was used to quantify the free sulfhydryl content of NAP-DE according to a previous protocol.⁴³ Briefly, a DTNB stock solution (2 mM DTNB, 50 mM NaAc) was used to create a working DTNB solution consisting of 50 μL of DTNB stock solution, 100 μL of PBS, 840 μL of H₂O, and 10 μL of sample. A UV–vis spectrophotometer (Thermo Scientific Genesys 10S UV–vis) recorded absorbance measurements at a previously reported wavelength of 412 nm. Using the DTNB working solution, a calibration curve of known l-cysteine hydrochloride monohydrate concentrations was created, and the sulfhydryl content of NAP-DE was determined.

2.6. Fourier Transform Infrared Spectroscopy (FTIR).

FTIR analysis was used to confirm the presence, absence, and modification of various functional groups throughout the synthesis of SNAP-DE. FTIR spectra of translucent KBr pellets prepared using a 1:100 mass ratio of DE particles:KBr were recorded with a Nicolet 6700 spectrometer (Thermo Electron Corporation, Madison, WI). For each sample, 128 scans were obtained at a resolution of 4 cm⁻¹ over the wavenumber range of 4000–400 cm⁻¹.

2.7. Nitric Oxide Release Measurements.

Measuring NO release from SNAP-DE was done in real-time via chemiluminescence using a Sievers Nitric Oxide Analyzer (NOA) model 280i (Boulder, CO). Samples were weighed and subsequently tested by submersion in 0.01 M PBS containing EDTA at 37 °C inside of an amber reaction vessel to protect from light. Nitrogen gas was continuously bubbled and swept from the vessel at a flow rate of 200 mL min⁻¹ to carry the NO being released to the NOA.

The NOA was used to measure both the total NO loading and continuous NO release of SNAP-DE. To measure total NO loading, alternating injections of 0.25 M copper(II) chloride and ascorbic acid were added to the NOA reaction vessel. These injections triggered the release of all NO present in the sample within a time frame measurable by NOA. The 24 h continuous NO release profile was determined by linearly interpolating between steady-state data recorded at the beginning and end of a 24 h period. Between the final and initial NOA measurements, physiological conditions were maintained by storing the samples in 0.01 M PBS with EDTA at 37 °C. Continuous release data were reported as both the instantaneous NO release (ppb/mg) and as a cumulative percent of the total NO initially loaded onto the DE particles.

2.8. Scanning Electron Microscopy and Energy-Dispersive X-ray Spectroscopy.

Scanning electron microscopy (SEM, FEI Teneo, FEI Co.) was employed at an accelerating voltage of 5.00 kV to examine the morphology of diatomaceous earth throughout the derivatization. The SEM was equipped with a large detector energy-dispersive X-ray spectroscopy (EDS, Oxford Instruments) system used for elemental analysis and mapping of modified diatomaceous earth.

2.9. Bacterial Inhibition Test.

In the current study, the antibacterial properties of SNAP-DE were tested against Gram-positive Staphylococcus aureus (S. aureus), a common causative agent of blood and nosocomial infections.^44–46 Single isolated colony forming units of S. aureus strains were obtained from a precultured LB agar Petri dish, inoculated in 10 mL of LB medium, and incubated at 37 °C, 120 rpm for 14 h. To ensure that the bacteria used in this study were in an actively dividing log phase, the optical density of the culture was measured at a wavelength of 600 nm (OD₆₀₀) using a UV–vis spectrophotometer (Thermo Scientific Genesys 10S UV–vis). The bacteria were then separated from the original media and suspended in PBS buffer. This provided the bacteria with an osmotic physiological environment and prevented bacterial proliferation. The separation of cells from the medium was achieved by centrifugation for 7 min at 3500 rpm. The supernatant was discarded, replaced with sterile PBS to eliminate traces of medium, and centrifuged for 7 min at 3500 rpm. The supernatant was again discarded, and the cells were resuspended in PBS.

