Low Dose Inflammatory Potential of Silica Particles in Human-Derived THP-1 Macrophage Cell Culture Studies – Mechanism and Effects of Particle Size and Iron

Abstract

Silica and iron are major constituents in ambient particulate matter, and iron is a common impurity in many engineered nanomaterials. The purpose of this work was to determine the pro-inflammatory and other biological effects and mechanism of particle size and iron presence under low dose, non-cytotoxic conditions that are likely to approximate actual exposure levels, in contrast with higher dose studies in which cytotoxicity occurs. Specifically, human-derived THP-1 macrophages were exposed to 1 μg/ml of pristine and iron-coated 50 nm and 2 μm engineered silica nanoparticles. Particles were first characterized for size, size distribution, surface area, iron concentration, phase and aggregation in cell culture media. Then, biological assays were conducted to determine a non-lethal dose used in subsequent experiments. Superoxide production, lipid peroxidation, and increased pro-inflammatory cytokine (TNF-α and IL-1β) mRNA expression were measured as a function of particle size and iron presence. Smaller particle size and the presence of iron increased superoxide production, lipid peroxidation, and the induction of pro-inflammatory cytokine mRNA expression. Separate addition of an iron-chelator, a scavenger of superoxide and hydrogen peroxide, and an inhibitor of phosphatidylcholine specific phospholipase C (PC-PLC), suppressed the increase in cytokine mRNA expression. Furthermore, free iron itself showed none of the aforementioned effects. The results highlight the importance of particle size and iron in lung inflammation for both natural and engineered nanomaterials, under low dose, non-toxic conditions, and support the role of an oxidant, lipid peroxidation and PC-PLC dependent inflammatory mechanism.

1. Introduction^*

Airborne, micron-sized silica has long been associated with lung diseases [1] and the presence of metals at low concentrations has been shown to contribute to its toxic and pro-inflammatory nature [2–4]. Further, the rapid development of engineered/manufactured nanophases, such as silica, titania, and carbon nanotubes, has added new potential sources of environmental and workplace airborne particulate pollution [5]. Engineered silica particles have desirable material properties and have been widely used for various applications, such as drug delivery, cancer therapies, biosensors, gene delivery, biomedical imaging, and as catalyst supports [6–9], however, their potential toxicity and pro-inflammatory properties merit closer study [10].

Transition metals such as iron are a common component of ambient particulate matter from various sources (street [11–15], street dust [16], automobiles [17–21], brush fire [22,23], coal combustion [24,25] and subway [26–28]). and contribute to the biologic effects of particles with different chemical properties [27, 29–33]. They also are intentionally added to engineered silica particles for biomedical and industrial applications, or are present as a common contaminant in other nanomaterials synthesized via catalytic routes. This raises the question of how iron presence influences the biological effects of silica particles.

Previous studies have also demonstrated a size-dependent toxicity for silica particles. Chen et al. [34] observed that both nano- and micro-sized silica caused pulmonary inflammation in rats. Eom et al., Akthar et al. and Lin et al. [35–37] demonstrated oxidative stress and depletion of glutathione levels in A549 and BEAS-2B cells upon exposure to amorphous silica nanoparticles in a dose- and time-dependent manner; and Guo et al. reported oxidative stress, inflammation, and endothelial dysfunction in HUVECs [38]. Both Napireska et al. [39] and Yu et al. [40] showed a size-dependent cytotoxicity of silica nanoparticles, with surface area and size being important determinants of toxicity, and Kasper et al. [41] determined that lung surfactant augments cytotoxicity. Wottrich et al. [42] reported toxicity for A549 cells treated with synthesized 60 nm silica nanoparticles.

The majority of these previous studies on the biologic effects of silica have mainly focused on engineered silica induced-cytotoxicity at high doses beyond the actual exposure range in humans. Further work is needed to understand the effect of particle size and iron at low non-cytotoxic doses on engineered silica’s pro-oxidative and pro-inflammatory effect. This is a key area to explore, considering the wide use of engineered silica particles in various applications and need for understanding of their biological effects, particularly at doses that mimic human exposure.

Previous work by our lab [43] showed that natural ambient silica particles induced the production of superoxide (O₂⁻⁻), and hydrogen peroxide (H₂O₂) by macrophages. In the Fenton reaction, Fe²⁺ can react with H₂O₂ to produce the hydroxyl radical, which can damage biomolecules such as lipids, proteins, and DNA [3,44,45]. Furthermore, it was demonstrated that phosphatidylcholine-specific phospholipase C (PC-PLC) may be involved in natural, crystalline silica-stimulated induction of TNF-α and IL-1β inflammatory molecules [45]. However, the extent of inflammatory production by engineered amorphous particles could be lower than natural crystalline silica particles, due to silica surface dehydroxylation, as shown by other studies [46].

