Imidazole-stabilized gold nanoparticles induce neuronal apoptosis: An in vitro and in vivo study Roberta Imperatore,1 Gianfranco Carotenuto,2 Maria Antonietta Di Grazia,1 Ida Ferrandino,3 Letizia Palomba,4 Raffaella Mariotti,5 Emilia Vitale,6 Sergio De Nicola,7 Angela Longo,2 Luigia Cristino1 1

Institute of Biomolecular Chemistry, CNR of Pozzuoli, Naples, Italy Institute for Composite and Biomedical Materials, CNR of Naples, Naples, Italy 3 Department of Biological Sciences, University of Naples Federico II, Naples, Italy 4 Department of Biomolecular Sciences, University of Urbino “Carlo Bo”, Urbino, Italy 5 Department of Neurological and Movement Sciences, University of Verona, Verona, Italy 6 Institute of Protein Biochemistry, CNR of Naples, Naples, Italy 7 SPIN Institute, National Research Council, Complesso Universitario of M.S. Angelo, Naples, Italy 2

Received 18 March 2014; revised 11 June 2014; accepted 18 July 2014 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.35289 Abstract: Gold nanoparticles are increasingly being employed in innovative biological applications thanks to their advantages of material- and size-dependent physics and chemical interactions with the cellular systems. On the other hand, growing concern has emerged on the toxicity which would render gold-based nanoparticles harmful to cell cultures, animals, and humans. Emerging attention is focused on the interaction of gold nanoparticles with nervous system, especially regarding the ability to overcome the blood–brain barrier (BBB) which represents the major impediment to the delivery of therapeutics into the brain. We synthesized highly stable 2-mercapto-1-methylimidazole-stabilized gold-nanoparticles (AuNPs)-mmi to investigate their entry, accumulation, and toxicity in vitro (SH-SY5Y human neuroblastoma cells) and in vivo (brain of C57BL/6 mice) through optical and electron microscopy. After incubation in the cell culture medium at the lowest dose of 0.1 mg/mL the (AuNPs)-mmi nanoparticles were found compacted and recruited into endosome/

lysosomes (1 h) before their fusion (2 h) and the onset of neuronal death by apoptosis (4 h) as proved by terminaltransferase-mediated dUTP nick end labeling assay and caspase-3 immunoreactivity. The ability of (AuNPs)-mmi to cross the BBB was assessed by injection in the caudal vein of C57BL/6 mice. Among different brain regions, the nanoparticles were found in the CaudatoPutamen area, mainly in the striatal neurons 4 h after injection. These neurons showed the typical hallmarks of apoptosis. Our findings provide, for the first time, the dynamic of 2-mercapto-1-methylimidazole nanogold uptake. The molecular mechanism which underlies the nanogold-driven apoptotic event is analyzed and discussed in order to take into account when designing nanoC 2014 Wiley materials to interface with biological structures. V Periodicals, Inc. J Biomed Mater Res Part A: 00A:000–000, 2014.

Key Words: blood–brain barrier, neuroblastoma, cytotoxicity, thiolato ligands, apoptosis

How to cite this article: Imperatore R, Carotenuto G, Di Grazia MA, Ferrandino I, Palomba L, Mariotti R, Vitale E, De Nicola S, Longo A, Cristino L. 2014. Imidazole-stabilized gold nanoparticles induce neuronal apoptosis: An in vitro and in vivo study. J Biomed Mater Res Part A 2014:00A:000–000.

INTRODUCTION

Gold nanoparticles (AuNPs) hold great interest in nanomedicine because of their built-in and size-dependent physicochemical features. Stability, monodispersity, and excellent contrasting properties for transmission electron microscopy (TEM) analysis enable AuNPs to be traced with high dynamic resolution of uptake and intracellular sorting. Recent advances in chemical synthesis and biomolecular functionality have led to a massive expansion of AuNPs biomedical applications, including imaging, clinical diagnostic, and therapy.1–4 Special

attention has been paid to central nervous system (CNS) diseases, in which the blood–brain barrier (BBB) provides the strongest impediment to drug delivery into the brain. Indeed, more than 95% of in vitro potentially active compounds for the treatment of brain diseases are not able to cross the BBB in a pharmacologically sufficient concentration. The endothelial cells of the brain capillaries are uniquely interconnected by tight junctions in order to exclude the uptake of nanoparticles (NPs) driven by hydrostatic and osmotic gradients into the brain, whereas passive or active transcellular uptake can

RI, GC, AL, and LC conceived the experimental design and analyzed the data. MADG, RM, IF, LP, and EV participated in the design of the experiment and developed the sample preparation. IF and RI performed the ultrastructural study. AL and GC developed the synthesis and characterization of nanoparticles. SDN developed the model of in vitro uptake. LC wrote the article. All authors read and approved the final article. Correspondence to: L. Cristino; e-mail: [email protected] and A. Longo; e-mail: [email protected]