The OD₆₀₀ of the cell suspension in PBS was measured and adjusted to keep the cell count in the range of 10⁸–10¹⁰ colony forming units (CFUs) per milliliter. The SNAP-DE and unmodified DE were suspended in triplicates (n = 3) in 1 mL of PBS-bacteria culture. The bacterial suspension without any DE exposure was taken as a positive control. Before suspension, SNAP-DE was weighed such that 0.8 μmol of NO was released per milliliter of PBS/cell solution. This weight was determined by calculating the total NO released per milligram of SNAP-DE over a 24 h period under conditions mimicking those of the bacterial suspension. The resulting mixture was incubated at 120 rpm and 37 °C for 24 h. After 24 h, the bacterial suspension was gently agitated with a pipet and serially diluted (10⁻¹–10⁻⁵) for plating in premade LB agar Petri dishes. The Petri dishes were incubated at 37 °C for 24 h. In parallel, serial dilutions of the bacteria were prepared just before suspending the diatoms in the bacteria culture and plated in LB agar Petri dishes. This verified the consistency of viable cell concentrations between experiments. Post incubation, the CFUs/mg were counted (eq 1) to observe the relative bactericidal effect shown by the diatoms and ultimately the relationship between NO release and bactericidal activity.

$$\%\text{ bacterial inhibition}=\frac{\text{colonies in control}-\text{colonies in test}}{\text{colonies in control}}\times 100$$ (1)

2.10. Formazan-Based Cell Cytotoxicity Test.

The cell cytotoxicity kit (CCK-8) (Sigma-Aldrich) provides a standard WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium monosodium salt] based cell viability assay. The CCK-8 test is nondestructive in nature and more sensitive than other tetrazolium salts such as MTT, XTT, WST-1, and MTS. The number of living cells is directly proportional to the amount of formazan dye (orange color) generated by the interaction of WST-8 with dehydrogenases in the cells and is detected at the absorbance maxima of 450 nm.

Preparation of Leachates.

The ISO 10993-5:2009 test for in vitro cytotoxicity was followed to generate leachates from unmodified control DE and SNAP-DE (concentration of 1 mg/mL of medium). This was done by soaking 10 mg of the sterilized SNAP-DE for 24 h at 37 °C in amber vials containing 10 mL of DMEM medium. After 24 h, the extracts were kept in the refrigerator (4 °C) prior to use in the cell culture experiment.

Cell Culture.

Mouse fibroblast cells were used to evaluate the mammalian cytotoxicity of SNAP-DE. A 3T3 mouse fibroblast cell line (ATCC-1658) was used, and leachates were obtained from the biomaterial in accordance with the ISO 10993 standard. A cryopreserved vial was thawed and cells were cultured in a 75 cm² T-flask containing complete DMEM medium with 10% fetal bovine serum (FBS). Additionally, 1% penicillin–streptomycin was added to prevent contamination. The T-flask with cells was incubated for eight days at 37 °C in a humidified atmosphere of 5% CO₂ to allow for monolayer formation. The culture medium was replaced intermittently and cells were checked daily for growth and contamination. After the confluence reached above 80%, cells were detached from the T-flask (trypsinized with 0.18% trypsin and 5 mM EDTA). Finally, the cells were counted under hemocytometer using Trypan blue (dye exclusion method). Around 5000 cells/mL were seeded in a cell culture grade 96 well plate and incubated for 24 h in a humidified incubator with 5% CO₂.

Cytotoxicity Test.

The manufacture’s protocol (Sigma-Aldrich) was followed to perform the cytocompatibility test using a CCK-8 kit on the mouse fibroblast cells. After 24 h of cell culture incubation in a 96-well plate, 10 μL of the leachates from control DE and SNAP-DE was added (n = 7) to the cells. The cells were allowed to respond to the leachates during a separate 24 h incubation period inside a cell culture incubator at physiological temperature. After 24 h, 10 μL of the WST-8 solution was added to the resulting solution and incubated for 4 h. During this time, dehydrogenase enzymes from live cells acted on the WST-8 solution, converting it to an orange product, formazan, measurable at 450 nm. The relative viability (%) of the cells in response to SNAP-DE leachates was reported relative to the control (without leachate exposure) using eq 2:

$$\%\text{ cell viability}=\frac{\text{absorbance of the test samples}}{\text{absorbance of the control samples}}\times 100$$ (2)

2.11. Statistical Analysis.

All data herein are reported as mean ± standard deviation with n = 3 data points unless otherwise noted. Any statements of statistical significance were made using standard two-tailed t-tests and comparison p-values of <0.05.