In this context, the aims of this paper were to a) determine the pro-inflammatory and other biological effects of particle size and iron presence in engineered silica nanoparticles at a low, non-cytotoxic dose, b) obtain a better understanding of the pro-inflammatory signaling consequently induced in macrophages, and c) examine whether the inflammatory pathways observed in cells with engineered silica particles resemble those observed with natural silica particles.

2. Materials and Methods

2.1. Reagents and materials

The engineered silica nano- and micro-sized particles used in this study were spherical amorphous particles with an average particle size of 50 nm and 2 μm, respectively, purchased from Microspheres-Nanospheres (Cold Spring, New York, USA). Materials used were diphenyl-1-pyrenylphosphine (DPPP) TRIZOL reagent (Invitrogen-Life Technologies, Grand Island, NY), Tricychodecan-9-yl-xanthate (D609) (BIOMOL-Enzo Life Sciences, Kelayres, PA), TaqMan reverse transcription reagent, and SYBR Green PCR Master Mix (Applied Biosystems Life Technologies, Grand Island, NY). Manganese (III) tetrakis (N-ethylpyridinium-2-yl) porphyrin (MnP) was kindly provided by Dr. James Crapo, National Jewish Health (Denver, USA). All other chemicals, unless specified, were analytical grade and purchased from Sigma-Aldrich (St. Louis, MO).

2.2. Silica particles and characterization

According to the manufacturer’s specifications, the particle solutions contained spherical silica micro- and nano-particles with an average primary particle size of 2 μm and 50 nm, respectively. Prior to exposure, the particle suspensions were cleaned of synthesis residues, solvents, and lipopolysaccharide contamination by autoclaving the suspensions. Particle suspensions were then concentrated and dried for 24 hours. The final silica material was weighed and stored. The particle sizes, size distribution, morphology, and iron presence were determined using a JEOL 2010 High-Resolution Transmission Electron Microscope (HRTEM) equipped with LaB₆ filament (200kV), FEI CM-200 HRTEM equipped with Field Emission Gun (FEG) source and Gatan Imaging Filter (GIF) with 1k×1k CCD camera and energy resolution of 0.9 eV with 1nm spatial resolution, and FEI Quanta 200 Scanning Electron Microscope (SEM) equipped with a tungsten filament and Energy-Dispersive X-ray Spectroscopy (EDXS) detector with ultrathin window and 132 eV energy-resolution. X-Ray Diffraction (XRD) patterns were collected on a PANalytical X’pert Pro X-Ray Diffractometer to confirm the amorphous phase of 2 μm silica particles. A Selected Area Diffraction (SAD) pattern was collected to confirm the amorphous phase of the 50 nm nanoparticles. Quantitative iron adsorption onto the silica was determined using an Agilent 7500 Inductively Coupled Plasma - Mass Spectrometer (ICP-MS) (detection limit for iron: 1.5 μg/L or ppb), along with colorimetric measurement as described by Ghio et al. [47]. The surface area of the particles was determined by the Brunauer-Emmett-Teller (BET) method. Both 2 μm and 50 nm suspensions were prepared in phosphate buffered saline (1× PBS) and dispersed by sonication to avoid aggregation. In each study, the suspension was freshly prepared, diluted, and then immediately added to the THP-1 cells.

2.3. Imaging of particle-treated cells

The 2 μm silica suspension was prepared in 1× PBS and dispersed by a sonicator to avoid aggregation. The suspensions were freshly prepared, diluted, and then immediately used to treat differentiated THP-1 macrophages for 24 hours. After 24 hours, the cells were imaged with a Nikon C1 confocal microscope.

For TEM sample preparation, cell culture plates were lined with aclar sheets and THP-1 monocytes (2 × 10⁶ cells) were added to each well to undergo differentiation with 50 ng/ml PMA. Following this, cells were treated with 1 and 100 μg/ml of 50 nm silica nanoparticles for 24 hours. Cells were then fixed with 2.5% glutaraldehyde in Phosphate Buffered Saline (PBS), pH 7.4 for 2–3 hours at room temperature and then cooled to 4 °C for 24 hours. Excess glutaraldehyde was removed and cells were washed with PBS buffer for 2–3 times. The culture dishes were then post fixed with osmium tetroxide in PBS buffer for 10 minutes at room temperature and cooled to 4 °C for 24 hours. The osmium tetroxide was then suctioned off and the culture plates were first washed with PBS for 2–3 times and then dehydrated at 50%, 75%, 95% and 100% ethanol (30 minutes each dehydration). The cells were then embedded by using a mixture of resin and propylene oxide in different series (1:1, 3:1 and 100% resin for 30 minutes each). The embedding capsules filled with resin were then inverted and pressed till the resin was in direct contact with the aclar sheets. The sample and the resin in the capsule were then heated in a 60 °C oven for 24 hours to allow the resin to polymerize. Next, the resin block was removed from the aclar sheets by using liquid nitrogen and subjected to thin sectioning using a ultramicrotome. Ultra thin sections (below 100 nm) were cut and placed on copper grids, and then post-stained with uranyl acetate and lead citrate and examined by TEM.

