Skip to main content
NCBI home page
Toxicology Research logoLink to Toxicology Research
. 2016 Feb 26;5(3):836–847. doi: 10.1039/c5tx00421g

Absence of cytotoxicity towards microglia of iron oxide (α-Fe2O3) nanorhombohedra†

Crystal S Lewis a,‡, Luisa Torres b,‡, Jeremy T Miyauchi b,‡, Cyrus Rastegar b, Jonathan M Patete a, Jacqueline M Smith a, Stanislaus S Wong a,c,✉, Stella E Tsirka b,✉
PMCID: PMC4890976  NIHMSID: NIHMS764529  PMID: 27274811

graphic file with name c5tx00421g-ga.jpgWe evaluated the cytotoxicity of iron oxide nanorhombohedra towards microglia cells, and found little to no production of either TNFα, ILIβ, NO, or reactive oxygen species.

Abstract

Understanding the nature of interactions between nanomaterials, such as commercially ubiquitous hematite (α-Fe2O3) nanorhombohedra (N-Rhomb) and biological systems is of critical importance for gaining insight into the practical applicability of nanomaterials. Microglia represent the first line of defense in the central nervous system (CNS) during severe injury or disease such as Parkinson's and Alzheimer's disease as illustrative examples. Hence, to analyze the potential cytotoxic effect of N-Rhomb exposure in the presence of microglia, we have synthesized Rhodamine B (RhB)-labeled α-Fe2O3 N-Rhomb, with lengths of 47 ± 10 nm and widths of 35 ± 8 nm. Internalization of RhB-labeled α-Fe2O3 N-Rhomb by microglia in the mouse brain was observed, and a dose-dependent increase in the cellular iron content as probed by cellular fluorescence was detected in cultured microglia after nanoparticle exposure. The cells maintained clear functional viability, exhibiting little to no cytotoxic effects after 24 and 48 hours at acceptable, physiological concentrations. Importantly, the nanoparticle exposure did not induce microglial cells to produce either tumor necrosis factor alpha (TNFα) or interleukin 1-beta (IL1β), two pro-inflammatory cytokines, nor did exposure stimulate the production of nitrites and reactive oxygen species (ROS), which are common indicators for the onset of inflammation. Finally, we propose that under the conditions of our experiments, i.e. in the presence of RhB labeled-α-Fe2O3 N-Rhomb maintaining concentrations of up to 100 μg mL–1 after 48 hours of incubation, the in vitro and in vivo internalization of RhB-labeled α-Fe2O3 N-Rhomb are likely to be clathrin-dependent, which represents a conventional mechanistic uptake route for most cells. Given the crucial role that microglia play in many neurological disorders, understanding the potential cytotoxic effects of these nanostructures is of fundamental importance if they are to be used in a therapeutic setting.

1. Introduction

Nanomaterials, comprising nanoscale structures measuring between 1 and 100 nm in size, have attracted significant research interest due to their unique structure-dependent physical properties. Recently, concerns have been raised over the potentially deleterious effects of these nanomaterials on human health and the environment.1–3 From a toxicological perspective, nanoscale materials can induce different types of cellular responses, characterized by a variety of distinctive uptake mechanisms, such as endocytosis, mediated for example by receptor-specific target sites.4–6

For a given nanomaterial, morphology (e.g. in terms of its size and shape) is thought to be one of the key factors that can decisively determine the observed degree of its cytotoxicity and cellular uptake. Indeed, significant effort, including from one of our groups in particular, has been involved with systematically synthesizing novel motifs of diverse classes of nanomaterials, such as but not limited to derivatized carbon nanotubes (CNTs), rare earth ion-doped cerium phosphate (CePO4) nanowires, silicon dioxide (SiO2) nanotubes, titanium dioxide (TiO2) nanostructures, and zinc oxide (ZnO) nanowires and nanoparticles, to analyze their potential for biomedical applications. The objective of that prior body of work had been to correlate size, shape, morphology, and chemical composition of nanomaterials with their corresponding uptake mechanisms in an effort to probe and understand their individual and collective impact upon cellular toxicity, in general.7–14 In effect, we had been interested in determining the specific factors that control nanoscale toxicity.

The model system we study herein is related to a family of magnetic iron oxide (Fe3O4) nanostructures that has already been well studied. Indeed, nanoparticulate magnetite have previously been extensively investigated for incorporation into diverse applications, including for biological fluids, tissue-specific release of therapeutic agents, anti-cancer drug delivery systems, hyperthermia, and contrast enhancement for magnetic resonance imaging (MRI).15–18 In this context, the study of their potential toxicology to cells has served as a valuable means of gauging the viability, biocompatibility, and overall practicality of this magnetic iron oxide platform for ubiquitous use in these assorted contexts.19 Nevertheless, the use of Fe3O4 for biomedical applications has been limited by issues associated not only with particle inhomogeneity and cost concerns but also with its inability to effectively differentiate between tumors and artifacts arising from bleeding, metal deposits, and/or calcification in T2-weighted MRI images.20

A common, companion material to Fe3O4, i.e. hematite (Fe2O3), possesses a rhombohedral crystal structure with a R3c space group.21 However, unlike Fe3O4, hematite can exist in different crystallographic forms such as alpha-hematite (α-Fe2O3), beta-hematite (β-Fe2O3), gamma-hematite (γ-Fe2O3), and epsilon-hematite (ε-Fe2O3), with α-Fe2O3 and γ-Fe2O3 as the most familiar motifs. In particular, α-Fe2O3 has been synthesized as different morphologies, including as particles, cubes, and rods, and has been incorporated as functional components of gas sensors, CO oxidation catalysts, lithium-ion batteries, and colloidal mediators for hyperthermia treatment.22–26 In a number of these aforementioned applications,27 the α-Fe2O3 nanoparticles have been employed as particulate, aerosolized motifs. Therefore, it is imperative to understand the potential toxicological effect of exposure to nanoscale hematite, as manifested by different intake routes such as inhalation, ingestion, and injection.

Nevertheless, the intrinsic toxicity of Fe2O3 nanostructures still remains a matter of considerable controversy. For example, in vitro studies have shown that α-Fe2O3 nanoparticles larger than 90 nm in diameter (i.e. ∼250 nm and ∼1.2 μm) gave rise to little if any toxicity with respect to human lung epithelial cells (A549) and murine alveolar macrophages (MH-S).28 Moreover, α-Fe2O3 nanotubes, characterized by ∼200 nm diameters, were found to be compatible with rat adrenal medulla cells (PC12), and in fact served as a potential delivery vehicle for nerve growth factor (NGF) in order to convert these cells into neurons.29 By contrast, animal studies using Fe2O3 nanoparticles have revealed that these nanostructures may detrimentally induce either airway inflammation in healthy mice or cellular reduction in alveolus and lymph nodes in allergic mice.30