3. RESULTS AND DISCUSSION

3.1. Amine Quantification of Functionalized Diatomaceous Earth.

The primary amine content of DE postsilylation was quantified to verify the presence of amine-functionalized intermediates and gauge overall reaction efficiencies. Because it is well documented that nonsurface bound silanes are readily eliminated by thorough washing, it can be safely assumed that the primary amines detected by FQ belong exclusively to surface bound coupling agents.^47–50 ATTO-TAG FQ reacts with the primary amines of these coupling agents to form a highly fluorescent product detectable to the attomole range.⁴² Post-silylation primary amine levels were found to be 1.10 ± 0.17, 0.15 ± 0.01, and 0.11 ± 0.03 μmol/mg for APTES, AHAMTES, and APDMES treated DE, respectively (Table 1).

Table 1.

Amine Contents, Thiol Contents, and Conversion Ratios for APTES, AHAMTES, and APDMES Treated DE^a

cross-linkerb	amine contentc (μmol/mg)	thiol contentd (μmol/mg)	conv ratioe (%)
APTES	1.10 ± 0.17	0.0312 ± 0.006	2.84
AHAMTES	0.15 ± 0.01	0.0181 ± 0.003	12.1
APDMES	0.11 ± 0.03	0.0130 ± 0.001	11.8

^aAll data are reported as mean ± standard deviation for n = 3 data points unless otherwise noted. APTES thiol content was determined using n = 9 data points.

^bType of aminosilane used to functionalize the surface of diatomaceous earth particles.

^cPrimary amine content of diatomaceous earth particles after silylation.

^dSulfhydryl content of silylated diatomaceous earth after NAP addition.

^eRatio between sulfhydryl content after NAP addition and amine content before NAP addition.

Interestingly, APTES-DE showed amine levels approximately 7 times higher than AHAMTES-DE and 10 times higher than APDMES-DE derived under the same conditions. The likely explanation for this variance stems from the differing molecular structures, and thus reactivities, of the three aminosilanes. It is widely theorized that aminosilane attachment proceeds via the primary amine catalyzed S_N2 exchange reaction between the ethoxy groups of silanes and the oxygens of silanols.^47–49 Because of this, APTES and AHAMTES, which possess three ethoxy groups each, are inherently more reactive than APDMES containing only a single ethoxy group. Three ethoxy moieties allow APTES and AHAMTES to self-polymerize with surface bound and free aminosilanes to form three-dimensional amine rich surface multilayers (Figure 3).

Figure 3.

Schematic representation of aminosilane self-polymerization occurring on the diatomaceous earth silica surface.

While this explains the tendency for APDMES to form only low amine content monolayers, it fails to explain why APTES and AHAMTES, despite their equivalent number of ethoxy groups, result in considerably different amine concentrations. Because aminosilylation relies upon intramolecular primary amine catalysis, it is essential that the terminal amines of APTES and AHAMTES be available to the sites of S_N2 exchange.^48,49 While APTES is believed to form a five-membered cyclic intermediate which places its primary amine adjacent to the site of S_N2 exchange, AHAMTES possesses a significantly longer alkyl chain which reduces its ability to undergo a similar intramolecular catalysis (Figure 2).^48,49 This conformational difference reduces the ability of AHAMTES to form high amine content surface coatings.⁴⁹

3.2. Sulfhydryl Quantification of Diatomaceous Earth.

The thiol content of silylated diatomaceous earth after reacting with NAP-thiolactone was determined using Ellman’s Reagent (DTNB). DTNB reacts with sulfhydryls to diffuse a yellow product into solution that is quantifiable by UV–vis.⁴³ Because free thiols arise only after the primary amine-initiated ring-opening of NAP-thiolactone, sulfhydryl content serves as a direct indicator of covalent NAP attachment. The sulfhydryl concentrations of APTES, AHAMTES, and APDMES were found to be 0.0312 ± 0.006 (n = 9), 0.0181 ± 0.003 (n = 3), and 0.0130 ± 0.001 μmol/mg (n = 3), respectively (Table 1).