2.4. Iron coating protocol

For iron coating onto the surface of silica, particles were exposed to FeCl₃ dissolved in degassed water. 1.0 ml of 1 mM FeCl₃ solution was added to 1.0 ml of silica (2 mg silica/ml). Suspensions were vortexed, agitated for 24 hours, and centrifuged at 1,200 g for 10 minutes. In order to avoid the formation of oxyhydroxides, water used for dilutions was degassed in argon gas for 24 hours and glassware was washed with 1% HNO₃. The supernatant was then assayed for iron by the method described by Ghio et al. [47]. To quantify the surface iron absorption, standards and samples (0.8 ml) were acidified with 0.1 ml of 25% (wt/vol) trichloroacetic acid; 0.1 ml of 2.5 M potassium thiocyanate was added, and the absorbance was determined at 480 nm. The amount of iron adsorbed was calculated as the difference between the initial and final concentrations in the solution. Engineered silica particles exposed to 1 mM FeCl₃ were washed with distilled water and dried. Measurements were done in triplicate.

2.5. Cell culture

THP-1 cells (Human acute monocytic leukemia cell line) purchased from American Type Culture Collection, TIB-202, were cultured in T25 flasks (Fisher Scientific) with RPMI 1640 culture medium supplemented with 10% Fetal Bovine Serum (FBS) (Omega Scientific), β-mercaptoethanol, 100 μg/ml penicillin, and 100 μg/ml streptomycin at 37 °C in a humidified 5% CO₂ atmosphere. In recent years, these cells have been widely established as an in vitro model for native monocyte-derived macrophages in studies of inflammatory disease by particle inhalation [48,49]. Prior to the experiments, the monocytes were differentiated into adherent macrophages by treatment with 50 ng/ml phorbol 12-myristate-13-acetate (PMA) for 3 days.

2.6. Determination of cell toxicity with 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazoliumbromide (MTT) assay

THP-1 cells were differentiated at a density of 3×10⁵ cells/ml per well in 12-well tissue culture plates. After 22 hours of particle exposure, percent viability was quantified by incubating the cells for 2 hours at 37 °C with MTT (5 mg/ml). The resulting colored product was solubilized in MTT solubilization solution (10% Triton-X 100 and 0.1N HCl in acidic isopropanol). The absorbance was measured at 550 nm and the background absorbance was measured at 690 nm using a microplate reader (Molecular Devices, Sunnyvale, CA).

2.7. Determination of superoxide production

To determine superoxide production, a modified colorimetric nitroblue tetrazolium (NBT) assay was performed as described by Rook et al. [50]. THP-1 monocytes (3×10⁵/well) were differentiated into macrophages in a 12-well culture plate. After pretreatment with 2.5 mM NBT solution in 37° C Krebs Ringer Phosphate Buffer (KRP) for 30 minutes, exposure to 0.05, 0.1, and 1 μg/ml of 2 μm and 50 nm silica and iron coated silica particles was carried out. 50 ng/ml PMA was used as a positive control. Cells were washed with warm PBS to remove the extracellular NBT and then air-dried. The amount of NBT deposited inside the cells was then quantified by adding 120 μL 2 M potassium hydroxide (KOH) and 140 μL dimethyl sulfoxide (DMSO) to dissolve the cell membrane. The resulting blue formazan was mixed and transferred to a 96-well plate to read the absorbance at 360 nm with a plate reader.

2.8. Lipid peroxidation assay with diphenyl-1-pyrenylphosphine (DPPP) fluorescence

THP-1 (100,000 cells/well) monocytes were differentiated into macrophages in a 96-well culture plate. Cells were incubated with 5 μM DPPP dissolved in 0.05% DMSO for 15 minutes in the dark and washed twice with 1× PBS buffer to completely remove extracellular DPPP. After an exposure of 1 μg/ml of 2 μm and 50 nm silica and iron coated silica particles for 15 minutes, the increased emission fluorescence intensity of DPPP oxide (DPPP=O), was measured at an excitation of 351 nm and an emission of 380 nm with a plate reader. As a positive control, 100 μM cumene hydroperoxide was added.

2.9. Quantification of mRNA with real-time polymerase chain reaction (Real-time PCR) assay

RNA extraction, cDNA reverse transcription, and measurement of mRNA levels and calculation of mRNA levels relative to control were performed according to a procedure as described before [51]. The primer sequences for real time PCR assay are as described previously [43].