According to Brunauer–Emmett–Teller (BET) analysis, iron oxide nanorhombohedra (N-Rhomb), normalized for geometric considerations, possess a higher surface area (∼45 m2 g–1) than either nanocubes (∼13.5 m2 g–1) or nanorods (∼39 m2 g–1), depending on their size.22,31 Therefore, since the surface area of α-Fe2O3 N-Rhomb is second only to that of spherical nanoparticles (∼133 m2 g–1), which have already been extensively explored in cytotoxic analysis, this observation provides us with a rationale to fully understand the shape dependence of α-Fe2O3 N-Rhomb's interaction with cells, especially when engulfed.32 Moreover, with various reports on the shape-dependent cytotoxic behavior of nanowires versus nanoparticles under various cellular conditions,33 it is therefore necessary to gain a similar insight into the analogous effects of rhombohedral α-Fe2O3 in a biological context. In terms of a prior report with comparable objectives to our own, it is worth noting that studies involving LiNbO3 nanorhombohedra have suggested that these nanostructures maintained cell viabilities of ∼80% after 48 hours of incubation within mouse macrophage cells.34

Prior size and morphology-specific studies of various nanoparticles have indicated the ability of these nanoscale sized motifs to cross the blood brain barrier and thereby enter the central nervous system (CNS) of higher order biological organisms, such as mammals.35,36 In light of this result, many metal oxides such as Fe2O3 and TiO2 have been previously probed for possible neurotoxic effects upon exposure.37–39 With α-Fe2O3 N-Rhomb's small size (<100 nm), large surface area, and chemical stability, nanoscale hematite possesses significant potential to overcome the challenges associated with passage through the blood brain barrier. Therefore, it represents an excellent system with which to probe cytotoxic effects associated with exposure of the CNS to nanomaterials, particularly nanostructures that have been surface modified through the attachment of different specific and judiciously chosen moieties.40

In this work, herein, we have synthesized visually traceable, dye-conjugated nanostructures, i.e. Rhodamine B (RhB)-labeled α-Fe2O3 N-Rhomb. Subsequently, we tested their uptake and possible toxicity in a key model system, i.e. microglia, the immuno-competent cells associated with the CNS. These cells are implicated in the pathology of many CNS disorders, including Alzheimer's disease, spinal cord injury, multiple sclerosis, Parkinson's disease, and ischemia.41–47 Hence, given the central function and critical biological importance of microglia, our objective herein has been to understand the implication of their exposure to our α-Fe2O3 rhombohedral nanostructures in order to determine any potential cytotoxic effects. To the best of our knowledge, our use of α-Fe2O3 N-Rhomb to assess the distinctive role of the rhombohedral shape (and associated surface area) in the context of cytotoxicity has not been previously demonstrated in the literature. Herein, we reveal that microglia successfully incorporate RhB-labeled α-Fe2O3 N-Rhomb under both in vivo and in vitro conditions. More importantly, we definitively demonstrate that incubation with these nanoscale metal oxides at physiological concentrations in fact do not induce either cellular toxicity or inflammatory reactions in cultured microglia.

2. Experimental procedures

X-ray diffraction (XRD)

The crystallographic purity of as-prepared α-Fe2O3 N-Rhomb was confirmed using powder XRD. To prepare a typical sample for analysis, a fixed quantity was dispersed in ethanol and sonicated for ∼1 min, prior to deposition onto a glass slide. Diffraction patterns were subsequently obtained using a Scintag diffractometer, operating in the Bragg configuration using Cu Kα radiation (λ = 1.54 Å) and with 2θ lattice parameters, ranging from 20 to 70° at a scanning rate of 0.25 degrees per minute for variously sized α-Fe2O3 N-Rhomb.

Primary microglial cultures and the N9 microglial cell line

Cerebral cortices from postnatal day 1 MacGreen mice, which express enhanced green fluorescent protein (eGFP) under the control of the microglia/macrophage-specific promoter, CSF1R, in the C57BL6 background,48 were dissected, digested with trypsin (0.25% in HBSS) for 15 minutes at 37 °C, and mechanically dissociated by trituration, as described previously.49 Mixed cortical cells were plated in DMEM medium, supplemented with 10% FBS, 1% sodium pyruvate, and gentamycin on poly-d-lysine coated tissue culture plates. After 10 days, microglial cells were separated from the astrocytic monolayer by the addition of 12 mM lidocaine, and the isolated microglia were seeded onto 24 well plates at a density of 15 000 cells per mL. The N9 immortalized murine microglial cell line was maintained under identical culture conditions.50

α-Fe2O3 N-Rhomb preparation for cell culture

Due to the direct mutual attraction between nanostructures via either van der Waals forces or chemical bonding, some degree of aggregation is expected.51,52 As a result, the α-Fe2O3 N-Rhomb stock solution (1 mg mL–1), which was prepared using cell culture medium containing fetal bovine serum, was sonicated for 24 hours using a sonicator probe (220–260 V, 7.5 A, Misonix Model XL2020) in order to break up the bulkier agglomerates of nanocrystals.

Quantification of cell fluorescence

MacGreen microglia were plated onto 24-well plates containing coverslips at a density of 15 000 cells per mL. Approximately 24 hours after plating, microglia were treated with 1 μg mL–1, 10 μg mL–1, and 100 μg mL–1 solutions of either bare RhB or RhB-labeled α-Fe2O3 N-Rhomb. After 24 hours, coverslips were fixed in 4% paraformaldehyde (PFA) and mounted onto slides using 4′,6-diamidino-2-phenylindole (DAPI) fluoromount. Five Z-Stack images were taken per cover slip at 63× magnification at a digital resolution of 1024 × 1024 with a Zeiss confocal microscope using LSM 510 Meta software. The fluorescence of each cell per image was quantified using ImageJ software. The corrected total cell fluorescence (CTCF) was calculated with the formula: CTCF = integrated density of cell – area of cell X mean gray area of a background sample. For experiments in which the mechanism of nanoparticle uptake was studied, MacGreen microglia were pre-treated with chlorpromazine (CPZ) (Sigma-Aldrich) at a final concentration of 30 μM, approximately two hours before RhB-labeled α-Fe2O3 N-Rhomb exposure.

Electron microscopy

The morphology and size of the bare α-Fe2O3 nanostructures were assessed using a field emission SEM (FE-SEM Leo 1550) and an analytical high-resolution SEM (JEOL 7600F) instrument operating at an accelerating voltage of 15 kV, both of which were equipped with EDX capabilities. To prepare these samples for structural characterization, fixed amounts were dispersed in water and sonicated for ∼1 min, prior to deposition onto a silicon (Si) wafer.

Low magnification transmission electron microscopy (TEM) was also used at an accelerating voltage of 120 kV using a JEOL JEM-1400 instrument, equipped with a 2048 × 2048 Gatan CCD digital camera as well as with energy dispersive X-ray (EDX) spectroscopy capabilities. High-resolution TEM (HR-TEM) images coupled with SAED patterns were recorded using a JEOL JEM-3000F microscope, equipped with a Gatan image filter (GIF) spectrometer operating at an accelerating voltage of 300 kV. All samples were then primed for analysis by dispersion in water followed by sonication. Subsequently, the solution was deposited drop-wise onto a 300 mesh Cu grid.

To prepare the corresponding microglia engulfed RhB-labeled α-Fe2O3 nanostructures for TEM observation, microglia were plated onto ACLAR sheets (EMS, Hatfield, PA.) and treated with 1 μg mL–1, 10 μg mL–1, and 100 μg mL–1 of bare α-Fe2O3 N-Rhomb, respectively. Samples were then consigned to 2% osmium tetroxide in 0.1 M PBS, pH 7.4, dehydrated in a graded series of ethyl alcohol, and embedded with Durcupan resin. Samples were subsequently placed onto formvar-coated slot copper grids, counter-stained with uranyl acetate and lead citrate, and later viewed with a FEI Tecnai12 BioTwinG2 electron microscope. Digital images were acquired with an AMT XR-60 CCD Digital Camera system.