While these results support the expectation that higher levels of surface bound amines result in increased NAP attachment, a direct proportionality between amine content and subsequent NAP attachment was not observed. Specifically, while one would expect APTES-DE to possess NAP levels 7 times higher than AHAMTES-DE and 10 times higher than APDMES-DE (based on amine content), APTES-DE instead demonstrated 1.7 and 2.4 times more NAP attachment than AHAMTES-DE and APDMES-DE, respectively. A closer examination of these results reveals that the percentages of surface amines tethered to NAP-thiolactone were 2.84, 12.1, and 11.8% for APTES, AHAMTES, and APDMES, respectively.

Because the thickness of aminosilane layers is the most meaningful difference between cross-linkers employed in this work, lower amine conversion ratios for APTES multilayers suggest that the deposition of aminosilane coupling agents past monolayer thicknesses improves NAP attachment only marginally. The authors offer the following discussion as a plausible, nonexperimental explanation for the observed data. It has been suggested that APTES routinely forms nonuniform, highly dense, interconnected silane networks (Figure 3).^47–49,51 In such an environment, NAP-thiolactone, with its highly substituted ring structure, likely experiences steric congestion toward nucleophilic attack.⁴¹ Accordingly, NAP-thiolactone ring-opening is unlikely to occur within the interconnected silane network believed to be present on the diatomaceous earth surface. However, penetration is not impossible, and although the aminosilane multilayers of APTES resulted in a lower overall conversion ratio, an increase in the sheer quantity of NAP attachment was observed.

While steric hindrance may explain differences in amine conversion between dense multilayers of APTES and thin layers of AHAMTES and APDMES, relatively low conversion ratios for even thin aminosilane layers suggest that more factors are at play in NAP-thiolactone binding than sterics alone. Similarly, low amine conversions were observed by Frost et al. when tethering NO-releasing groups to amine-modified fumed silica particles.²⁶ An alternative explanation for the presence of high APTES levels with comparatively low thiol levels is that APTES, possessing a shorter, more reactive alkyl chain, may self-react to produce a bulk product distinct from diatomaceous earth particles. While plausible, it has been previously shown that the synthetic conditions used in this work (low silane concentration, moderate reaction temperature, and thorough washing with organic solvents) minimize the formation of aminosilane-based oligomers and polymers.⁴⁹ Additional evidence for a DE surface coating rather than separate bulk particles will be discussed in the SEM and EDS results sections. Future work will explore ways to maximize the binding of amines to NAP by fine-tuning synthetic parameters such as reagent concentration, reaction time, and pH.

The central goal of this investigation was to develop diatomaceous earth particles possessing maximal NO release. High NO loading per DE particle ensures that the therapeutic effects of nitric oxide are elicited with minimal material input. Although a higher percentage of AHAMTES and APDMES cross-linkers were converted to AHAMTES-NAP and APDMES-NAP, the total amount of NO donor immobilized with these cross-linkers was considerably less than the quantity immobilized when APTES cross-linker was employed. For this reason, APTES-NAP, with its high quantity of nitrosatable sulfhydryl groups, was selected for further characterization.