C 2014 WILEY PERIODICALS, INC. V

1

still occur. Therefore, the development of new strategies to enhance drug delivery to the brain is of great importance, and nanomaterial chemistry has dealt with this problem by developing nanoparticles able to cross the BBB and accumulate in the CNS.5–7 Limited to gold and bimetallic nanoparticles, low levels of 1-methylimidazole additives strongly improve the resulting stability in ionic liquid suspensions by preventing passivation of the catalytic surface.8 Many studies described the synthesis of imidazole derivatives, emphasizing their attractive fluorescence emitting properties.9–11 However, although the clinical use of gold nanoparticles has been proposed, only little is known about the potential toxicological effects, such as induction of inflammatory immune responses, apoptotic cell death or developmental problems in embryos. In the present study, we designed gold-nanoparticles stabilized with 2-mercapto-1-methylimidazole (-mmi), hereafter named (AuNPs)-mmi, and exploited their high stability and advantageous spectral properties to investigate their delivery, bioaccumulation and toxicity, in vitro (SH-SY5Y human neuroblastoma cells) and in vivo (nervous tissue from the brain of C57BL/6 mice). The electron dense properties of the metallic core enable the (AuNPs)-mmi nanoparticles to be easily traced at nanoscale resolution after cellular uptake in SH-SY5Y human neuroblastoma. Therefore, we compared the (AuNPs)-mmi cytotoxic effects [i.e., reactivity to caspase-3 and terminaltransferase-mediated dUTP nick end labeling (TUNEL) with the intracellular trafficking and accumulation of (AuNPs)mmi into putative endosome/lysosome bodies up to the final cellular death. Finally, we tested the (AuNPs)-mmi ability to cross the BBB by revealing their distribution into the midbrain area of brain, and induction of reactive gliosis after intravenous injection in the caudal vein of C57BL/6 mice compared to vehicle-injected mice. Our results show high efficiency of nanogold intracellular uptake, both in SHSY5Y human neuroblastoma and brain neurons. After incubation in the cell culture medium the (AuNPs)-mmi nanoparticles were rapidly (20 min) internalized into the cytoplasm, compacted, and recruited into endosome/lysosomes (1 h) before their fusion (2 h) and the onset of neuronal death by apoptosis (4 h) as proved by TUNEL assay and caspase-3 immunoreactivity. Giving the wide application of nano-tools able to enter BBB during in vivo clinical trials, a basic knowledge of their effects after neuronal interaction is crucial. Our findings provide, for the first time, the dynamic of 2-mercapto-1methylimidazole stabilized gold nanoparticles uptake and their neurotoxic effects by imaging their cellular tracing at light and electron microscopy. The molecular mechanism which underlies the nanogold-driven apoptotic event is analyzed and discussed in order to take into account when designing nanomaterials to interface with biological structures. METHODS

Synthesis and characterization of (AuNPs)-mmi The synthesis of (AuNPs)-mmi was carried out by Au13 reduction and precipitation in polar solvent according to

2

IMPERATORE ET AL.

standardized protocols.8,12,13 In particular, nontoxic, odorless, and high soluble in polar solvent thiol, 2-mercapto-1methylimidazole (C4H6N2S, Aldrich, 98%) was used. At the beginning, a bloody-red colored solution was obtained after mixing the acetonitrile thiol solution with a tetrachloroauric(III) acid trihydrate (HAuCl43H2O, Aldrich, 99.9%) acetonitrile solution under magnetic stirring, at room temperature without precipitate formation. Successively, addition of few drops of an ammonium hydroxide aqueous solution (28.0– 30.0% NH3 basis Aldrich) caused the precipitation of a white microcrystalline product. The reaction between thiol and tetrachloroauric(III) acids was given by nHAuCl4 13nC4 H5 N2 2SH ! ðAuÞn 2ðS2C4 H5 N2 Þn 1 ðC4 H5 N2 SÞ2n 1 4nHCl The system was left under magnetic stirring for two days. This reaction allowed to obtain stable gold nanoparticles passivated by organic shell of mercapto-methylimidazole to prevent sintering. The white solid precipitate was separated by vacuum-filtration from other products by washing in acetonitrile. Morphological characterization of (AuNPs)-mmi was performed by means of the TEM microscopy by using the Philips EM208S microscope operating through accelerating voltage of 100 kV and equipped with a MegaView digital Camera. The optical properties of (AuNPs)-mmi were examined by recording the photoluminescence spectra using an UV-Visible spectrofluorometer (Perkin-Elmer-LS55) which allows the direct investigation of the as-prepared samples as power specimens. In vivo injection of (AuNPs)-mmi The experiments were performed according to the guidelines of the institutional ethical code and the Italian (D.L. 116/92) and European Directive 2010/63/EU revising Directive 86/609/EEC on the protection of animals used for scientific purposes. N 5 24 C57BL/6 adult male mice were housed, three per cage, under controlled illumination (12-h light/12-h dark cycle; lights on at 6:00 AM) and standard environmental conditions (ambient temperature 20–22 C, humidity 55–60%). Mice chow and tap water were available ad libitum. All efforts were made to reduce both animal numbers and suffering during the experiments. The nanoparticles were suspended in 0.9% NaCl saline to a final concentration of 1 mg/mL and 0.5 mL of suspension was injected in the caudal vein resulting in a dose of 25 mg/kg body weight. Evaluation of dose-response of (AuNPs)-mmi emission fluorescence was carried out from slices of brain tissue of mice injected also with lower nanoparticles concentration (15 mg/kg) without any detection up to 6 h from treatment. N 5 12 mice were injected with the (AuNPs)-mmi suspension; n 5 6 controls were injected with equal volume of nonimidazole coated AuNPs (hereafter named AuNPs) suspension and n 5 6 controls were injected with equal volume of saline. The mice were sacrificed 4 h (n 5 3) and 6 h (n 5 3) after nanoparticles or vehicle injection by cervical dislocation and the brains were removed