Enzyme linked immuno-absorbent assay (ELISA)

Conditioned medium, obtained from primary cell cultures respectively treated with 1 μg mL–1, 10 μg mL–1, and 100 μg mL–1 of α-Fe2O3 N-Rhomb, was used for ELISA analysis. Levels of tumor necrosis factor alpha (TNFα) and interleukin 1-beta (IL1β) were determined using the eBiosciences quantitative sandwich enzyme immunoassay following the manufacturer's protocol. Briefly, 96 well plates were coated with diluted capture antibody and incubated overnight at 4 °C. The plates were sealed for 1 hour at room temperature followed by a two-hour incubation period with either the standard or the sample. The wells were then incubated for 1 h with a working detector solution followed by incubation with the substrate solution for 30 min. The reaction was stopped with 50 μL of 1 N H2SO4. The absorbance of each well was recorded using 450 nm wavelength light on an ELISA plate reader.

Lactate dehydrogenase (LDH) cytotoxicity

An LDH Cytotoxicity Detection Kit (Roche Diagnostics Ltd) was used, according to the manufacturer's instructions. Briefly, primary microglia were incubated for either 24 or 48 hours with different concentrations of α-Fe2O3 N-Rhomb. Samples were run in triplicate to determine LDH release by microglia in the presence of α-Fe2O3 N-Rhomb. Untreated cells served as a ‘low control’, whereas detergent-lysed cells served as a ‘high control’ for LDH release. 100 μL of the kit ‘reaction mixture’ was added to the wells. The plate was then maintained in the dark for 30 minutes at room temperature, to allow for the tetrazolium salt, INT, in the presence of LDH, to become reduced to formazan. The reactions were terminated by addition of ‘stop’ solution. Absorbance values of the formazan dye were measured at 490 nm using an ELISA plate reader. Cytotoxicity was calculated using the following equation: Cytotoxicity (%) = (experimental reading – low control reading)/(high control reading – low control reading) × 100%.

Nitric oxide (NO) assay

The concentration of NO produced by microglia cultured in the presence of N-Rhomb was determined by the reaction of nitrite present in cell supernatant with 2,3-diaminonaphthalene (DAN). Briefly, primary microglia were seeded in 96 well plates and exposed to the α-Fe2O3 N-Rhomb motifs for 24 hours. The cultured supernatant was then collected and centrifuged to remove the N-Rhomb. 100 μL of cell supernatant and 20 μL DAN (0.05 mg ml–1 in 0.62 M HCl) were re-suspended together. The reaction was terminated after 20 minutes with the addition of 100 μL of 0.28 M NaOH. A Mithras LB 940 Multimode Microplate Reader (Berthold Technologies) was then used to measure 2,3-naphthyltriazole formation, using a 355 nm excitation/460 nm emission filter pair. The nitrite concentrations in samples were calculated, based upon a sodium nitrite standardization curve.

DCFDA assay for the detection of ROS

The production of reactive oxygen species (ROS) by microglia was measured using a 2′,7′-dichlorofluorescein diacetate (DCFDA) assay, as previously described.53 Non-fluorescent DCFDA is converted to 2′,7′-dichlorofluorescein (DCF) in the presence of ROS. Briefly, microglia were plated onto a 96 well plate and incubated for 24 hours with different concentrations of N-Rhomb. After 24 hours, the media from the plates were aspirated off, and the cells were washed with 1× PBS followed by incubation with 25 μM DCFDA dissolved in 1× PBS for 30 minutes. The cells were then washed in 1× PBS, and the amount of DCF present within the cells was quantified on a Fluoroskan Ascent Microplate Fluorometer (Thermo Scientific) using a 485 nm excitation/530 nm emission filter pair. The fluorescence of treated samples was standardized relative to untreated (i.e. negative) control cells. Cells incubated for 24 hours with 100 ng mL–1 lipopolysaccharide (LPS) served as a positive control for the production of ROS and the subsequent oxidation of DCFDA to DCF.

Animals

All experiments conducted had prior approval from the Institutional Animal Care and Use Committee (IACUC) as well as the Department of Laboratory Animal Research at Stony Brook University.

Intrahippocampal RhB- labeled α-Fe2O3 N-Rhomb injection

Injections were performed bilaterally in the hippocampus. Mice were anesthetized with 1.25% Avertin and injected with 100 μg mL–1 of RhB-labeled α-Fe2O3 N-Rhomb at stereotactic coordinates –2.5 mm from Bregma and –1.7 mm lateral using a Hamilton syringe (0.485 mm I.D., Hamilton, Reno, NV) connected to a motorized stereotaxic injector (Stoelting, Wood Dale, IL). Following surgery, animals were injected intraperitoneally (i.p.) with 0.03 mg kg–1 of buprenorphine (Bedford labs) and left on a heating pad, until they were fully recovered from anesthesia. After 24 hours post-injection, mice were anesthetized and perfused with 4% PFA. The brains were collected, post-fixed, cryo-protected, and cut into 40 μm thick sections. The sections were then mounted with DAPI fluoromount, and photographed at a digital resolution of 1024 × 1024 with a Zeiss confocal microscope using LSM 510 Meta software.

Statistics

All statistics were performed using either Statview (v. 4.0) or GraphPad Prism 6 for Windows. Data are presented as mean ± SEM (i.e. standard error on the mean). One-way ANOVA was used to determine the level of significance between groups in the fluorescence quantification analysis, the LDH test, and the TNFα ELISA assay, respectively. A Bonferroni post-test was used to control for multiple comparisons. Data were considered to be statistically significant, when p < 0.05.

3. Results

Product characterization of bare α-Fe2O3 N-Rhomb

We discuss the preparative protocols of our α-Fe2O3 N-Rhomb nanostructures in significant detail in the ESI.† Specifically, using the hydrothermal technique at 120 °C for 12 hours, we were able to generate both average-sized and small-sized α-Fe2O3 N-Rhomb, as determined by XRD (Fig. 1C & D), with all of the expected diffraction peaks observed, corresponding to the standard JCPDS pattern for phase-pure hematite α-Fe2O3 (JCPDS #86-0550). Typical images associated with the SEM analysis of the smaller-sized and average-sized N-Rhomb, as shown in Fig. 1, revealed that the nanostructures possessed the correct morphology and were uniform in size, with associated measured lengths of 47 ± 10 nm and 75 ± 8 nm, and corresponding widths of 35 ± 8 nm and 50 ± 8 nm, respectively.

Fig. 1. Characterization of small (i.e. ∼47 nm) and average sized (i.e. ∼75 nm) bare α-Fe2O3 N-Rhomb. SEM and XRD images of small (A & C) and average-sized (B & D) α-Fe2O3 N-Rhomb. High-resolution TEM image (E) of bare, ∼75 nm α-Fe2O3 N-Rhomb. A low magnification image in the lower right-hand inset is shown. The upper left-hand inset shows the electron diffraction pattern.