2.10. Statistical analysis

All data were expressed as the mean ± standard deviation. One-way ANOVA and one-tailed unpaired student’s t-test were used for significance testing, using *p < 0.05. All the experiments were conducted with replicates, so the data are the means of three experiments, each using three experimental points, with control values subtracted from that of the sample, when compared to uncoated silica.

3. Results

The biological response induced by particles is dictated by several factors, including size, surface area, morphology, phase, charge, and chemical composition [52]. The focus of this study was the relationship between size and the presence of surface iron in well-controlled engineered silica particles, and the production of inflammatory mediators and the role of oxidative signaling in it at a low non-cytotoxic dose. Consequently, particles were characterized by multiple techniques and a range of bioassays were carried out, as detailed below.

3.1. Particle characterization

The physical properties of micro- and nano- engineered silica particles used in this study are summarized in Table 1, and in Figures 1–3. The mean size and size distribution of the “as received/uncoated” particles were observed by Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM) by measuring the diameter of about 400 particles in the micrographs (Table 1; Figures 2a, 2b). The results were consistent with the measurements provided by the suppliers, within reported variances. The surface areas for 2 μm and 50 nm particles were evaluated by Brunauer-Emmett-Teller (BET) method (Table 1). Dynamic Light Scattering (DLS) measurements conducted for the 50 nm particles in the presence or absence of RPMI 1640 media after sonication indicated that their hydrodynamic diameters were not larger than the size range estimates measured by TEM (Table 1).

Table 1.

Engineered silica 2 μm and 50 nm nanoparticle characteristics. Sizing data were obtained using SEM, TEM, and DLS. Iron concentration and surface area were measured by ICP-MS and BET analysis.

Engineered Silica Size	Primary Diameter	Hydrodynamic Diametera	Hydrodynamic Diameterc	BET Surface Area	Iron Concentration	Surface Iron Concentration
				(m2/mg)	(μMol/g silica)	(Mol/m2 silica)
2 μm	2.05±0.18 μm	n.d.b	n.d.	0.00136	34.01	0.027
50 nm	41.53±10.63 nm	95±10 nm	138±15 nm	0.0545	54.03	0.0099

^aMean (±SD) DLS measurement conducted in PBS

^bData determined with optical microscopy images

^cMean (±SD) DLS measurement conducted in PBS plus cell culture media

Figure 1. Engineered 2 μm silica characterization.

(a) EDX pattern and (b) SEM micrograph, and (c) XRD pattern of 2 μm engineered silica. Particle size and morphology were found to be uniform, no measurable iron was observed, and the phase was found to be amorphous.

Figure 3. Engineered 50 nm Silica characterization.

(a) EDX pattern, (b) TEM micrograph and (c) selected area diffraction of 50 nm engineered silica. Particle size and morphology were found to be uniform, while iron contaminants were below the detection limit and the phase was found to be amorphous.

Figure 2. Particle characterization of size distribution and aggregation.

(a) 2 μm and (b) 50 nm silica particle size distributions as determined by AnalysisPRO software using SEM micrographs. Optical micrographs of 2 μm engineered silica in (c) PBS and (d) cell culture media. Particles were well dispersed in the buffer, with few aggregates observed in the media.

Brunner et al. [53] have discussed and showed the need for using reliable methods to observe particle dispersions as they can cause discrepancies in data interpretation. In the present study, particles were characterized before and after dispersion in the cell culture media and just prior to cell exposure. Due to the shorter exposure times used in this study, in the cell culture media and sonication prior to it, no dispersing agents were used since no aggregate formation was observed. Uniform dispersal of particles allowed for control over determining size-dependent effects (Table 1; Figures 1b, 2c, 2d, 3b).

The phase of the 2 μm and 50 nm particles was determined to be amorphous, as expected and clearly shown from the X-ray Diffraction (XRD) broadened peaks and diffuse or halo electron diffraction pattern [54] (Figures 1c, 3c). The use of characterized silica particles with uniformity in size and distribution provided a controlled platform for data comparison.