IMIDAZOLE-STABILIZED GOLD NANOPARTICLES

ORIGINAL ARTICLE

and fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4) up to 24 h before cryoprotection in 30% sucrose solution, freezing, and cutting. Cell culture and (AuNPs)-mmi uptake The human neuroblastoma cells SH-SY5Y were grown on 100 mm diameter Petri dishes as monolayers in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 2 mM L-glutamine,1% penicillin (20 units/mL), streptomycin (20 mg/mL), and containing 10% (vol/vol) heat-inactivated fetal bovine serum (FBS). Cells were maintained at 37 C in a saturated humidity atmosphere containing 95% air and 5% CO2, plated overnight on poly-D-lysine coated cover slips and, then, incubated for 20 min, 1 h, 2 h, and 4 h at 37 C with 1 mL of cellular culture medium containing (AuNPs)-mmi (0.1 mg/350.000 cells) or nonimidazole coated AuNPs or in medium only. This (AuNPs)-mmi concentration was fixed according to the dose-response determination of the quenching effect on nanoparticles emission spectra. Notably, 0.1 mg/mL (AuNPs)-mmi was the lowest biologically efficient concentration without any quenching effect on the fluorescence emission spectra. Furthermore, this dose was comparable to that used in vivo considering that a putative dispersion of (AuNPs)-mmi in the extracellular matrix, interstitial fluid and perivascular space can occur in living animals.14 The optical density (OD) of (AuNPs)mmi subcellular fluorescence emission was quantified on n 5 500 SH-SY5Y cells, for each treatment and time point, by exposition to excitation wavelength 510–576 nm and quantification by Leica-MetaMorph LASAF2.2.0 software. Moreover, the samples were observed in DIC/FLUO modality by coupling the bright differential interference contrast (DIC/DMI6000-Leica) with the fluorescence fields. All images were recorded by a digital camera Leica DFC 340FX. Lysosomal uptake of (AuNPs)-mmi The lysosomal acidic organelles of SH-SY5Y cells were stained and tracked with LysoTrackerV Green DND-26 (Invitrogen) 200 nM after 30 min of incubation in serum free media. After the lysosomal staining, the cells were rinsed in PBS and incubated in culture medium containing (AuNPs)mmi (0.1 mg/350.000 cells) or nonimidazole coated AuNPs (0.1 mg/350.000 cells) or medium only for 20 min, 1 h, 2 h, and 4 h. Counterstaining with DAPI (40 ,6-diamidino-2-phenylindole), a cell permeable fluorescent dye that binds to DNA, was performed before observation at microscope. R

Ultrastructural localization To study the cytological effect of (AuNPs)-mmi at the ultrastructural level, SH-SY5Y cells were incubated for 20 min, 1 h, 2 h, and 4 h at 37 C with 1 mL of cellular culture medium containing (AuNPs)-mmi (0.1 mg/350.000 cells) or nonimidazole coated AuNPs or in medium only. The pellet was concentrated by centrifugation at 3000g, collected, and processed for TEM analysis. Briefly, the pellet of nanoparticle-treated or medium-treated cells was fixed in a mixture of 4% paraformaldeyde/2.5% glutaraldehyde and postfixed with OsO4 (0.5% in PB, 10 min at 4 C), dehydrated in an ascending

series of ethanol and propylene oxide and counteracted with 1% uranyl acetate in 70% ethanol (15 min at 4 C). Finally, cells were embedded in TAAB 812 resin (TAAB, England), sectioned in ultrathin sections (60 nm) at ultramicrotome LKB (Leitz, Germany) and collected on Formvar-coated slot grids before treatment with 0.65% lead citrate. Electron micrographs were taken with the TEM microscope LEO 912AB (Zeiss, Germany). Cytotoxicity assays: Caspase-3 immunoreactivity and TUNEL To test the apoptotic effect of nanoparticles, caspase-3 immunostaining and the TUNEL assay were performed on cells and brain slice tissue. The brains of AuNPs-mmi-injected mice and respective controls, were dissected 4 h and 6 h after injection and cut in serial coronal frozen sections (10 lm-thick) collected in three alternate series and processed, respectively, for NeuN (a neuronal marker; dilution 1:500; Abcam, Cambridge, UK), caspase-3 (dilution 1:100; Santa Cruz Biotechnology, Santa Cruz, CA). SH-SY5Y cells were incubated 20 min, 1 h, 2 h, and 4 h in a culture medium containing (AuNPs)-mmi (0.1 mg/350.000 cells) or nonimidazole coated AuNPs (0.1 mg/350.000 cells) or medium only and caspase-3 reactivity was revealed by Alexa-488 donkey anti-IgG secondary coupled with DAPI. In a parallel study the TUNEL assay was determined using a commercial kit (Roche-Diagnostics GmbH, Germany) in accordance with the manufacturer’s instructions. Statistical analyses Data were expressed as mean of standard deviation. A repeated two-way analysis of variance was performed using SPSS version 8.2. Student’s or Bonferroni’s two tail t-test were used for statistical analysis (*p < 0.05, **p < 0.001, and ***p < 0.0001). In vitro experiments were performed in triplicate specimens and repeated 5–6 times for each selected time point. The quantitative analysis of TUNEL assay or intensity of fluorescence emission were carried out on n 5 500 cells randomly sampled from these in vitro experiments. RESULTS