3.3. Fourier Transform Infrared Spectroscopy.

FTIR spectra of unmodified-DE and NAP-DE are shown in Figure 4 and suggest the successful chemical modification of diatomaceous earth. In the unmodified-DE spectrum, the vibration observed at 3321 cm⁻¹ corresponds with both the Si–OH bonds abundant on the silica surface and the −OH bonds of water physically absorbed to the silica surface. In the same spectrum, the band at 1632 cm⁻¹ is consistent with bending vibrations of surface bound H₂O. The disappearance of the broad peak at 3321 cm⁻¹ in the NAP-DE spectrum indicates the elimination of Si–OH groups and physically absorbed water upon silylation with APTES and NAP attachment. Moreover, the presence of conjugated amides consistent with the structure of NAP-DE in this spectrum is suggested by carbonyl vibrations at 1759 and 1657 cm⁻¹, N–H stretching at 3251 and 3197 cm⁻¹, and N–H bending at 1562 cm⁻¹. The sp³ C–H bonds consistent with the alkyl chain of APTES and methyl groups in NAP are seen in subtle vibrations at 3051, 2966, and 2916 cm⁻¹.

Figure 4.

FTIR spectra of unmodified DE (red) and NAP-DE (blue).

3.4. Nitric Oxide Content and Release Kinetics of SNAP-DE.

Chemiluminescence, one of most popular means of quantifying nitric oxide release from materials, was used to determine the total NO content and release kinetics of SNAP-DE. Total NO release from SNAP-DE was found to be 0.0372 ± 0.00791 μmol/mg using alternating injections of 0.25 M copper(II) chloride and ascorbic acid. This value of NO loading is within range of the sulfhydryl levels quantified by Ellman’s Assay (0.0312 ± 0.0061 μmol/mg), indicating efficient nitrosation. Nitric oxide release of 0.0372 ± 0.00791 μmol/mg attained through the covalent attachment of NO donor minimizes the chance of toxic NO levels occurring when SNAP-DE is substituted for traditional DE in applications requiring bulk quantities of material. Furthermore, because targeted NO release levels vary greatly across applications, the ability to fine-tune NO flux by modulating the mass of SNAP-DE incorporated into materials is a tremendous asset.

Physiological conditions were chosen for NO release testing to mimic the in vivo conditions of biomedical applications and for facile comparison with previously reported NO-releasing particles. Figure 5 illustrates the nitric oxide release from SNAP-DE as both an instantaneous value and a cumulative percentage of total NO loading over a 24 h period. The cumulative percentage was obtained by dividing the total NO released at any time point by the total amount of NO initially loaded onto the DE (0.0372 ± 0.00791 μmol/mg).

Figure 5.

Representative 24 h SNAP-DE NO release profile as both an instantaneous PPB value and cumulative percentage. Because SNAP-DE NO-release levels plateaued quickly, this profile was prepared by linearly interpolating between steady-state data recorded at the beginning and end of the 24 h period.

Notably, it can be inferred from the red dotted line in Figure 5 that the release half-lives of SNAP-DE routinely exceeded 24 h. This marks a significant improvement upon the half-lives of previously reported diazeniumdiolate-based silica particles (6 min–12 h).^27,28,30

Sustained (relative to the release durations of similar silica particles) release of NO offers a unique combination of NO release kinetics and loading which expands upon the applications for currently existing NO technology, particularly in the areas of platelet inhibition and bacteria killing. Because the NO release levels of SNAP-DE plateaued shortly after addition to the reaction chamber, a 24 h release profile was prepared by linearly interpolating between steady-state data recorded at the beginning and end of a 24 h period. The assumption of linear NO release between measurements is reasonable because the particles were stored in the same solution (0.01 M PBS with EDTA) and at the same temperature (37 °C) throughout the 24 h period. Under these constant conditions, the observation of release kinetics differing markedly from those at the beginning and end of the study is unlikely.

Nitric oxide release under physiological conditions proceeds primarily by catalytically mediated interactions with solution ions. In Figure 5, the large release of NO observed within the first hour is likely due to either loosely bound silane–SNAP complexes trapped within DE particles or less entangled SNAP functional groups rapidly interacting with solution ions. After the initial burst release of NO, the SNAP-DE material follows zero-order kinetics as a consistent release of NO is interpolated over the 24 h period (0.0485 μmol of NO/mg·h).