Batches of 2 μm and 50 nm particles were coated with iron and comparisons were made with their respective uncoated silica particles. The concentration of iron, coated onto the surface of the silica particles, was measured by both calorimetric assay and Inductively Coupled Plasma Mass spectroscopy (ICP-MS). Iron was mainly chosen as the metal dopant to silica as it is a common metal added to (i.e., for catalysis) or found as a contaminant in nanomaterials, and is capable of redox cycling [55]. In addition, iron impurities are also present in the raw materials or precursors used for particle synthesis. Procedures for iron-oxide coating of silica particles are well established and generally involve precipitation of the Fe³⁺-oxide phase from aqueous solution. The calorimetric measurements for iron concentrations were calculated by a method described by Ghio et al. [47], and for 2 μm and 50 nm silica particles, were 34.01 μM and 54.03 μM of iron per gram of silica, respectively. The ICP-MS iron concentrations for 2 μm (36.75 μM/g silica) and 50 nm (52.34 μM/g silica) coated particles were comparable to the calorimetric data. As a function of surface area, the iron concentration was greater for the 2 μm particles (0.027 mol/m² of silica versus the 50 nm particles (0.0099 mol/m²). (Moles of iron were normalized.) Both 2 μm and 50 nm silica particles contained significant amounts of complexed iron on the surface from the FeCl₃ solution,_, although 50 nm particles complexed more Fe³⁺ than 2 μm particles due to their larger surface area to volume ratio. HRTEM and Scanning Transmission Electron Microscopy (STEM) images of iron coated 50 nm silica particles are shown in Figure 4, where Electron Energy-Loss Spectroscopy (EELS) confirms that iron precipitates as 5 nm particles on the surface of silica. In comparison, based on the manufacturer’s data and ICP-MS analysis, iron on the uncoated 2 μm and 50 nm particles was below detectable limits. Iron was also not detected on uncoated 2 μm and 50 nm particles when measured by Energy-dispersive X-ray spectroscopy (EDX) analysis (Figures 1a, 3a).

Figure 4. Iron-doped Engineered 50 nm Silica characterization.

(a) HRTEM image, (b) STEM image and (c) nanoprobe EELS confirming ~5 nm iron oxide particle deposition on silica.

3.2. Imaging of Particle-Treated Cells

Optical microscopy (Figure 5) shows that at the low doses used in this study of 1 μg/ml, 2 μm silica particles were actively taken up by the macrophages, mostly around the membrane periphery and some in the cell cytosol. At a higher dose of 100 μg/ml that is more typically used in toxicity studies, macrophage over burden can be observed. TEM images (Figure 6) reveal 50 nm silica uptake at different stages. Particles are observed interacting with the cell membrane, inside phagosomes, inside phagosomes in the process of fusing with lysozomes, and interestingly, in the cell cytosol.

Figure 5. 2 μm silica particle uptake by THP-1 macrophages.

2 μm silica particles were actively taken up by the macrophages, mostly around the membrane periphery and some in the cell cytosol. Macrophage over burden; excess phagocytosable particles, can be observed when cells were treated with 100 μg/ml (C) of silica as compared to 1 μg/ml (B) of silica. White arrows indicate silica particles. Red arrow indicates macrophage overload.

Figure 6. 50 nm silica particle uptake by THP-1 macrophages at different stages as studied by TEM.

A) Numerous phagosomes with nanoparticle aggregates encapsulated within them indicated by the arrow. B) Some particles are evident outside phagosomes in the cytosol. C) Particles inside fused phagosomes with lysosomes. D) Silica nanoparticle-cell membrane surface interaction. E) Depicts the fusion of lysosome with phagosome-containing particles. F) Shows the interaction of the macrophage pseudopods with silica nanoparticles.

3.3. Determination of non-cytotoxic dose

The non-cytotoxic dose of 2 μm and 50 nm silica particles was determined by 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay (Figure 7). For 2 μm particles, doses of 50 μg/ml and below did not differ significantly from the control (no particles), while for 50 nm particles, doses of 10 μg/ml and below did not differ significantly from the control sample. To ensure that a non-cytotoxic dose was used, 1μg/ml was used in all subsequent experiments for both particle sizes.

Figure 7. Determination of non-cytotoxic silica concentrations.

MTT assay was performed to determine the non-cytotoxic exposure levels of engineered (a) 2 μm and (b) 50 nm silica particles. The cell viability was determined as a percentage control. Values are mean ± SD from three independent experiments. Significance indicated by *p < 0.05 when compared to control (no particles).

3.4. Effect of size on engineered silica in O₂⁻⁻ and lipid peroxidation production

In the first set of experiments, both 2 μm and 50 nm particles, along with their respective iron coated versions, were investigated for their stimulation of O₂⁻⁻ in cultures of human monocytes-derived macrophages in vitro. SOD-inhibitable NBT formazan absorbance, indicative of O₂⁻⁻ production in the cells, doubled significantly in macrophages after a 10 minute exposure to uncoated 50 nm silica particles, when compared to uncoated 2 μm particles of the same mass concentration of 1 μg/ml (Figure 8). The O₂⁻⁻ production was 2- and 3-fold higher in macrophages exposed to 2 μm and 50 nm particles than in the control cells (no particles). At lower doses of 0.05 and 0.1 μg/ml, the difference was not significant. Levels of O₂⁻⁻ observed were comparable to 50 ng/ml PMA, a known stimulus of the respiratory burst; i.e. stimulated O₂⁻⁻ production [56]. No comparisons of size-dependent effects of O₂ production from the iron-coated batches of the micro- and nano-sized particles were performed due to the difference in their iron concentrations. When expressed on a mass concentration (μg/ml) basis, the dose-.response relationship for O₂⁻⁻ production supported a size-dependent effect, with increased production as particle diameter decreases (Figure 8).