Morphological and optical characterization of (AuNPs)-mmi The nanoparticles appeared to be spherically shaped by TEM imaging [Fig. 1(A)]. The size distribution histogram and corresponding Gaussian curve [Fig. 1(B)] were obtained by analysis of the TEM images (Sigma Scan Pro 5 image software). The average diameter of the (AuNPs)-mmi was 2.7 6 0.7 nm. The photoluminescence phenomenon which occurs by UV excitation determined an high intensity color transition of nanoparticles from white to green-yellow. The fluorescence of the pure powder was stable at room temperature over six months. The optical properties of (AuNPs)-mmi were examined by recording their photoluminescence (PL) spectra by using an UV-visible spectrofluorimeter (Perkin Elmer, LS55). The PL spectra exhibited

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00

3

FIGURE 1. A: TEM micrograph of (AuNPs)-mmi nanoparticles; Scale bar 5 20 nm. B: Size distribution of AuNPs-mmi nanoparticles; the red curve represents a Gaussian fitting with the experimental results. C: Excitation (triangle-line) and emission (square-line) spectra of the AuNPs-mmi. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

an excitation band extending from 300 to 400 nm with a maximum at 384 nm [Fig. 1(C), triangle-line] and a broad emission band extending from 400 to 650 nm, with a peak at 530 nm [Fig. 1(B), square–line]. The red curve represents the curve fitting with Gaussian function to the experimental recorded emission spectrum [Fig. 1(C), square–line]. The PL spectra of (AuNPs)-mmi exhibited a clear Stokes shift, that is, the difference in wavelength between positions of the maximum excitation and maximum emission peaks, of Dk 5 146 nm. Neuronal in vitro uptake of (AuNPs)-mmi As shown in Figure 2, a fast interaction between (AuNPs)mmi nanoparticles and SH-SY5Y cells was detected after 20 min of incubation at concentration of 0.1 mg/mL, when many cells (80.5 6 9.4%) exhibited a strong red punctuate signal in the cytoplasm under selective excitation at 510– 576k and not at 410–500k or 320–360k. This effect became stronger 1 h after incubation and peaked 2 h after incubation (96.1 6 10.9%). Over a longer period (4 h), loss of (AuNPs)-mmi bioaccumulation was observed (Fig. 2). Incubation with medium or nonimidazole coated AuNPs did not reveal any specific fluorescent signal during the same time course. (AuNPs)-mmi were frequently found in lysosomes because of coexpression of the green punctuate fluorescent signal of lysotracker with the red signal of (AuNPs)-mmi

4

IMPERATORE ET AL.

[Fig. 3(A,B); yellow dots]. The DIC/FLUO microscopy analysis provided a topological map of the intracellular distribution and accumulation of nanoparticles in terms of intracellular fluorescence density [Fig. 3(C,D)], showing a time-dependent increase of (AuNPs)-mmi intracellular uptake and a peak of intensity at 2 h by comparison with nonimidazole coated AuNPs [Fig. 3(E)]. This first set of results provided evidence of high advantageous spectral fluorescence emitting properties of (AuNPs)-mmi compared with bare AuNPs. In order to test the putative toxicological effects of (AuNPs)-mmi on biological systems, a morphological study at ultrastructural investigation was performed by a time course treatment of SH-SY5Y with (AuNPs)-mmi compared to bare nanoparticles. We found that (AuNPs)-mmi lead to cytotoxicity and apoptosis by inducing loss of regular shape and cellular morphology 4 h after treatment (Fig. 4). Notably, at difference with (AuNPs)-mmi, the bare AuNPs were found as electron dense clusters inside the cytoplasm of SHSY5Y cells without any associated toxicological or apoptotic morphological hallmarks, up to 4 h of treatment. In particular, by comparing the morphology of the vehicle-treated cells [Fig. 4(A)] with that of (AuNPs)-mmi-treated cells [Fig. 4(B–D)] we found the latter with a typical scalloped cell membrane equipped with branching extending into the extracellular space, 1 h and 2 h after incubation

IMIDAZOLE-STABILIZED GOLD NANOPARTICLES

ORIGINAL ARTICLE

FIGURE 2. Time course of (AuNPs)-mmi nanoparticles uptake into SH-SY5Y cells. A: Representative micrographs of merged images acquired by differential interference contrast microscopy (DIC) coupled to different wavelengths excitation: 410–500k (B; control wavelengths excitation), 510–576k (C; nanoparticles wavelengths excitation), 320–360k (D; DAPI wavelengths excitation). Only the DAPI fluorescent signal (D) was recorded from cell medium-treated samples excited at 410–500k (B) and 510–576k (C) wavelengths. The nanoparticles (red signals) were evident inside the cells after 20 min (AuNPs)-mmi incubation and fluorescence became stronger after 1 h or 2 h. The intracellular fluorescent emitting signal was lost after 4 h incubation with (AuNPs)-mmi. Scale bar 5 10 mm. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00