There are several strategies to finely tune the NO release of the proposed SNAP-DE material. To inhibit NO release, the material can be blended within polymers that prevent ion diffusion. This would make thermal degradation the primary mechanism of NO release from SNAP. Additionally, certain transition metal ions or metal nanoparticles shown to have high catalytic activity for S-nitrosothiols could be incorporated into the nanopores of DE to increase NO release.⁵ Because selective amounts of SNAP–DE particles can be added to materials to modulate NO release, nitric oxide-releasing DE offers a naturally based, bio-inspired, and tunable means of incorporating NO into polymers, hydrogels, pastes, and creams.

3.5. Scanning Electron Microscopy and Energy-Dispersive X-ray Spectroscopy.

SEM images of diatomaceous earth seen in Figure 6 illustrate the retention of particle morphology throughout the SNAP–DE derivatization. The small fragments observed on the particle surfaces in Figure 6a are likely pulverized DE removed in Figure 6b after either extensive washing steps or altered surface chemistry. It is noteworthy that discrete particles consistent with the undesirable formation of bulk product from aminosilane self-reaction do not appear after surface functionalization (Figures 6b,c). This, when considered with other characterizations, suggests that the surface of diatomaceous earth was modified successfully.

Figure 6.

SEM images of diatomaceous earth (a) before, (b) after APTES silylation but before NAP attachment, and (c) after covalent SNAP attachment. The width of the white scale box is the number located within it.

Morphological conservation is imperative to the maintenance of diatomaceous earth’s uniqueness as an NO donor system. In the past, fumed silica has been used in the derivatization of NO-releasing silica particles.^26,27 A known issue with fumed silica, however, is its aggregation into course, irregular clusters during pyrolytic production.⁵² These irregularities render particle morphology highly unpredictable and conversion to an NO-releasing product with consistent release kinetics challenging.^26,27 The diatomaceous earth used in this work, however, has been previously shown to consist mostly of discrete centric diatom species 4–6 μm in diameter and 10–20 μm in length possessing large openings on either end and highly order rows of 400–500 nm pores.⁵³ Such an ordered, porous structure would be extremely difficult to reproduce by the synthetic means employed in previous studies and affords diatomaceous earth unique physical properties invaluable to the fields of biosensing, filtration, immunoprecipitation, microfluidics, nanofabrication, protein catalysis, and drug delivery.^28,30,53,54

Virtually all previously reported NO-releasing silica particles have been nonporous and on the nanoscale.^{26–28,30,54} Although mesoporous NO-releasing silica particles have been described by Soto et al., the diatomaceous earth particles highlighted herein feature systematic pore and particle structures hundreds of times larger than those reported previously.³⁰ Because porosity and size have been shown to directly affect the loading efficiency, release kinetics, and degradation rates of therapeutic materials, such a substantial divergence from previous particle morphologies stands to broaden the application range and utility of NO-releasing silica.⁵⁵ Specifically, porous silicon microparticles are being explored as “mother ships” capable of carrying therapeutic payloads such as nanoparticles, proteins, enzymes, drugs, and genes.⁵⁶

Energy-dispersive X-ray spectroscopy spectra of both unmodified-DE and SNAP-DE along with chemical mapping of elemental sulfur can be seen in Figure 7. Because SNAP is the only sulfur-containing material used in the synthesis of SNAP-DE, the presence of sulfur in modified DE samples serves as a direct indicator of SNAP presence. The EDS spectra of SNAP-DE clearly indicate the appearance of an elemental sulfur peak and thus the incorporation of SNAP into the sample. The spectra sulfur mapping of SNAP-DE substantiates and visually complements this finding by showing high concentrations of elemental sulfur in the same location as SNAP-DE particles. This suggests that the modified surfaces of DE are responsible for the appearance of a sulfur peak. The appearance of aluminum signals in the spectra is not surprising because the element is a known component of diatomaceous earth.³⁴ Regions of particularly high sulfur intensity in Figure 6d could be the result of smaller, irregularly shaped particles responding better to NAP functionalization, but this has not been evaluated experimentally. Similarly, overlay regions which lack large particles, yet possess sulfur signals may contain nanosized fragments of functionalized DE not visible on the scale of the image.