Figure 8. Iron presence and reduction in particle size increases O₂⁻⁻ production.

Effect of size and surface iron of engineered silica (a) 2 μm and (b) 50 nm particles on O₂⁻⁻ production after 10 minutes at low non-cytotoxic concentrations. The absorbance for control (no particles) was subtracted from the sample absorbance values. Values are the mean ± SD from three independent experiments. *p < 0.05 when compared to uncoated silica.

As in the O₂⁻⁻ production results, lipid peroxide production showed a size-dependent effect. Lipid peroxidation after 15 minutes of exposure to a low non-cytotoxic dose of 1 μg/ml was significantly increased for 50 nm uncoated particles over the same mass concentration of 2 μm particles (Figure 9). As a positive control, cumene hydroperoxide (100 μM) was used as demonstrated in previous studies [57]. To establish the relationship between production of O₂⁻⁻ and lipid peroxidation, the cells were pretreated with 100 μM manganese (III) tetrakis (N-ethylpyridinium-2-yl) porphyrin (MnP). MnP is a superoxide dismutase (SOD)/catalase mimic that scavenges both O₂⁻⁻ and H₂O₂ that is formed by O₂⁻⁻ dismutation [58]. MnP significantly prevented lipid peroxidation in the macrophages caused by silica particles.

Figure 9. Iron presence and reduction in particle size increases lipid peroxidation production.

Effect of size and surface iron of engineered silica (a) 2 μm and (b) 50 nm on peroxidation production. To determine the involvement of H₂O₂ in lipid peroxidation production, cells were also separately pretreated with 100 μM MnP, a SOD/catalase mimic, for 15 minutes, prior to silica treatment (10 minutes). The absorbance for control (no particles) was subtracted from the sample absorbance values. CHO was cumene hydroperoxide, which was the positive control. Values are the mean ± SD from three independent experiments. Significance indicated by *p < 0.05 when compared to uncoated silica.

3.5. Effect of surface iron on engineered silica in O₂⁻⁻ and lipid peroxidation production

Previous investigators including Limbach et al. [59] have shown that iron-containing nanoparticles promoted oxidant production within human lung epithelial cells. In this study with human derived macrophages, even the presence of low levels of iron on the 2 μm and 50 nm particles, 34.01 μM/g and 54.03 μM/g of iron/g respectively, increased O₂⁻⁻ production significantly by THP-1 cells, compared to cells without any particles and cells with uncoated particles (Figure 8). The difference in the iron loading onto the surfaces of nano- and microsized particles as seen in Table 1 could be mainly because of the increase in total surface area as particle size is decreased while the total particle mass remains the same [60]. This variance, however, did not affect the current study as the main focus was to investigate the size-dependent effects of as received/uncoated engineered silica, and to understand the influence of metals on both engineered nano- and micro-sized silica’s pro-inflammatory effects individually. As in the results with size dependency, the presence of iron on the micro- and nanoparticles leads to a higher O₂⁻⁻ production by cells in contrast to their counterparts. To investigate whether the presence of iron or change in particle diameter also caused lipid peroxidation, lipid hydroperoxides were quantified using DPPP, which is specific to lipid hydroperoxides and does not react with hydrogen peroxide [61]. Iron coated on the 2 μm and 50 nm particles caused a 2- and 3- fold increase in lipid peroxidation over the control, as compared to uncoated 2 μm and 50 nm particles that caused a 1.3 and 1.7-fold increase over the control (Figure 9).

3.6. PC-PLC dependent iron mediated-engineered silica-induced pro-inflammatory cytokine mRNA expression

To investigate the mechanism of particle-induced pro-inflammatory cytokine mRNA expression, relative mRNA levels of TNF-α and IL-1β were measured by Real-time PCR. The levels of mRNA expression of both cytokines were higher in THP-1 cells within 3 hours of exposure to 1 μg/ml of 2 μm and 50 nm silica particles, but the expression levels did not increase at further time points of 6 and 12 hours (data not shown). As in the studies of O₂⁻⁻ and lipid peroxidation, in the presence of coated iron and a smaller particle size (50 nm), both of the inflammatory mediators elevated strikingly as shown in Figure 10. Addition of the iron chelator DTPA, along with engineered coated 2 μm and 50 nm silica, significantly lowered the induction of both cytokine’s mRNA (Figure 10). Lipopolysaccharide (LPS) was used as a positive control as it is a known inducer of cytokine mRNA expression. Pretreatment with MnP (100 μM) decreased TNF-α and IL-1β induction by silica (1 μg/ml) exposure in the macrophages, showing the dependence of cytokine mRNA induction on O₂⁻⁻ generation.

