Science of the Total Environment 432 (2012) 382–388

Contents lists available at SciVerse ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r .com/ locate /sc i totenv

In vitro biocompatibility of solid lipid nanoparticles

Adny Henrique Silva a, Fabíola Branco Filippin-Monteiro a, Bruno Mattei a,Betina G. Zanetti-Ramos b, Tânia Beatiz Creczynski-Pasa a,⁎a Departamento de Ciências Farmacêuticas, Universidade Federal de Santa Catarina P.O. Box 476, Florianópolis, SC, 88040-900, Brazilb Nanovetores - Encapsulados de Alta Tecnologia, SC 401, km1, ParqTecAlfa, Ed. Celta, 4° andar, Florianópolis, SC, 88030‐000, Brazil

⁎ Corresponding author. Tel.: +55 48 3721 2212; faxE-mail addresses: tania.pasa@ufsc.br, taniabcp@gma

0048-9697/$ – see front matter © 2012 Elsevier B.V. Alldoi:10.1016/j.scitotenv.2012.06.018

a b s t r a c t

a r t i c l e i n f o

Article history:Received 1 July 2011Received in revised form 1 June 2012Accepted 6 June 2012Available online xxxx

Keywords:SLNCytotoxicityVero cellsMDCK cells

This study was undertaken to address the current deficient knowledge of cellular response to solid lipidnanoparticles (SLNs) exposure. We investigated the cytotoxicity of several SLNs formulations in two fibro-blast cell lineages, Vero and MDCK. Several methods were used to explore the mechanisms involved in thiscytotoxic process, including cell viability assays, flow cytometry and ROS generation assessment. Amongnanoparticles tested, two of them (F4 and F5) demonstrated more cytotoxic effects in both cell lineages.The cell viability assays suggested that F4 and F5 interfere in cell mitochondrial metabolism and in lysosomalactivity. In addition, F5 decreased the percentage of MDCK cells in G0/G1 and G2/M phases, with a markedincrease in the Sub/G1 population, suggesting DNA fragmentation. Regarding F4, although IC50 was higher(~700 μg/mL), this formulation affected mitochondrial membrane potential for Vero cells. However, theIC50 of F5 was around 250 μg/mL, suggesting the effect of SDS (sodium dodecyl sulfate) present in the formu-lation. In summary, the nanoparticles tested here appears to be biocompatible, with the exception of F5.Further studies are required to elucidate the in vivo effects of these nanoscale structures, in order to evaluateor predict the connotation of their increased and widespread use.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Nanotechnology has a wide range of potential applications inmany different areas and the safety of nanomaterials became amajor concern. Nanoparticles are present in commercially availableproducts, including cosmetics, food, clothes, and military equipment,as well as being increasingly used in medicine for diagnostic pur-poses, imaging and drug delivery. Current statistics show that morethan 1000 products or product lines are available on the marketusing nanotechnology in their production (Hsiao and Huang, 2011).Nanoparticles have unusual physicochemical properties, such assmall size, surface area and size distribution, chemical composition,purity, crystallinity, electronic properties, surface reactivity, solubili-ty, shape and aggregation (Nel et al., 2006); and it is precisely theseproperties that make nanoparticles so attractive. It is believed thatthe small size of nanoparticles allows them to enter and pass throughthe tissues, cells and organelles, and the size of some nanoparticlesis similar to many biological molecules such as proteins, and to microor-ganisms such as viruses (Fadeel and Garcia-Bennett, 2010). In biomedicalarea, the conception of nanotechnology embraces structures sized below1 μm and can consist of different materials (Elzoghby et al., 2012; Kayseret al., 2005; Martins et al., 2012; Videira et al., in press).

: +55 48 3721 2212.il.com (T.B. Creczynski-Pasa).

rights reserved.

Solid lipid nanoparticles (SLNs) have been widely used innanomedicine, e.g. cosmetic and skin formulations (Pardeike et al.,2009), for cancer treatment (Joshi and Müller, 2009; Wang andThanou, 2010), for lung cancer therapy (Chattopadhyay et al.,2007), for oral and ocular drug delivery (Müller et al., 2006), forgene therapy (Kwon et al., 2008) and for imaging, among others.SLNs consist of a matrix composed of lipids, which remain in solidstate at both room and human body temperatures, designed to encap-sulate lipophilic and hydrophilic drugs. Their preparation is basedon an original melt homogenization technique, which allows the pro-duction of nanocarriers in an aqueous surfactant solution (Joshi andMüller, 2009).