Fig. 1

Open in a new tab

Cultured microglia Engulf bare α-Fe2O3 N-Rhomb

When the CNS undergoes either injury, infection, or disease, microglia, i.e. the immuno-competent cells of the CNS, act as the first line of defense. They migrate to the site of injury, assume antigen-presenting properties, secrete cytokines, and trigger phagocytosis of dead cells and cell debris.54

To test the ability of microglia to internalize nano-sized α-Fe2O3 particles, N9 immortalized microglia were exposed to increasing concentrations of the smaller sized bare α-Fe2O3 N-Rhomb (Fig. 2A). The cells were imaged with a light microscope, approximately 24 hours after exposure. The light microscopy images show that N9 microglia did indeed internalize α-Fe2O3 N-Rhomb and consequentially, remained viable after treatment, as the structural integrity of the membranes of the microglia cells had been maintained, even upon nanostructure incorporation (Fig. 2A).

Fig. 2. Cultured primary microglia engulf bare ∼47 nm α-Fe2O3 N-Rhomb. Light microscopy images (A) of untreated cells and of cells exposed to 1, 10, and 100 μg mL–1, respectively, of bare α-Fe2O3 N-Rhomb. Images were taken after 24 h after α-Fe2O3 N-Rhomb exposure. Red arrows point towards cells that have internalized α-Fe2O3 N-Rhomb. TEM and high magnification TEM images (B) of untreated cells and of cells exposed to 1, 10, and 100 μg mL–1, respectively, of bare α-Fe2O3 N-Rhomb. Scale bars are either 2 μm or 500 nm.

Fig. 2

Open in a new tab

Electron microscopy images of primary C57BL6 microglia were also consistent with this observation. For each nanoparticulate concentration tested, these microglia cells were found to have incorporated ∼47 nm α-Fe2O3 N-Rhomb primarily within the cellular vesicles as opposed to the nucleus (Fig. 2B). In general, we found that the higher the initial incubation concentration, the greater the number of N-Rhomb particles observed within the microglia. Moreover, there was no sign of cytotoxicity, as the cells retained their resting state and shape (i.e. long branches with small cellular bodies). It is noteworthy that the larger-sized ∼75 nm N-Rhomb were detected and incorporated to a lesser extent within the microglia cells as compared with the correspondingly smaller ∼47 nm N-Rhomb (Fig. S1†). As a result, additional experiments were performed on the smaller-sized ∼47 nm nanorhombohedra.

To further confirm the presence of α-Fe2O3 N-Rhomb within the microglia cells themselves, chemically-sensitive EDX spectroscopy data were taken on two representative regions of the microglia (Fig. S2†). In one area, the hematite N-Rhomb structures appear to be clearly engulfed within the microglia cells (Free Draw 1), whereas in another part of the sample, no hematite nanorhombohedra are apparently visible (Free Draw 2). Based upon the EDX spectrum, the signals associated with the N-Rhomb-containing area gave rise to significantly higher peak intensities for Fe, i.e. ∼7× larger, than for the area without N-Rhomb present. Therefore, these data are consistent with the idea of the iron oxide nanostructures as being localized and engulfed within the microglia cells, as expected. Other peaks such as copper (Cu), lead (Pb), and osmium (Os) emanate from the TEM copper grid as well as from the cross-sectional staining agents of lead citrate and osmium tetroxide, respectively.

To visualize the in vitro engulfment of the α-Fe2O3 N-Rhomb by microglia from an optical perspective, α-Fe2O3 N-Rhomb were labeled with the fluorescent dye RhB (95%, Aldrich) for easy detection (Fig. 3A). We describe the chemical modification protocol used to conjugate α-Fe2O3 N-Rhomb with RhB both in words as well as schematically (Fig. S3) in the ESI.† Spectroscopic confirmation of the successful attachment and binding of the dye onto the iron oxide surface was provided by UV-visible and infrared (IR) spectroscopy data. The expected absorption and bond signatures noted in these results are consistent with the generation of RhB-labeled α-Fe2O3 N-Rhomb. These data are provided in the ESI (Fig. S4†). Additionally, the engulfment behavior of RhB labeled N-Rhomb by primary microglia was compared with that of a bare RhB control (Fig. S5†). Primary microglia were obtained from neonatal MacGreen mice, which express eGFP under the control of the microglia/macrophage promoter CSF1R in the C57BL6 background.48 MacGreen microglia were exposed to increasing concentrations of both RhB-labeled α-Fe2O3 N-Rhomb and bare RhB in separate runs for 24 hours. The cells were fixed, mounted on slides, and imaged using a confocal microscope. The fluorescence intensity, emanating from both the RhB-labeled α-Fe2O3 N-Rhomb and bare RhB contained within each cell, was quantified using ImageJ (Fig. 3C & Fig. S5†).

Fig. 3. Primary microglia internalize RhB-labeled α-Fe2O3 N-Rhomb using a clathrin-dependent mechanism. Confocal images (A) of eGFP expressing microglia exposed to 1, 10, and 100 μg mL–1, respectively, of RhB-labeled α-Fe2O3 N-Rhomb, stained with DAPI for nuclear staining. Confocal images (B) of cells exposed to 0, 1, 10, and 100 μg mL–1, respectively, of RhB-labeled α-Fe2O3 N-Rhomb. Cells were pre-treated with 30 μM CPZ, 2 hours prior to nanoparticle exposure. Images were taken 24 hours after nanoparticle exposure. (C). Quantification of the RhB fluorescence of microglia treated either with or without 30 μM CPZ, followed by incubation with RhB-labeled Fe2O3 N-Rhomb. Scale bars = 20 μm. Data are shown as mean ± SEM. **** p < 0.0001.

Fig. 3

Open in a new tab

Fig. 3 is consistent with increasing fluorescence intensity with increasing concentrations of RhB-labeled α-Fe2O3 N-Rhomb analyzed. Cells treated with 100 μg mL–1 of RhB-labeled α-Fe2O3 N-Rhomb possessed significantly higher total cell fluorescence especially when compared with not only untreated cells but also cells treated with 1 μg mL–1 and 10 μg mL–1 of RhB-labeled α-Fe2O3 N-Rhomb, thereby indicating that microglia do successfully internalize these nanostructures (Fig. 3).

By comparison, the control RhB samples also substantiated a trend of increasing fluorescent intensity with increasing concentration (Fig. S5†). However, changes in the microglial morphology from the original resting to a more activated, amoeboid form were evident, when cells treated with sample controls at concentrations of 10 μg mL–1 and higher were imaged (Fig. S5†). However, this did not appear to be true in the presence of RhB-labeled Fe2O3 N-Rhomb, as these microglia cells all maintained their inactive, ramified morphology.

Microglia engulf Rh-B-labeled α-Fe2O3 N-Rhomb in a clathrin-dependent manner

To investigate the mechanism underlying the internalization of RhB functionalized α-Fe2O3 N-Rhomb, we tested the effect of chlorpromazine (CPZ), a specific clathrin-mediated endocytosis inhibitor, on the ability of microglia to internalize RhB-labeled α-Fe2O3 N-Rhomb (Fig. 3B).55 Cultured MacGreen microglia were pre-treated with a high concentration (30 μM) of CPZ for 2 hours prior to RhB-labeled α-Fe2O3 N-Rhomb exposure, as previously described.55 Approximately 24 hours later, the cells were fixed, mounted, and imaged under a confocal microscope. Microglia maintained their resting morphology after CPZ treatment, and in effect, we observed that the concentration of CPZ induced little if any apparent toxic effects (Fig. 3B).