Figure 10. Iron presence and reduction in particle size increases cytokine mRNA expression via PC-PLC.

PC-PLC dependent iron mediated-engineered silica-induced inflammatory mediator mRNA expression after 3 hours. A similar response was observed even after 6 and 12 hour treatments (data not shown). (a) and (b) represent IL-1β and TNF-α response induced by exposure to engineered 2 μm silica. (c) and (d) represent IL-1β and TNF-α response induced by exposure to engineered 50 nm silica. Values are the mean ± SD from three independent experiments (*p < 0.05). Separately, cells were also pretreated with 1 μM DTPA (30 minutes) before the addition of silica. The relative mRNA levels of IL-1β and TNF-α were determined with Real-time PCR. Similar to O₂⁻⁻ and lipid peroxidation, expression of both cytokine mRNAs was amplified significantly in the presence of iron. Addition of chelator DTPA decreased this effect. PC-PLC dependence was determined by pre-treating the cells with 40 μM D609 followed by addition of 1 μg/ml of natural silica for 6 hours. Additionally, cells were also pretreated with 100 μM MnP for 1 hour before silica treatment. Both MnP and D609 decreased the silica initiated cytokine production significantly (*p<0.05).

To determine the involvement of PC-PLC, which was previously shown to be associated with natural silica-induced cytokine mRNA expression by our lab [43], the macrophages were pretreated with 40 μM D609 (tricyclodecan-9-yl-xanthogenate), a PC-PLC inhibitor, for 10 minutes prior to exposure to 1 μg/ml of iron coated 50 nm and 2 μm engineered silica particles. This inhibited engineered silica-induced IL-1β and TNF-α transcription in THP-1 cells (Figure 10 A, B, C, D), indicating likely PC-PLC involvement.

4. Discussion

In the past decade, there has been an increase in the number of studies reporting nanoparticle toxicity and oxidative stress [28, 59, 62, 63]. However, to date very few studies have addressed the cellular consequences of oxidative challenge leading to inflammation and antioxidant adaptation at non-cytotoxic doses. Using a high toxic dose could overwhelm macrophages and impair their phagocytic function, gene expression, and even cause cell death [64–66]. Indeed, for 2 μm silica particles we observed macrophage overburden at a higher dose (100 μg/ml) more typically used in the literature vs. the dose of 1 μg/ml used here (Figure 4); and for 50 nm silica, some particles are observed outside the cytosol where they can interact with cell organelles, even at a low dose of 1 μg/ml. Our bioassay results with engineered micro- and nano-sized silica particles overall show that at a low non-cytotoxic dose of 1 μg/ml, production of O₂⁻⁻, lipid hydroperoxide and cytokine increased in macrophages in a size- and iron-dependent manner.

To determine the upstream micro- and nano-sized engineered silica-macrophage interaction mechanism, the peroxidation of lipid membrane, increased mRNA expression of inflammatory mediators (IL-1β and TNF-α), and role of PC-PLC enzyme were investigated and comparisons drawn with natural silica particle-induced inflammatory response, which we had previously studied using the same cell model and parallel exposure protocols [43]. A non-cytotoxic dose of 1 μg/ml of silica was selected to study the inflammatory effects of both 2 μm and 50 nm silica particles, based on MTT reduction assay results (Figure 7), and these results were consistent with earlier reports for onset of cytotoxicity of engineered silica particles between the range of 10–100 μg/ml [39–41]. This dose was similar to that used for our earlier studies of natural silica [43] and also relatively close to the realistic exposure levels of natural silica in ambient particulate matter (PM) as shown by Paur et al. [67]. Studies by Davis et al. [68] and Chow et al. [69] found crystalline silica levels in ambient PM in 25 US metropolitan cities and the San Joaquin Valley, both urban and rural areas, ranging from 0.2 to 10 μg/m³. To relate the approximations of the ambient levels to that of in vitro studies, Paur et al. [67] used the following parameters: air inhaled/day - 25 m³, concentration of particles in air - 10 μg/m³, lung surface area - 100 cm², and average deposition efficiency of 30%. Based on these parameters, they estimated that for a particle of 100 nm size, with an average surface area of 10 m²/g, the realistic average dose per day would be closer to 7.5 × 10⁻⁴ μg/cm² of the lung surface area and a lifetime dose would be closer to 6.6 μg/cm² in a non-occupational setting for a 100 nm particle. This lifetime dose (80 years) corresponds to the one-time doses (100 μg/ml to 1 mg/ml) that are delivered to cells in most in vitro studies. Therefore, such studies might not correlate well with realistic exposure scenarios, and may not accurately indicate the particle physiochemical properties and biological response mechanisms relevant under conditions of chronic exposure, where inflammation is likely to play a key role. To study the effects of nanomaterial particulates on the respiratory system, it is essential to perform experiments with low, non-cytotoxic doses that more closely resemble realistic exposure levels.