Humans have been exposed to airborne nanosized particlesthroughout their evolutionary stages, but exposure has increaseddramatically over the last century due to anthropogenic sources.The rapid growth and development of nanotechnology have probablyresulted in increased exposure to these nanostructures, leadingto inhalation, ingestion, injection and absorption of nanomaterials(Oberdörster et al., 2005). Due to the increase in investment andproduction of goods at the nanoscale, combined with the increasedoccupational exposure of workers and consumers, it is of great impor-tance to assess the safety of these nanomaterials (Hsiao and Huang,2011; Hu et al., 2009).

Many lipid nanoparticles, including SLN, have been shown to benon-cytotoxic (Liu et al., 2008); (Joshi and Müller, 2009; Yuan et al.,2010). However, this biocompatibility is difficult to evaluate due to

Table 1Characterization of SLNs.

SLN Lipid Surfactant Particlesize (nm)

PDI Zetapotential(mV)

F1 Solid white vaseline USP Tween 80 128±4.1 0.32±0.02 −18F2 Butyrospermum parkii Tween 80 144±3.2 0.21±0.05 −23F3 Virola surinamensis Tween 80 135±2.5 0.17±0.08 −30F4 Tristearin Tween 80 110±4.5 0.22±0.06 −26F5 Tristearin SDS 116±3.5 0.21±0.05 −39F7 Theobroma cacao Tween 80 167±5.2 0.18±0.03 −41F8 Platonia esculenta Tween 80 173±2.4 0.19±0.02 −20

383A.H. Silva et al. / Science of the Total Environment 432 (2012) 382–388

the variable composition of formulations that differs both in natureand percentage of lipids. Due to the poor available data about toxicityin the literature, it is necessary to evaluate the toxicity of SLN firstin vitro, then ex vivo, and finally in vivo for comparison with othernanostructures (Nassimi et al., 2010).

Toxicity generated by nanostructured materials is not limited tohumans but also extended to the environment. According to Ricoet al. (2011), environmental conditions can influence the concentra-tions of ions present in plants, which can also determine the absorp-tion and accumulation of nanoparticles. Plants can absorb essentialand non-essential elements by soil and water, which above certainconcentrations can cause toxicity. When stored in the plants, toxicelements can be transferred to the humans by the consumption ofvegetables or roots (Rico et al., 2011). This exposure of toxic elementscan also happen by fish consumption, according to recent reviewreporting the toxicity induced by nanomaterials in aquatic environ-ment (Kahru and Dubourguier, 2010).

In an attempt to cover the toxicity of nanoparticles, cell culturemethods have been used to predict the cytotoxicity of nanoparticlesand to explore the mechanisms involved in these phenomena. Viabilityassays such as 3-[4, 5‐dimethylthiazol-2-yl]-2, 5‐diphenyl tetrazoliumbromide (MTT) and Neutral Red (NR), give information about cytotox-icity and reflect the capacity of some compounds to cause cell death asa consequence of damage or alterations with basic cellular functions(Ivanova and Uhlig, 2008). Besides that, evaluation of mitochondrialmembrane potential is crucial, since any disturbance can dramaticallychange the maintenance of bio-energetic state of cells (Szabo et al.,2011) leading to power generation deficiency and cell death (Bernardi,1999; Teodoro et al., 2011). Additionally, oxidative stress has beenimplicated in the mechanism of nanoparticles toxicity (Ahamed, 2011;Shukla et al., 2011; Wang et al., 2009). When generation of reactiveoxygen species (ROS) exceeds the capacity of antioxidant defense, apotential impact on cell metabolism can occur, leading to cell damageand death afterwards (Ahamed, 2011).

Therefore, in this work we explored the effect of several SLN ontwo fibroblast cell lineages using different viability assays, mitochon-drial potential and cell cycle disturbance, as well as ROS generation.