The localization of RhB labeled Fe2O3 N-Rhomb outside the cellular membrane of the microglia cells under confocal microscopy conditions suggested a lack of engulfment of the nanostructures (Fig. 3B). Fluorescence quantification shows that treatment with 30 μM CPZ significantly prevented as much as 96% of the potential uptake of RhB-labeled α-Fe2O3 N-Rhomb at all of the concentrations of nanoparticles tested from 1 μg mL–1 to 100 μg mL–1, thereby supporting the idea that microglia primarily internalize these particles through a clathrin-dependent mechanism (Fig. 3C).

RhB-labeled α-Fe2O3 N-Rhomb are not toxic to microglia at therapeutic concentrations

To test the potential cytotoxicity of the RhB functionalized α-Fe2O3 N-Rhomb on cultured microglia, a lactate dehydrogenase (LDH) assay was used. LDH is rapidly released when the membranes of cells rupture, and hence, the presence of LDH in the supernatant is indicative of cell death.56 As such, primary microglia were treated with increasing concentrations of RhB-labeled α-Fe2O3 N-Rhomb (Fig. 4A), and the media were collected 24 and 48 hours later so as to measure LDH release. As Fig. 4A shows, both untreated and treated microglia exhibited similar levels of LDH release and displayed less than ∼4% cytotoxicity at 24 hours. After 48 hours, a significant increase of ∼30% cytotoxicity was observed in microglia treated with the highest concentration of NRhomb (i.e. 100 μg mL–1). Nevertheless, it should be noted that this concentration is approximately 626 μM, which far exceeds a normal therapeutic dose. All of the other concentrations tested exhibited no apparent cytotoxicity over the period of time tested.

Fig. 4. RhB-labeled α-Fe2O3 N-Rhomb are minimally cytotoxic and do not result in either the upregulation of pro-inflammatory factors or nitrite production within cultured microglia. (A). Conditioned media from primary microglia, treated with 1, 10, and 100 μg mL–1, respectively, of α-Fe2O3 N-Rhomb were collected, and levels of LDH were measured at 24 and 48 hours. Untreated cells (Ctrl) served as a negative control and lysed cells were used as a positive control for LDH release. (B and C). Primary microglia were treated with either 0 (Ctrl), 1, 10, or 100 μg mL–1, respectively, of α-Fe2O3 N-Rhomb, or with 100 ng mL–1 of lipopolysaccharide (LPS). Approximately 24 hours after treatment, media isolated from the cells were used for the detection of TNFα (B). or IL1β (C). (D). Nitrite production by primary microglia after 24 hours of incubation with 0 (Ctrl), 1, 10, or 100 μg mL–1 of α-Fe2O3 N-Rhomb, or with 100 ng mL–1 LPS. (E). Oxidation of DCFDA by microglia after a 24 hour incubation with 0 (Ctrl), 1, 10, or 100 μg mL–1 of α-Fe2O3 N-Rhomb, or with 100 ng mL–1 LPS. Data are shown as mean ± SEM. * p < 0.05, ** p < 0.01, **** p < 0.0001.

Fig. 4

Open in a new tab

RhB-labeled α-Fe2O3 N-Rhomb do not cause in vitro microglial activation

Microglia can give rise to at least two different activation states, depending on the signals they receive: the pro-inflammatory M1 state and the anti-inflammatory M2 state. In the M1 state, microglia secrete TNFα, IL1β, and other pro-inflammatory cytokines, while in the M2 state, microglia produce IL-10, IL-4, TGF-β, as well as other anti-inflammatory factors. In several injury models, M2 microglia have been deemed to be beneficial for tissue regeneration,57,58 whereas M1 microglia are considered to inhibit tissue healing and repair. Upregulation of TNFα has been found in previous literature to be implicated as a factor in various ailments such as Alzheimer's disease, cancer, major depression, and inflammatory bowel disease.59–63

To evaluate whether RhB-labeled α-Fe2O3 N-Rhomb result in microglial release of pro-inflammatory factors, ELISAs were performed to quantify the levels of the pro-inflammatory cytokines TNFα and IL1β. Primary microglia were treated with increasing concentrations of RhB-labeled α-Fe2O3 N-Rhomb, and the media were collected approximately 24 hours later to measure corresponding levels of TNFα and IL1β (Fig. 4B & C).

Exposure to lipopolysaccharide (LPS) was used as a positive control since it is a potent inducer of pro-inflammatory cytokines in microglia.64 After 24 hours, there was no significant increase in the levels of either TNFα or IL1β produced by microglia that had been treated with systematically greater concentrations (i.e. 1, 10, or 100 μg mL–1) of RhB-labeled α-Fe2O3 N-Rhomb, relative to the control. As expected, the levels of TNFα and IL1β released by LPS-treated cells were significantly higher than those produced by both untreated cells (i.e. control) as well as the cells treated with RhB-labeled α-Fe2O3 N-Rhomb, thereby confirming that the presence of these nanoparticles does not necessarily give rise to the expression of pro-inflammatory agents.

RhB-labeled α-Fe2O3 N-Rhomb do not trigger either nitric oxide or ROS production in vitro

It is well known that nitric oxide (NO) is associated with various key functions within the CNS, such as regulation of synaptic plasticity, the sleep-wake cycle, and hormone secretion.65,66 However, when produced in excess, NO can undergo oxidation reduction reactions through the formation of reactive oxygen species (ROS), thereby generating reactive nitrogen-containing species that can result in nitrosative stress and cellular damage.65,66 Nitrite production by microglia after engulfment of N-Rhomb remained low, suggesting that the presence of varying concentrations of N-Rhomb did not result in the production of NO by microglia (Fig. 4D). All concentrations of RhB-labeled N-Rhomb particles tested gave rise to insignificant changes in NO production by microglia, relative to untreated, control cells. Moreover, their NO production was significantly lower than that of a positive LPS control. Hence, it can be concluded that the presence of RhB labeled N-Rhomb alone does not induce noticeable NO production in microglia.

The production of ROS by microglia was also assessed using a DCFDA assay. DCFDA is esterified and oxidized by cells in the presence of ROS, giving rise to the fluorescent compound, DCF. Microglia treated with LPS served as a positive control for the production of ROS and, by extension, the intracellular accumulation of DCF. Cells treated with increasing concentrations of N-Rhomb showed no significant elevation in DCF accumulation relative to that of control samples, i.e. untreated cells, a finding indicative of negligible production of ROS (Fig. 4E). Thus, it was determined that concentrations of NRhomb of up to 100 μg mL–1 did not induce ROS production in microglia in vitro.

Microglia internalize RhB-labeled α-Fe2O3 N-Rhomb in vivo

To test whether microglia can internalize RhB-labeled α-Fe2O3 N-Rhomb in vivo, MacGreen mice were injected with 100 μg mL–1 of RhB-labeled α-Fe2O3 N-Rhomb bilaterally into the dorsal hippocampus. Approximately 24 hours after the injection, the mice were transcardially perfused with 4% PFA; the brains were then collected, cryo-protected, sectioned into 40 μm thick slices, mounted, and imaged under a confocal microscope. White arrows in Fig. 5 point to microglia that have internalized RhB-labeled α-Fe2O3 N-Rhomb. Taken together, these data indicate that microglia can engulf α-Fe2O3 nanostructures under both in vitro and in vivo conditions in a clathrin-dependent manner without causing microglia to release pro-inflammatory factors which might have thereby compromised the viability of the cells.