Exposure to 50 nm uncoated silica particles significantly increased O₂⁻⁻, lipid peroxide and pro-inflammatory cytokine mRNA expression over 2 μm uncoated particles, which is expected due to the larger surface area per unit mass of the 50 nm particles, indicating an influence of particle size on the biological effects (Figures 8, 9, 10). Previous toxicological studies have indicated a possible relation between particle size and cytotoxicity [70, 71]. This study shows that even at a low non-cytotoxic mass dose of 1 μg/ml there is evidence of particle size influence over oxidative stress, lipid peroxidation and inflammation. The evidence of particle size dependency was prominent, as even with the lower surface iron concentration for 50 nm coated particles, these particles resulted in higher productions compared to the 2 μm particles (Figures 8, 9, 10). Our group and others in the past have observed that the presence of iron in natural silica particles exacerbates reactive oxygen species production [40, 72, 73]. The findings in Figure 8 confirm that engineered silica particles follow a similar behavior to natural silica particles with respect to the influence of iron on O₂⁻⁻ production. The iron coated micro- and nano-sized silica particles also showed a statistically significant increase in lipid peroxide and cytokine (IL-1β and TNF-α) mRNA expression, over uncoated silica particles, their counterparts (Figures 9 and 10).

The role of iron in the biological effects of silica particles was further confirmed with the finding that the iron chelator DTPA reduced the induction of pro-inflammatory cytokines caused by iron-coated silica (Figure 10). This data suggests that iron present on the silica particle exacerbates the O₂⁻⁻, and thus H₂O₂ production in the macrophages. Further disruption of lipid rafts in the macrophages could be due to lipid peroxidation stemming from the production of hydroxyl radical (HO·) through Fenton reaction (Fe²⁺ + H₂O₂ → Fe³⁺ + OH⁻ + HO·) as shown in our previous work [43]. The disruption of lipid rafts, which occurs with no cytotoxicity, stimulates a signaling pathway through PC-PLC activation that results in increased cytokine gene transcription [43]. To further clarify the possible upstream oxidative signaling pathway, the cultured macrophages were pretreated with a O₂⁻⁻ and H₂O₂ scavenger, MnP. Interestingly, pretreatment with MnP decreased the subsequent lipid peroxide and pro-inflammatory cytokine mRNA expression in the macrophages caused by both 2 μm and 50 nm silica particles (Figures 9 and 10). This was similar to the previous observation with natural silica. Further studies were conducted to elucidate the mechanism between lipid peroxidation and pro-inflammatory mediator production. Our previous work with natural silica showed that PC-PLC enzyme is required for the induction of cytokine mRNA expression [43, 45]. Others have shown that PC-PLC acts through nuclear factor-kappa B (NF-κB) signaling pathways, which are known to be involved in the up-regulation of key inflammatory factors [74, 75]. D609, a relatively specific PC-PLC inhibitor [76, 77], inhibited both micro- and nano-sized silica-induced IL-1β and TNFα transcription (Figure 10). Further work is, however, warranted to understand the PC-PLC activation from lipid peroxidation products. Involvement of lipid raft disruption is likely involved [43].

5. Conclusion

The present investigation has demonstrated that engineered silica particles at low non-cytotoxic doses exhibit a strong particle size and iron-dependent lipid peroxidation-mediated inflammatory response in THP-1 human macrophages. Particle size, along with iron, aggravated pro-inflammatory mediator production even at a low non-cytotoxic dose of 1 μg/ml. Furthermore, the results demonstrate that as in exposure to natural silica, increased engineered silica-induced cytokine (TNF-α, IL-1β) mRNA expression is induced through an iron-dependent activation of the PC-PLC signaling pathway.

This study also helps to highlight some key particle parameters triggering the underlying pro-inflammatory mechanism that are largely unknown for engineered particles at a dose that avoids cell death and alteration of gene expression levels. Elucidation of this pathway will provide useful information to assess hazards for people and workers exposed to engineered silica in day-to-day and occupational scenarios, as well as in pharmacotherapy. It also provides additional analysis from a more controlled (with respect to composition and size) materials system for assessing the effect of surface transition metals on particle-induced inflammation.

Highlights.

• Engineered and natural silica signal same PC-PLC dependent inflammatory pathway.

• Fifty nm silica induces pro-inflammatory response at lower dose than 2 μm silica.

• Non-toxic pro-inflammatory response to silica nanoparticles is iron-dependent.

• Soluble iron does not induce the pro-inflammatory response of iron on silica.