5

FIGURE 3. Cellular uptake of (AuNPs)-mmi. A and B: Inward trafficking of (AuNPs)-mmi into SH-SY5Y cells lysosomes. Representative microV graphs of SH-SY5Y cells co-stained with DAPI (blue) and LysoTracker (green) in the absence (A) or presence (B) of (AuNPs)-mmi (red). C and D: Representative images of DIC/FLUO microscopy and topographical map of living SH-SY5Y cells (C1,D1) treated for 4 h with cell medium without (C) or with (D) (AuNPs)-mmi 0.1 mg/350.000 cells. Each pixel color-value represents the measure of light interferential amount in the matching point of the sample shown in the boxed areas C1 and D1, respectively. Codified-color rendering improved the estimate of topographical and densitometric interferential value. Scale bar 5 15 mm in (A)–(D). E: Time course of the (AuNPs)-mmi red fluorescence in SH-SY5Y cells. The peaks of intensity were recorded after 1 h or 2 h incubation with (AuNPs)-mmi . Values of intensity were obtained from n 5 5100 cells per each treatment and time point analyzed as in (D). ##p < 0.001 1 h and 2 h (AuNPs)-mmi-treated cells versus 20 min or 4 h-treated cells; *p < 0.05 and ***p < 0.0001 (AuNPs)-mmi-treated versus respective nonmmi-coated AuNP-treated cells. Each value was normalized with respect to the optical density measured from medium-treated cells at different time points. The intensity of fluorescence was expressed in arbitrary units and the mean value for each time points and treatment was normalized to the corresponding baseline value of fluorescence emitted by SH-SY5Y cells cultured in medium only, without nanoparticles. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] R

6

IMPERATORE ET AL.

IMIDAZOLE-STABILIZED GOLD NANOPARTICLES

ORIGINAL ARTICLE

FIGURE 4. Representative TEM micrographas of SH-SY5Y cells treated 4 h in the absence (A) or presence of (AuNPs)-mmi for 1 h (B), 2 h (C), and 4 h (D). Numerous highly dense and dark putative endosome-lysosome structures were present inside the cytoplasm after (AuNPs)-mmi incubation. Scale bar5 2 mm in (A)–(C). Representative endosome–lysosome structures of untreated (A1) or nanoparticle-treated (C1) cells were shown as high magnification in respective insets of (A) and (C). Scale bar 5 0.15 mm in (A1) and (C1). E: Analysis of different size distribution of aggregated (AuNPs)-mmi nanoparticles as a function of the intracellular distance from the plasma membrane. Scale bar 5 0.5 mm. F: Distribution of the average particle size in different cell areas (1, 2, 3, and 4) shown in (E). The straight line in (E) was drawn starting from the cell membrane through four different compartments labeled as 1, 2, 3, and 4 on the TEM image. The distribution of the average size of each aggregate plotted against its intracellular distance from the cell membrane fitted with an exponential function (red line) and showed the highest frequency of distribution (10–20%) for the largest size of nanoclusters (140–240 nm diameter) in the inner cellular compartments 3 and 4 (F). The exponential increase in aggregate size with the intracellular distance was suggestive of a growth mechanism in which the rate of aggregation was directly proportional to the aggregate size and to the intracellular distance travelled according to a constant value of 4.2 nm/lm. Quantitative analysis of (AuNPs)-mmi aggregation performed 4 h after cell incubation with (AuNPs)-mmi after processing TEM images with the Sigma Scan Pro 5 software (E and F). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00

7

FIGURE 5. (AuNPs)-mmi induce apoptosis in SH-SY5Y cells. A–E: Representative micrographs of cells treated with (AuNPs)-mmi for 4 h and then processed for caspase-3 immunoreactivity. DIC analysis is shown in (A). B: Merged image of red (AuNPs)-mmi fluorescence (C) colocalizing with the green signal of caspase 3 (D) and DAPI (E). Scale bar 5 10 mm. F and G: TUNEL assay in SH-SY5Y cells. Table I summarizes the percentage of TUNEL positive or negative cells. Most of the (AuNPs)-mmi-treated cells were TUNEL positive. The control cells were mainly TUNEL negative. Scale bar 5 20 mm. The TUNEL assay was performed by comparing the effect on the cell viability of 4 h treatment in medium containing (AuNPs)-mmi (0.1 mg/350.000 cells) to 4 h treatment in medium containing nonimidazole coated AuNPs (0.1 mg/350.000 cells) or medium only, as controls. As summarized in Table I, the majority of TUNEL negative cells (93.3 6 15.5%) was found in the control samples, both treated with medium or AuNPs; few cell bodies were TUNEL positive for medium or AuNPs treatment as 6.3 6 2.2% and 8.5 6 2.6%, respectively (F, arrows). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