Fig. 5. Microglia engulf RhB-labeled α-Fe2O3 N-Rhomb in vivo. Confocal images of brain sections from MacGreen mice treated with 100 μg mL–1 RhB-labeled α-Fe2O3 N-Rhomb. White arrows point to microglia that have internalized the RhB-labeled α-Fe2O3 N-Rhomb. Scale = 20 μm.

Fig. 5

Open in a new tab

4. Discussion

We have synthesized RhB-labeled α-Fe2O3 N-Rhomb, and tested the ability of microglia to internalize them. These cells appear to efficiently engulf the RhB-functionalized α-Fe2O3 N-Rhomb without any noticeable membrane damage, as suggested by electron and confocal microscopy. In terms of addressing potential shape-dependent toxicity, RhB-labeled α-Fe2O3 N-Rhomb are non-inflammatory and non-cytotoxic at therapeutic concentrations, suggesting that α-Fe2O3 N-Rhomb have the potential to be used as drug carriers. Drugs that either reduce M1 activation or induce M2 activation such as minocycline, tuftsin (TKPR), or microglia inhibitory factor (MIF/TKP)67–69 could be conjugated onto nanoparticles in order to alter the phenotype of microglia. For instance, Papa et al. showed that nanostructures conjugated onto minocycline are engulfed by microglia and reduce inflammation in a model of spinal cord injury.55

Our results thus show promise for future studies involving the conjugation of anti-inflammatory compounds onto nanostructures that can be engulfed by microglia as well as for the tracking of cell behavior using imaging techniques such as confocal microscopy and MRI, for example. This point has been demonstrated by some of our unpublished work (data not shown), in which we have observed that α-Fe2O3 N-Rhomb can be detected within a mouse brain by using T2-weighted MRI scans.

We have been able to demonstrate that microglia can internalize both bare and RhB-labeled α-Fe2O3 N-Rhomb. Particle aggregation is expected due to the direct mutual attraction between nanostructures occurring via either van der Waals forces or chemical bonding.51,52 Sonication of the RhB-labeled N-Rhomb prior to exposure to microglia reduced the degree of particle aggregation, although it did not completely prevent clustering. Due to the low TNFα, IL-1β, ROS, and nitrite levels in microglia after the RhB-labeled N-Rhomb treatment, it is unlikely that microglia themselves became over-activated, thereby resulting in an engulfment of a large amount of particles. The particle clusters observed in Fig. 2 are more likely the result of aggregation, due to strong mutual attraction between these nanostructures.

Microglia are highly phagocytic and can clear away dead cells and debris using a variety of endocytic mechanisms, including receptor-mediated endocytosis, pinocytosis, and phagocytosis.70–72 Moreover, microglia play key roles in several neurological diseases and can quickly respond to either infection or injury.73,74 Chemically modified nanostructures that can be easily engulfed by microglia represent therefore a potentially viable strategy with which to manipulate the functional properties of the microglia themselves.

Specifically, we have shown that microglia used a clathrin-dependent endocytic pathway to internalize RhB-labeled α-Fe2O3 N-Rhomb, as evidenced by the lack of nanoparticle uptake even at varying concentration levels, in the presence of the endocytosis inhibitor, CPZ. This observation is in agreement with other studies that have successfully demonstrated internalization of other types of nanostructures by microglia.55,75,76 Additionally, we have highlighted that internalization does not lead to aberrant activation in cultured microglia, i.e. in the presence of RhB-labeled α-Fe2O3 N-Rhomb maintaining concentrations of up to 100 μg mL–1, and that microglia in the mouse brain are equally efficient at internalizing hematite nanostructures, thereby indicating that α-Fe2O3 N-Rhomb may be suitable for coupling anti-inflammatory agents as a form of drug therapy.

Supplementary Material

Click here for additional data file. (959.5KB, pdf)

Acknowledgments

Research (including support for LT and CSL) was provided by the Turner Dissertation fellowship (LT) and the Alliance for Graduate Education and Professoriate – Transformation (AGEP-T), which is funded by National Science Foundation – Division of Human Resources Development Contract No. HRD 1311318. Research funds were provided by the U.S. Department of Energy, Basic Energy Sciences, Materials Sciences and Engineering Division (JMP and SSW), NIH T32GM007518 as well as NIH F30CA196110 (JTM), and NIH R01NS42168 (SET), respectively. Experiments were performed in part at the Center for Functional Nanomaterials located at Brookhaven National Laboratory, which is supported by the U.S. Department of Energy under contract number DE-SC-00112704. Transmission electron microscopy data acquired for our microglia experiments were collected at the Center Microscopy Imaging Center (C-MIC) at Stony Brook University, Stony Brook, NY 11794 under the direction of Ms. Susan Van Horn.

Footnotes

†Electronic supplementary information (ESI) available. See DOI: 10.1039/c5tx00421g