2. Materials and methods

2.1. Materials

Several types of lipid were employed in nanoparticles preparation,such as tristearin (Dynasan 118®), solid white vaseline USP (DEG),and vegetal lipids from Butyrospermum Parkii, Theobroma cacao, Virolasurinamensis, Platonia esculenta (Naturais da Amazônia, Brazil). Theselipid materials were chosen due to their relevance in the potentialuse as drug nanocarries, and since these raw materials were alreadyapproved to be used by the cosmetic industry. The surfactants poly-sorbate 80 (Tween 80®) and sodium dodecylsulfate (SDS) werepurchased from Sigma-Aldrich. The cell culture media and fetalbovine serum were purchased from Cultilab. The antibiotics penicillin/streptomycin were purchased from GIBCO; JC-1 probe (5,5′,6′6-tetrachloro-,1′,3,3′ tetraethylbenzymidazolcarbocianyne iodide) andDCFH-DA (2′,7′‐dichlorofluorescein diacetate) were purchased fromInvitrogen. Dimethyl sulfoxide (DMSO) was from Merck and all otherreagents were purchased from Sigma-Aldrich.

2.2. SLN preparation

Tristearin SLN and solid white vaseline USP SLN were prepared bysolubilization to obtain 200 mg of lipid in a solution of chloroform/methanol (1:1). Organic solvents were removed and the lipid layerwas melted by heating to 5 °C above the lipid melting point.After that, an aqueous phase was prepared by dissolving Tween80® or SDS (1%) in ultra-pure water (MilliQ–Millipore®) to produce

20 mL of the preparation and heated to the molten lipid phase tem-perature. The hot aqueous phase was added to the molten lipidphase and ultrasonicated (Vibracells) for 2 min, at 20 W. The particleswere obtained by allowing the hot nanoemulsion to cool at roomtemperature.

The SLN obtained from vegetal lipids (F2, F3, F7, F8) were patentedand produced by Nanovetores® PI 0801545-7A2. The techniquedescribed above was employed in SLN preparation, except for theuse of organic solvents. The formulations were named as F1 to F8(Table 1). However, we tested only seven formulations because one ofthe eight formulations prepared (F1 to F8), named F6 was not stable.For this reason F6 was not studied.

2.3. SLN characterization

2.3.1. Particle size and surface chargeThe average particle size/distribution and surface charge (zeta

potential) were determined by dynamic light scattering and laser-Doppler anemometry, respectively, using a Zetasizer Nano ZS (MalvernInstruments), equipped with 173° scattering angle. The measurementsweremade at 25 °C after appropriate dilution of the samples in distilledwater. The particle size distribution was given by the polydispersityindex (PDI). To measure the zeta potential, samples were placed ina specific cuvette where a potential of ±150 mV was established. Thepotential values were calculated from the mean electrophoretic mo-bility values using Smoluchowski's equation.

2.3.2. Transmission electron microscopy (TEM)The particle shape and sizewere evaluated by transmission electron

microscopy (TEM). Samples were deposited on carbon-coated coppergrids with 200 meshes (CF200-Cu, EMS) and dried for 24 h at roomtemperature. The microscope (JEM-1011 TEM) was operated at an ac-celerating voltage of 100 kV. For morphological analysis the TEM wasoperated in bright field mode with magnification above 10,000×.

2.4. Cell culture

Monkey kidney fibroblasts (Vero) and dog kidney fibroblasts(MDCK) were obtained from American Type Culture Cell (ATCC). Thecells were cultured in DMEM supplemented with 10% heat-inactivatedfetal bovine serum, 100 U/mL penicillin, 100 μg/mL streptomycin and10mM HEPES. The cells were maintained at 37 °C in a 5% CO2 humidi-fied atmosphere and pH 7.4. Every 2–3 days, cells were passaged byremoving 90% of the supernatant and replacing it with fresh medium.In all experiments, viable cells were checked at the beginning of theexperiment by Trypan Blue exclusion.

2.5. Viability assay

The cytotoxicity of SLN was evaluated by MTT (Mosmann, 1983)and NR (Repetto et al., 2008). Vero cells (1×104/well) and MDCKcells (2×104/well) were exposed to SLN for 24 h with concentrationsranging from 100 to 1000 μg/mL. The effect of two antioxidants on

384 A.H. Silva et al. / Science of the Total Environment 432 (2012) 382–388

cell viability was also evaluated. Cells were treated with catalase(20.000 UI/mL) and pretreated with N-acetylcysteine (2 mM, 1 hprior the treatment of SLNs). After incubation, cells were washedwith fresh culture medium and 5 mg/mL of MTT was added, followedby incubation for 2 h at 37 °C. The precipitated formazan was dis-solved in 100 μL of DMSO and the absorbance was measured at540 nm using a micro-well system reader. For NR assay, the mediumcontaining the particles was replaced with 200 μL of culture mediumcontaining 0.05 mg/mL of NR and incubated for 30 min. After that,cells were washed twice with 200 μL of PBS. The extraction wasperformed with 200 μL of a solution containing 1% acetic acid, 50%ethanol and 49% water. The absorbance was measured at 540 nm.The IC50 values (a concentration that produces 50% reduction in theviable cell number) were calculated through a Hill concentration–response curve. A control was run in parallel in order to evaluatethe toxicity of surfactants used in SLN preparation.