[Fig. 4(B,C)]. Noteworthy, numerous large vacuoles in progress of fusion/division [Fig. 4(B) and arrowheads in (E)] were found in the cytoplasm, very close to the plasma membrane from which they originate by invagination and fusion (macropynocytosis). Interestingly, starting 20 min after incubation, the (AuNPs)-mmi were detected inside the cells and were found progressively packed into micrometer-sized granular structures like endosome/lysosome bodies after 1 h and 2 h incubation [Fig. 4(B,C)]. All the typical ultrastructural features of apoptosis, that is, cell shrinkage, chromatin condensation, nuclear collapse, and membrane blebbing, were detected 4 h after (AuNPs)-mmi incubation, probably due to the self-folding and fusion of multiple units of AuNP aggregates [Fig. 4(D)]. Quantitative analysis of (AuNPs)-mmi aggregation 4 h after cell incubation showed the highest

8

IMPERATORE ET AL.

frequency of distribution (10–20%) for the largest size of nanoclusters (140–240 nm diameter) in the inner cellular compartments 3 and 4 [Fig. 4(F)]. The statistical analysis was performed over 10 TEM images of (AuNPs)-mmi nanoparticles distribution within the cells. The exponential increase in aggregate size with the intracellular distance was calculated as a constant rate value of growth of 4.2 nm/lm. (AuNPs)-mmi lead cell death by apoptosis A significant increase of caspase-3 expression was found in 92.6 6 10.5% of SH-SY5Y cells 4 h after treatment with nanoparticles, revealing the activation of apoptosis by the (AuNPs)-mmi, and confirmed by the highest percentage of TUNEL positive cells (92.8 6 14.5%) among the cells

IMIDAZOLE-STABILIZED GOLD NANOPARTICLES

ORIGINAL ARTICLE

FIGURE 6. Brain distribution of (AuNPs)-mmi after intravenous injection in the caudal vein of mice. A–A3: Representative images of forebrain slices (Caudato Putamen region, CPu) of control samples after 4 h of vehicle injection. Note the neuronal body showing the neuronal marker expression of NeuN. Time-dependent AuNPs-mmi-induced apoptosis in the brain. B–B3: Note the pink-fuchsia fluorescent signal in the vast majority of striatal neurons 4 h after AuNPs-mmi injection (arrows) as coexpression of fluorescent nanoparticles signal (red) with NeuN reactivity (blue) whereas caspase3-reactivity was not found. C–C3: Note the occurrence of apoptosis 6 h after AuNPs-mmi injection, as indicated in neurons by the light blue/white fluorescent signal of merged blue (NeuN), red (AuNPs), and green (caspase-3) reactivity (arrows in C3). Scale bar 5 40 mm. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

exhibiting (AuNPs)-mmi internalization (85.2 6 9.2%). Moreover, a small percentage of TUNEL negative cells was also found (10.56 3.5%) including cells without (AuNPs)-mmi internalization [(6.6 6 1.4%; Fig. 5(G), black arrowhead]. TUNEL positive cells (AuNPs)-mmi-negative [7.6 6 1.8%; Fig. 5(G), black arrows] and TUNEL negative cells (AuNPs)mmi-positive [3.2 6 0.5%; Fig. 5(G), white arrowheads] were also found. Notably, the majority of TUNEL negative cells was found in control samples as 95.5 6 12.5% and 93.3 6 15.5% up to 4 h of treatment respectively in cell

medium alone or with bare AuNPs. Moreover, few cell bodies (6.3 6 2.2% and 8.5 6 2.6%, respectively, for medium or AuNPs treatment) were TUNEL positive [Fig. 5(F), arrows]. In vivo distribution of (AuNPs)-mmi in mouse brain The forebrain and midbrain regions of the mouse brain studied for analysis of (AuNPs)-mmi distribution 4 h and 6 h from intravenous injection showed the strongest particleassociated fluorescence starting from 4 h after injection [Fig. 6(B–B3)]. The fluorescence was mainly detected from

TABLE I. Percentage of TUNEL Reactive Cells n5500 Cells

Cell Medium (4h)

1AuNPs (4h)

1(AuNPs)-mmi(4h)

6.3%62.2% 95.5%612.5%

8.5%62.6% 93.3%615.5%

92.8%614.5% [85.2%69.2% (AuNPs)-mmi positive] 10.5%63.5% [3.6%61.4% (AuNPs)-mmi negative]

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00

9

TUNEL positive TUNEL negative

neurons of the Caudate Putamen area (CPu) with a more evenly spread expression of reactive gliosis (data not shown). The immunohistochemical analysis of caspase-3 reactivity coupled to Neuronal NeuN markers showed the highest percentage of NeuN/Capsase3-reactive striatal neurons 6 h after (AuNPs)-mmi injection [arrows in Fig. 6(C– C3)]. The parallel experiment performed with bare AuNPs injection didn’t exhibit particle-associated fluorescence in any brain regions 4 h and 6 h from intravenous injection. Notably, no caspase-3 immunoreactivity was found at the same time points (data not shown).