Figure 7.

EDS spectra of (a) unmodified DE and (b) SNAP-DE. (c, e) SEM images of SNAP-DE. (d) SEM image from (c) with an elemental sulfur mapping overlay. (f) SEM image from (e) with an elemental sulfur mapping overlay. The width of the white scale box is the number located within it.

3.6. Bactericidal Properties of SNAP-DE.

The successful development of new-age therapeutic biomaterials hinges, in large part, on their antibacterial properties. S. aureus, a major source of hospital acquired infections, forms a matrix on substrate surfaces and often results in biofilms resistant to antimicrobial agents such as antibiotics and silver nanoparticles.⁵⁷ Accordingly, S. aureus was selected in this study to serve as a proof-of-concept evaluation of the antibacterial properties of SNAP-DE in biomaterial applications. The bactericidal properties of SNAP-DE were evaluated using an NO concentration of 0.8 μmol NO/mL. Unmodified DE showed a slight reduction in bacterial CFUs as compared to the control, whose bacterial reduction was enhanced to the log scale in the presence of SNAP (Figure 8). The NO releasing SNAP-DE showed 92.95 ± 2.6% bacterial reduction as compared to the positive control sample (bacteria grown directly in the absence of DE or SNAP-DE). A similar bactericidal trend has been observed in the past with other SNAP-based NO releasing materials. Because the presence of nitric oxide-releasing moieties marks the only difference between SNAP-DE and positive and DE controls, bacterial inhibition is attributable to the toxic effects of NO against bacteria.

Figure 8.

A graphical comparison of the viable bacterial colony forming units per mg (CFU/mg) for unmodified DE and SNAP-DE after 24 h of exposure. Bacteria grown without exposure to either DE or SNAP-DE were used as a control. As a proof of concept, Gram-positive S. aureus, one of the major causal agents of biofilm formation and nosocomial infections, was used to test the antibacterial property of SNAP-DE. The bactericidal nature of NO killed the bacteria on a logarithmic scale (a) and resulted in 92.95 ± 2.6% of reduction (b). The data are reported as a mean ± standard deviation for n = 3 samples for (a) and mean ± standard error for (b), and the significance with a p-value <0.05 is stated for comparison.

Nitric oxide kills bacteria via nonspecific mechanisms which involve thiol and amine nitrosation in the extracellular matrices of bacteria, DNA cleavage, lipid peroxidation, and tyrosine nitrosation.⁵⁸ Moreover, unlike antibiotics and silver nanoparticles, bacterial resistance to NO is unlikely to develop due to the molecule’s rapid and nonspecific action.^6,59 In the past, our group and others have shown the antimicrobial effects of NO releasing biomaterials against P. aeruginosa, S. aureus, E. coli, A. baumanni, E. coli, L. monocytogenes, and E. faecalis.^{5,21,24,60,61} In many of these studies, NO donors were either blended or chemically linked to polymers while bacterial growth and inhibition were observed on the surface. The previous success of NO against the aforementioned bacteria, along with the bactericidal effects of SNAP-DE shown directly in this work, suggests that SNAP-DE can be an efficacious and versatile biomaterial both in its own right and as an additive to currently existing technologies. This work marks the first time an NO donor has been covalently linked to diatomaceous earth to form a next-generation NO-delivery scaffold of natural origin.

3.7. Cytotoxic Effects of SNAP-DE.

The in vitro cytotoxicity assay serves as a proof-of-concept evaluation of the material’s biocompatibility. The current study was performed per ISO standards for cytotoxicity using a WST-8 dye based CCK-8 kit (Sigma-Aldrich). Although research groups in the past have shown the antibacterial properties of antibiotics, silver nanoparticles, and NO-releasing materials, the toxic nature of these materials was either not tested or found to be cytotoxic to mammalian cells.^57,62–64 Therefore, evaluating the cytotoxicity of SNAP-DE in addition to its antibacterial properties was a major objective of this work. Mouse fibroblast cells were exposed to 10 μL of SNAP-DE material (1 mg/mL). Results for the test indicated that SNAP-DE possessed levels of fibroblast cell viability similar to those of the control (cells in the cell culture well without any material). No significant (n = 7) differences were found in the cytotoxicity analysis in the presence or absence of SNAP (Figure 9). In addition to the observed viability, the medium color remained red (phenol red indicator), showing that the metabolism of the fibroblast cells did not result in an acidic pH.