References

  1. Colvin V. Nat. Biotechnol. 2003;21:1166–1170. doi: 10.1038/nbt875. [DOI] [PubMed] [Google Scholar]
  2. Nanomaterials: toxicity, health and environmental issues, ed. C. S. S. R. Kumar, Wiley-VCH, Weinheim, 2006. [Google Scholar]
  3. Lewinski N., Colvin V., Drezek R. Small. 2008;4:26–49. doi: 10.1002/smll.200700595. [DOI] [PubMed] [Google Scholar]
  4. Li W., Chen C., Ye C., Wei T., Zhao Y., Lao F., Chen Z., Meng H., Gao Y., Yuan H., Xing G., Zhao F., Chai Z., Zhang X., Yang F., Han D., Tang X., Zhang Y. Nanotechnology. 2008;19:145102. doi: 10.1088/0957-4484/19/14/145102. [DOI] [PubMed] [Google Scholar]
  5. Sudimack J., Lee R. J. Adv. Drug Delivery Rev. 2000;41:147–162. doi: 10.1016/s0169-409x(99)00062-9. [DOI] [PubMed] [Google Scholar]
  6. Suh W. H., Suslick K. S., Stucky G. D., Suh Y.-H. Prog. Neurobiol. 2009;87:133–170. doi: 10.1016/j.pneurobio.2008.09.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen S., Zhao X., Chen J., Chen J., Wong S. S., Ojima I. Bioconjugate Chem. 2010;21:979–987. doi: 10.1021/bc9005656. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Zhang F., Wong S. S. ACS Nano. 2010;4:99–112. doi: 10.1021/nn901057y. [DOI] [PubMed] [Google Scholar]
  9. Zhou H., Chen J., Sutter E., Feygenson M., Aronson M. C., Wong S. S. Small. 2010;6:412–420. doi: 10.1002/smll.200901276. [DOI] [PubMed] [Google Scholar]
  10. Chen J., Zhou H., Santulli A. C., Wong S. S. Chem. Res. Toxicol. 2010;23:871–879. doi: 10.1021/tx900418b. [DOI] [PubMed] [Google Scholar]
  11. Panessa-Warren B. J., Warren J., Wong S. S., Misewich J. A. J. Phys.: Condens. Matter. 2006;18:S2185–S2201. [Google Scholar]
  12. Magrez A., Horváth L., Smajda R., Salicio V., Pasquier N., Forró L., Schwaller B. ACS Nanosci. Nanotechnol. 2009;3:2274–2280. doi: 10.1021/nn9002067. [DOI] [PubMed] [Google Scholar]
  13. Jin C., Tang Y., Yang F., Li X., Xu S., Fan X., Huang Y., Yang Y. Biol. Trace Elem. Res. 2011;141:3–15. doi: 10.1007/s12011-010-8707-0. [DOI] [PubMed] [Google Scholar]
  14. Tejral G., Panyala N. R., Havel J. J. Appl. Biomed. 2009;7:1–13. [Google Scholar]
  15. Samra Z. Q., Shabir S., Rehmat Z., Zaman M., Nazir A., Dar N., Athar M. A. Appl. Biochem. Biotechnol. 2010;162:671–686. doi: 10.1007/s12010-009-8840-4. [DOI] [PubMed] [Google Scholar]
  16. Ling Y., Wei K., Luo Y., Gao X., Zhong S. Biomaterials. 2011;32:7139–7150. doi: 10.1016/j.biomaterials.2011.05.089. [DOI] [PubMed] [Google Scholar]
  17. Polyak B., Friedman G. Expert Opin. Drug Delivery. 2009;6:53–70. doi: 10.1517/17425240802662795. [DOI] [PubMed] [Google Scholar]
  18. Wang Y. Quant. Imaging Med. Surg. 2011;1:35–40. doi: 10.3978/j.issn.2223-4292.2011.08.03. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Xu S., Yang F., Zhou X., Zhuang Y., Liu B., Mu Y., Wang X., Shen H., Zhi G., Wu D. ACS Appl. Mater. Interfaces. 2015;7:20460–20468. doi: 10.1021/acsami.5b06594. [DOI] [PubMed] [Google Scholar]
  20. Iqbal M. Z., Ma X., Chen T., Zhang L., Ren W., Xiang L., Wu A. J. Mater. Chem. B. 2015;3:5172–5181. doi: 10.1039/c5tb00300h. [DOI] [PubMed] [Google Scholar]
  21. Catti M., Valerio G., Dovesi R. Phys. Rev. B: Condens. Matter. 1995;51:7441–7450. doi: 10.1103/physrevb.51.7441. [DOI] [PubMed] [Google Scholar]
  22. Liang H., Jiang X., Qi Z., Chen W., Wu Z., Xu B., Wang Z., Mi J., Li Q. Nanoscale. 2014;6:7199–7203. doi: 10.1039/c4nr00552j. [DOI] [PubMed] [Google Scholar]
  23. Wang G., Li W., Jia K., Spliethoff B., Schüth F., Lu A. Appl. Catal., A. 2009;364:42–47. [Google Scholar]
  24. Xu S., Hessel C. M., Ren H., Yu R., Jin Q., Yang M., Zhao H., Wang D. Energy Environ. Sci. 2014;7:632–637. [Google Scholar]
  25. Tong G., Guan J., Zhang Q. Mater. Chem. Phys. 2011;127:371–378. [Google Scholar]
  26. Basnet P., Larsen G. K., Jadeja R. P., Hung Y.-C., Zhao Y. ACS Appl. Mater. Interfaces. 2013;5:2085–2095. doi: 10.1021/am303017c. [DOI] [PubMed] [Google Scholar]
  27. Berg J. M., Ho S., Hwang W., Zebda R., Cummins K., Soriaga M. P., Taylor R., Guo B., Sayes C. M. Chem. Res. Toxicol. 2010;23:1874–1882. doi: 10.1021/tx100307h. [DOI] [PubMed] [Google Scholar]
  28. Freyria F. S., Bonelli B., Tomatis M., Ghiazza M., Gazzano E., Ghigo D., Garrone E., Fubini B. Chem. Res. Toxicol. 2012;25:850–861. doi: 10.1021/tx2004294. [DOI] [PubMed] [Google Scholar]
  29. Chen L., Xie J., Yancey J., Srivatsan M., Varadan V. K. J. Nanotechnol. Eng. Med. 2010;1:041014. [Google Scholar]
  30. Gustafsson Å., Bergström U., Ågren L., Österlund L., Sandström T., Buchta A. Toxicol. Appl. Pharmacol. 2015;288:1–11. doi: 10.1016/j.taap.2015.07.001. [DOI] [PubMed] [Google Scholar]
  31. Zhao Y., Pan F., Li H., Niu T., Xu G., Chen W. J. Mater. Chem. A. 2013;1:7242–7246. [Google Scholar]
  32. Barton L. E., Quicksall A. N., Maurice P. A. Geomicrobiol. J. 2012;29:314–322. [Google Scholar]
  33. Adili A., Crowe S., Beaux M. F., Cantrell T., Shapiro P. J., McIlroy D. N., Gustin K. E. Nanotoxicology. 2008;2:1–8. [Google Scholar]
  34. Wang Y., Zhou X. Y., Chen Z., Cai B., Ye Z. Z., Gao C. Y., Huang J. Y. Appl. Phys. A: Solid Surf. 2014;117:2121–2126. [Google Scholar]
  35. Lockman P. R., Oyewumi M. O., Koziara J. M., Roder K. E., Mumper R. J., Allen D. D. J. Controlled Release. 2003;93:271–282. doi: 10.1016/j.jconrel.2003.08.006. [DOI] [PubMed] [Google Scholar]
  36. Farrell B. T., Hamilton B. E., Dósa E., Rimely E., Nasseri M., Gahramanov S., Lacy C. A., Frenkel E. P., Doolittle N. D., Jacobs P. M., Neuwelt E. A. Neurology. 2013;81:256–263. doi: 10.1212/WNL.0b013e31829bfd8f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wang Y., Wang B., Zhu M. T., Li M., Wang H. J., Wang M., Ouyang H., Chai Z. F., Feng W. Y., Zhao Y. L. Toxicol. Lett. 2011;205:26–37. doi: 10.1016/j.toxlet.2011.05.001. [DOI] [PubMed] [Google Scholar]
  38. Wang B., Feng W. Y., Zhu M. T., Wang Y., Wang M., Gu Y. Q., Ouyang H., Wang H. J., Li M., Zhao Y. L., Chai Z. F., Wang H. F. J. Nanopart. Res. 2009;11:41–53. [Google Scholar]
  39. Wang J. X., Chen C. Y., Liu Y., Jiao F., Li W., Lao F., Li Y. F., Li B., Ge C. C., Zhou G. Q., Gao Y. X., Zhao Y. L., Chai Z. F. Toxicol. Lett. 2008;183:72–80. doi: 10.1016/j.toxlet.2008.10.001. [DOI] [PubMed] [Google Scholar]
  40. Mahmoudi M., Sant S., Wang B., Laurent S., Sen T. Adv. Drug Delivery Rev. 2011;63:24–46. doi: 10.1016/j.addr.2010.05.006. [DOI] [PubMed] [Google Scholar]