DISCUSSION

The TEM analysis shows the property of 2.7 nm (AuNPs)mmi to aggregate in larger molecular clusters with a strong emission of red fluorescence after cellular internalization. According to these data, we have hypothesized a possible model of (AuNPs)-mmi cellular internalization in which single AuNP-mmi, after endocytosis, accumulate as multiple aggregates of 200–250 nm into endosomial and/or lysosomal bodies, before cell degradation. The (AuNPs)-mmi aggregates trigger the apoptotic pathway as characterized by caspase-3 activation, TUNEL reactivity and emergence of morphological hallmarks of apoptosis. Over 4 h of (AuNPs)mmi incubation a significant reduction of (AuNPs)-mmi fluorescence is observed as possible consequence of the lysosomal degrading effect on 200–250 nm sized (AuNPs)-mmi clusters rather than the occurrence of quenching phenomena. Indeed, 0.1 mg/mL (AuNPs)-mmi was the lowest biologically efficient concentration for which any quenching effect on fluorescence emission spectra was observed. The reduction of (AuNPs)-mmi fluorescence over 4 h of (AuNPs)-mmi is further supported by ultrastructural data showing, simultaneously, the presence of large empty vacuoles in the cells. These empty vacuoles are observed as possible consequence of the lysosomal degrading effect on 200–250 nm sized (AuNPs)-mmi clusters, coupled to activation of caspase-3 reactivity. All together, these morphological features identify the occurrence of apoptosis which is classically characterized by cell chromatin condensation, plasma-membrane asymmetry, and cell detachment from the matrix. This cellular suicide program is initiated by environmental stresses, and results in adaptor-protein-induced activation of initiator caspases, which culminates in the orchestrated disassembly and phagocytosis of the dying cell.15 The exponential increase of the size of (AuNPs)-mmi clusters with a constant value of 4.2 nm/lm is suggestive of a growth mechanism in which the rate of aggregation is directly proportional to the intracellular distance from the site of entry and the timing of onset of apoptotic events (i.e., caspase-3 and TUNEL reactivities). The most intriguing finding is the punctuate pattern exhibited by (AuNPs)-mmi 1 h after cellular incubation with SH-SY5Y cells (Fig. 2), which may arise from a putative mechanism of inward budding of the membrane limiting endosomal-like structures, as described in mammals.16 This assumption is supported by the occurrence of electron-dense globular substances of nm

10

IMPERATORE ET AL.

size (20–30 nm) detected in self-folding clusters of 200 nm diameter inside endosomal bodies such as lysosomes, wherein these aggregates can be clearly distinguished. In all cases, most of the 20–30 nm smaller aggregated particles are clearly identified also in the cytosol according to the kinetic study of De Jong et al.17 Many drugs are not able to permeate the BBB, and thus the development of strategies to deliver drugs to the CNS is very important.18 The present data indicate that (AuNPs)-mmi are strikingly able in vivo to cross the BBB after intravenous injection and form clusters when inside cells. However, (AuNPs)-mmi are neurotoxic, both in vitro and in vivo, and can induce apoptosis by accumulating and freely diffusing into the cytoplasm over a long period of time. According to previous in vitro studies, showing (NPs)–SiO2- and (NPs)–TiO2-induced cellular dysfunction and damage selectively in neurons through activation of the P53 and/or JNK pathways,19–21 we show here that (AuNPs)-mmi induce neuronal toxicity also in vivo 6 h after treatment. As intravenous route of injection requires that (AuNPs)mmi has to enter in blood circle before reach the CNS, cross the BBB and enter inside the neurons, it is likely to observe apoptotic events in vivo 2 h after their appearance in vitro. Studies of the biodistribution after intravenous injection of small (1.4-nm) AuNPs show that they are excreted by the kidneys as well as by the hepatobiliary system and no specific uptake in any organs could be observed. However, a significant amounts of small AuNPs circulate in the blood even at 24 h postintravenous injection.22,23 In summary, we have here shown, for the first time, a fast neurotoxic effect of small 2.7 nm sized (AuNPs)-mmi able to translocate through the blood into the brain, overcoming the BBB in significant amounts to lead apoptotic effects. The rapid neurotoxic effect of (AuNPs)-mmi is likely due to their stabilization by imidazole derivatives since it does not occur 4–6 h after injection of nonimidazole-coated nanoparticles, that is, bare AuNPs. The imidazole-coated AuNPs offer advantageous detection features being directly imaged by fluorescence at difference with bare, nonimidazole-coated, AuNPs. Both properties might be important in the discussion concerning risk assessment of gold nanoparticles. Further studies will be necessary to answer the remaining questions regarding the uptake and the biodistribution of (AuNPs)-mmi nanoparticles. In fact, the potentially toxic effects of various inorganic nanoparticles on neuronal function mediated by glial cells remain largely unknown, especially for those nanoparticles, like (AuNPs)-mmi, with starting diameters of less than 20 nm. It is unclear if these nanoparticles cause injury to neurons by an indirect pathway, such as the activation of microglial cells and subsequent production of proinflammatory factors. Indeed, various inorganic nanoparticles (SiO2-NPs, TiO2-NPs, HAP-NPs, and Fe3O4-NPs) of 20– 60 nm of diameter were found to be neurotoxic through microglial activation in vitro by activation of the NF-jBrelated signal cascade and subsequent increase in the expression of downstream proinflammatory mediators.21