Figure 9.

Cytotoxic potential of SNAP-DE leachates in solution was tested on 3T3 mouse fibroblast cells using a WST-8 dye based CCK-8 kit. Cells exposed to the leachates from DE and SNAP-DE demonstrated cell viabilities similar to those of control cells not exposed to leachates, demonstrating the nontoxic effects of the leachates. The data is reported as a mean ± standard deviation for n = 3 samples, and the significance with a p-value <0.05 is stated for comparison.

These results are consistent with previous studies demonstrating the cytocompatibility of both SNAP and diatomaceous earth. Specifically, diatomaceous earth has been utilized in personal hygiene and dietary applications for decades without issue, and SNAP has been shown to be noncytotoxic, biocompatible, and hemocompatible both in vivo and in vitro.^{4,5,53,65–67} Moreover, the major degradation product of SNAP, NAP, has been used in the safe treatment of medical conditions at doses of 2–4 g/day.¹⁴ Many studies have evaluated the toxicity of silica particles synthesized by various means, often with conflicting results.^55,68,69 A general consensus, however, exists that high concentrations of silanol functionalities (≡SiOH) on silica surfaces leads to increased toxicity. It is theorized that surface silanols compromise cell integrity by hydrogen bonding to key membrane components and/or dissociating above pHs of 2–3 to electrostatically interact with positively charged tetraalkylammonium-containing phospholipids.⁶⁹ This silanol associated toxicity stands to be minimized in SNAP-DE through high particle porosity and the encapsulation of silanol groups with amine layers. Porosity reduces the solid fraction of modified silica particles and thus the number of silanols available to negatively affect cell membranes.^68–70 Furthermore, the coverage of silaceous diatomaceous earth with aminosilanes likely forms a protective layer that limits cell accessibility to unreacted surface silanols.⁷⁰ These properties contrast previously reported and largely nonporous NO-releasing silica particles formed by co-condensation methods that homogeneously incorporate silanols throughout particles.^28,30,54

The SNAP-DE showed no toxicity toward mammalian cells, allowing for flexible material testing beyond the concentrations used in this study. Overall, SNAP-DE presents a tunable NO delivery vehicle with effective antibacterial and noncytotoxic properties suitable for graduation to in vivo animal models. A more comprehensive study of SNAP-DE’s cytotoxic properties at higher concentrations both in vitro and in vivo should be carried out in the future to prove the preclinical potential of SNAP-DE.

4. CONCLUSION

In this work, the synthesis and characterization of biotemplated mesoporous nitric oxide-releasing diatomaceous earth were described for the first time. By quantifying primary amine and thiol groups present on the surface of functionalized DE, APTES was shown to maximize NAP attachment and thus NO loading. Sulfhydryl levels of APTES-NAP and nitric oxide levels of APTES-SNAP were found to be 0.0312 ± 0.006 and 0.0372 ± 0.00791 μmol/mg, respectively. Fourier-transform infrared spectroscopy and energy-dispersive X-ray spectroscopy suggested successful modification of DE through the appearance of functional groups and atoms consistent with those of SNAP. Scanning electron microscopy confirmed the retention of diatomaceous earth’s unique morphology throughout the synthesis. Successful SNAP tethering was further demonstrated via sustained NO release over a 24 h period of real-time chemiluminescence measurement. Lastly, SNAP-DE particles were shown to reduce bacterial colonies, delivering a 92.95% relative killing efficiency against Staphylococcus aureus without negatively affecting mammalian cells. The results of this study suggest a promising new biotemplated NO donor system leverageable in antithrombogenic and antimicrobial biomaterials such as catheters, creams, extracorporeal circuitry, biosensors, and device coatings.