  41. Chew L. J., Fusar-Poli P., Schmitz T. Dev. Neurosci. 2013;35:102–129. doi: 10.1159/000346157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Deng Y., Lu J., Sivakumar V., Ling E. A., Kaur C. Brain Pathol. 2008;18:387–400. doi: 10.1111/j.1750-3639.2008.00138.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Merrill J. E., Scolding N. J. Neuropathol. Appl. Neurobiol. 1999;25:435–458. doi: 10.1046/j.1365-2990.1999.00200.x. [DOI] [PubMed] [Google Scholar]
  44. Peferoen L., Kipp M., van der Valk P., van Noort J. M., Amor S. Immunology. 2014;141:302–313. doi: 10.1111/imm.12163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Domercq M., Perez-Samartin A., Aparicio D., Alberdi E., Pampliega O., Matute C. Glia. 2010;58:730–740. doi: 10.1002/glia.20958. [DOI] [PubMed] [Google Scholar]
  46. Varnum M. M., Ikezu T. Arch. Immunol. Ther. Exp. 2012;60:251–266. doi: 10.1007/s00005-012-0181-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Gao Z., Tsirka S. E. Neurol. Res. Int. 2011;2011:383087. doi: 10.1155/2011/383087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sasmono R. T., Ehrnsperger A., Cronau S. L., Ravasi T., Kandane R., Hickey M. J., Cook A. D., Himes S. R., Hamilton J. A., Hume D. A. J. Leukocyte Biol. 2007;82:111–123. doi: 10.1189/jlb.1206713. [DOI] [PubMed] [Google Scholar]
  49. Bronstein R., Torres L., Nissen J. C., Tsirka S. E. J. Visualized Exp. 2013;78 doi: 10.3791/50647. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Stansley B., Post J., Hensley K. J. Neuroinflammation. 2012;9:115. doi: 10.1186/1742-2094-9-115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Hunter R. J., Foundations of Colloid Science, Oxford University Press, New York, 1987. [Google Scholar]
  52. Li D., Kaner R. B. J. Am. Chem. Soc. 2006;128:968–975. doi: 10.1021/ja056609n. [DOI] [PubMed] [Google Scholar]
  53. Harrington A. D., Tsirka S. E., Schoonen M. A. A. Chemosphere. 2013;93:1216–1221. doi: 10.1016/j.chemosphere.2013.06.082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Nakajima K., Kohsaka S. J. Biochem. 2001;130:169–175. doi: 10.1093/oxfordjournals.jbchem.a002969. [DOI] [PubMed] [Google Scholar]
  55. Papa S., Rossi F., Ferrari R., Mariani A., De Paola M., Caron I., Fiordaliso F., Bisighini C., Sammali E., Colombo C., Gobbi M., Canovi M., Lucchetti J., Peviani M., Morbidelli M., Forloni G., Perale G., Moscatelli D., Veglianese P. ACS Nano. 2013;7:9881–9895. doi: 10.1021/nn4036014. [DOI] [PubMed] [Google Scholar]
  56. Chan F. K., Moriwaki K., De Rosa M. J. Methods Mol. Biol. 2013;979:65–70. doi: 10.1007/978-1-62703-290-2_7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Kigerl K. A., Gensel J. C., Ankeny D. P., Alexander J. K., Donnelly D. J., Popovich P. G. J. Neurosci. 2009;29:13435–13444. doi: 10.1523/JNEUROSCI.3257-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Miron V. E., Boyd A., Zhao J. W., Yuen T. J., Ruckh J. M., Shadrach J. L., van Wijngaarden P., Wagers A. J., Williams A., Franklin R. J., ffrench-Constant C. Nat. Neurosci. 2013;16:1211–1218. doi: 10.1038/nn.3469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Brenner D., Blaser H., Mak T. W. Nat. Rev. Immunol. 2015;15:362–374. doi: 10.1038/nri3834. [DOI] [PubMed] [Google Scholar]
  60. Locksley R. M., Killeen N., Lenardo M. J. Cell. 2001;104:487–501. doi: 10.1016/s0092-8674(01)00237-9. [DOI] [PubMed] [Google Scholar]
  61. Swardfager W., Lanctot K., Rothenburg L., Wong A., Cappell J., Herrmann N. Biol. Psychiatry. 2010;68:930–941. doi: 10.1016/j.biopsych.2010.06.012. [DOI] [PubMed] [Google Scholar]
  62. Dowlati Y., Herrmann N., Swardfager W., Liu H., Sham L., Reim E. K., Lanctot K. L. Biol. Psychiatry. 2010;67:446–457. doi: 10.1016/j.biopsych.2009.09.033. [DOI] [PubMed] [Google Scholar]
  63. Brynskov J., Foegh P., Pedersen G., Ellervik C., Kirkegaard T., Bingham A., Saermark T. Gut. 2002;51:37–43. doi: 10.1136/gut.51.1.37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Nakamura Y., Si Q. S., Kataoka K. Neurosci. Res. 1999;35:95–100. doi: 10.1016/s0168-0102(99)00071-1. [DOI] [PubMed] [Google Scholar]
  65. Pacher P., Beckman J. S., Liaudet L. Physiol. Rev. 2007;87:315–424. doi: 10.1152/physrev.00029.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Calabrese V., Mancuso C., Calvani M., Rizzarelli E., Butterfield D. A., Stella A. M. Nat. Rev. Neurosci. 2007;8:766–775. doi: 10.1038/nrn2214. [DOI] [PubMed] [Google Scholar]
  67. Nissen J. C., Selwood D. L., Tsirka S. E. J. Neurochem. 2013;127:394–402. doi: 10.1111/jnc.12404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Kobayashi K., Imagama S., Ohgomori T., Hirano K., Uchimura K., Sakamoto K., Hirakawa A., Takeuchi H., Suzumura A., Ishiguro N., Kadomatsu K. Cell Death Dis. 2013;4:e525. doi: 10.1038/cddis.2013.54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Emmetsberger J., Tsirka S. E. Neurobiol. Dis. 2012;47:295–309. doi: 10.1016/j.nbd.2012.05.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Caberoy N. B., Alvarado G., Li W. J. Neuroimmunol. 2012;252:40–48. doi: 10.1016/j.jneuroim.2012.07.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Ranson P. A., Thomas W. E. J. Histochem. Cytochem. 1991;39:853–858. doi: 10.1177/39.6.2033242. [DOI] [PubMed] [Google Scholar]
  72. Papa S., Ferrari R., De Paola M., Rossi F., Mariani A., Caron I., Sammali E., Peviani M., Dell'Oro V., Colombo C., Morbidelli M., Forloni G., Perale G., Moscatelli D., Veglianese P. J. Controlled Release. 2014;174:15–26. doi: 10.1016/j.jconrel.2013.11.001. [DOI] [PubMed] [Google Scholar]
  73. Colton C., Wilcock D. M. CNS Neurol. Disord.: Drug Targets. 2010;9:174–191. doi: 10.2174/187152710791012053. [DOI] [PubMed] [Google Scholar]
  74. Kraft A. D., Harry G. J. Int. J. Environ. Res. Public Health. 2011;8:2980–3018. doi: 10.3390/ijerph8072980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Pickard M. R., Chari D. M. Int. J. Mol. Sci. 2010;11:967–981. doi: 10.3390/ijms11030967. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Luther E. M., Petters C., Bulcke F., Kaltz A., Thiel K., Bickmeyer U., Dringen R. Acta Biomater. 2013;9:8454–8465. doi: 10.1016/j.actbio.2013.05.022. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Click here for additional data file. (959.5KB, pdf)

Articles from Toxicology Research are provided here courtesy of Oxford University Press

RESOURCES