IMIDAZOLE-STABILIZED GOLD NANOPARTICLES

ORIGINAL ARTICLE

ACKNOWLEDGMENTS

We are grateful to the Italian National Research Council (CNR) for financial support. We are grateful to Endocannabinoid Research Group of ICIB-CNR and Prof. Vincenzo Di Marzo for scientific support. REFERENCES 1. Daniel MC, Astruc D. Gold nanoparticles: Assembly, supramolecular chemistry, quantum-size-related properties, and applications toward biology, catalysis, and nanotechnology. Chem Rev 2004; 104:293–346. 2. Kim CK, Ghosh P, Rotello V. Multimodal drug delivery using gold nanoparticles. Nanoscale 2009;1:61–68. 3. Chithrani DB. Intracellular uptake, transport, and processing of gold nanostructures. Mol Membr Biol 2010;27:299–311. 4. Pelaz B, Grazu V, Ibarra A, Magen C, del Pino P, de la Fuente JM. Tailoring the synthesis and heating ability of gold nanoprisms for bioapplications. Langmuir 2012;28:8965–8970. €le D, D’Angelo J, Noel JP, 5. Calvo P, Gouritin B, Chacun H, Desmae Georgin D, Fattal E, Andreux JP, Couvreur P. Long-circulating PEGylated polycyanoacrylate nanoparticles as new drug carrier for brain delivery. Pharm Res 2001;18:1157–1166. 6. Kreuter J, Shamenkov D, Petrov V, Ramge P, Cychutek K, KochBrandt C, Alyautdin R. Apolipoprotein-mediated transport of nanoparticle-bound drugs across the blood–brain barrier. J Drug Targeting 2002;10:317–325. 7. Ulbrich K, Hekmatara T, Herbert E, Kreuter J. Transferrin- and transferrin-receptor-antibody-modified nanoparticles enable drug delivery across the blood–brain barrier (BBB). Eur J Pharm Biopharm 2009;7:251–56. 8. Dash P, Scott RWJ. 1-Methylimidazole stabilization of gold nanoparticles in imidazolium ionic liquids. Chem Commun 2009;7:812–814. 9. Fabretti AC, Giusti A. 2-mercapto-5-methyl-1,3,4-thiadiazole as a ligand in Ru(III), Rh(III), Os(III) and Ir(III) trichloride complexes. Polyhedron 1986;12:1927–1930. 10. Surga WJ, Wisniewski MZ, Adach AG. Thermal and structural properties of coordination compounds of 2-mercapto-1Methylimidazole with palladium(II) and platinum(II). J Thermal Anal 1995;44:697–705.

11. Kalimuthu P, John SA. Studies on ligand exchange reaction of functionalized mercaptothiadiazole compounds onto citrate capped gold nanoparticles. Mater Chem Phys 2010;122:380–385. ~ oz Javier A, Gaub 12. Kirchner C, Liedl T, Kudera S, Pellegrino T, Mun € lzle S, Fertig N, Parak WJ. Cytotoxicity of colloidal CdSe HE, Sto and CdSe/ZnS nanoparticles. Nano Lett 2004;5:331–338. 13. Pan Y, Neuss S, Leifert A, Fischler M, Wen F, Simon U, Schmid G, Brandau W, Jhanen-Dechent J. Size-dependent cytotoxicity of gold nanoparticles. Small 2007;3:1941–1949. 14. Huang K, Ma H, Liu J, Huo S, Kumar A, Wei T, Zhang X, Jin S, Gan Y, Wang PC, He S, Zhang X, Liang XJ. Size-dependent localization and penetration of ultrasmall gold nanoparticles in cancer cells, multicellular spheroids, and tumors in vivo. ACS Nano 2012;6:4483–4493. 15. Chipuk JE, Green DR. Do inducers of apoptosis trigger caspaseindependent cell death? Nat Rev Mol Cell Biol 2005;6:268–275. 16. Piper RC, Katzmann DJ. Biogenesis and function of multivesicular bodies. Ann Rev Cell Dev Biol 2007;23:519–547. 17. De Jong WH, Hagens WI, Krystek P, Burger MC, Sips AJ, Geertsma RE. Particle size-dependent organ distribution of gold nanoparticles after intravenous administration. Biomaterials 2008; 29:1912–1919. 18. Shilo M, Motiei M, Hana P, Popovtzer R. Transport of nanoparticles through the blood–brain barrier for imaging and therapeutic applications. Nanoscale 2014;6:2146–2152. 19. Wu J, Sun J, Xue Y. Involvement of JNK and P53 activation in G2/M cell cycle arrest and apoptosis induced by titanium dioxide nanoparticles in neuron cells. Toxicol Lett 2010;199: 269–276. 20. Wu J, Sun J. Investigation on mechanism of growth arrest induced by iron oxide nanoparticles in PC12 cells. J Nanosci Nanotechnol 2011;11:11079–11083. 21. Xue Y, Wu J, Sun J. Four types of inorganic nanoparticles stimulate the inflammatory reaction in brain microglia and damage neurons in vitro. Toxicol Lett 2012;214:91–98. 22. Semmler-Behnke M, Kreyling WG, Lipka J, Fertsch S, Wenk A, Takenaka S, Schmid G, Brandau W. Biodistribution of 1.4- and 18nm gold particles in rats. Small 2008;4:2108–2111. 23. Khlebtsov N, Dykman L. Biodistribution and toxicity of engineered gold nanoparticles: A review of in vitro and in vivo studies. Chem Soc Rev 2011;40:1647–1671.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00

11