2.6. Mitochondrial membrane potential

To assay the mitochondrial membrane potential, the lipophiliccationicfluorochrome5,5′,6′6-tetrachloro-,1′,3,3′tetraethylbenzymida-zolcarbocianyne iodide (JC-1) was used. JC-1 is a green fluorescentmonomer at depolarized membrane potential or a red fluorescentJ-aggregate at hyperpolarized membrane potential. Cells wereseeded at 5×105/mL in 12-well dishes and incubated with the for-mulations for 24 h. After that, JC-1 (10 μg/mL) was added and cellswere incubated for 20 min at 37 °C (5% CO2), then cells werewashed twice with PBS and resuspended in 500 μL of PBS. A vol-ume of 100 μL was used to measure the fluorescence by a spectro-fluorimeter (Perkin‐Elmer LS55). JC-1 was excited at 488 nm, thered emission fluorescence was detected at 590 nm and the greenfluorescence was detected at 527 nm. The mitochondrial potentialwas presented as a ratio of 590/527 fluorescence and comparedwith the control cells that were considered to have 100% mito-chondrial membrane potential. An electron transport chain uncou-pler (FCCP 1 μM) was used as a positive control.

2.7. Cell cycle analysis

To analyze the cell cycle of the cells treatedwith SLN, flow cytometrywas used following themethod described elsewhere (Yang et al., 2007).Briefly, cells (4×105/mL) were incubated with the SLNs for 24 h, in12-well plates. After incubation, cells were harvested with trypsinand after 10 min of centrifugation at 400 g at room temperature.After that, cells were washed with PBS and centrifuged again. Thesupernatant was discarded and 200 μL of 70% ethanol was addedfollowing by incubation for 30 min at 4 °C. After incubation, 1 mL ofPBS with 2% of BSA was added and centrifuged for 10 min at 400 g.The supernatants were withdrawn and 0.5 μL of RNase (100 μg/mL)in lysis buffer (0.1% Triton‐X in PBS) was added. DNA content wasanalyzed using the FACSCanto flow cytometry equipment (BectonDickinson). The cell population in each phase of the cell cycle was de-termined using WinMDI 2.9 software.

2.8. ROS generation

Intracellular reactive species formation was evaluated using20,70-dichlorofluorescein diacetate (DCFH-DA), which is oxidized todichlorofluorescein (DCF) in the presence of ROS (Sauer et al.,2003). Cells (1.5×104/mL) were incubated with the SLN for 24 h.Subsequently, cells were incubated with 10 μM of DCFH-DA for30 min at 37 °C and then washed four times with PBS. The DCFfluorescence signal was measured using a spectrofluorimeter (Perkin‐Elmer LS55). The results obtained as fluorescence units were expressedas percentage of ROS, compared with non-treated cells and normalizedto total protein content.

2.9. Statistical analysis

Results were shown as means±SD of triplicates from three-independent experiments. Statistical significance was assessed byone-way ANOVA, followed by Dunnet's test, and *pb0.05 was takenas statistical significance.

3. Results

3.1. Particles characterization

Dynamic light scattering, distribution (PDI), and surface charge(zeta potential) measurements were used to characterize the SLNs.As shown in Table 1, the average SLN sizes were in a range of 110and 180 nm for all formulations with a narrow size distribution, asevidenced by low PDI value, indicating monodisperse populations.Additionally, all formulations showed negative charged surface dueto their lipid composition, and the presence of Tween 80®, a non-ionic surfactant, did not alter the surface charge. However, whenSDS was used, an increase in the zeta potential values was observed(from −26 mV to −39 mV), due to the presence of this anionicsurfactant on the surface of F5. To confirm particle size and obtain in-formation about morphology, TEM analysis was performed (Fig. 1).Through the TEM images, it was possible to determine and correlatesize with dynamic light scattering and to observe that the majorityof the particles had roughly spherical shape and reasonable smoothsurface.

3.2. Cell toxicity

To determine the cytotoxicity of SLNs, formulations were tested inVero and MDCK cell lineages. Firstly, cells were exposed to 500 μg/mLof each SLN for 24 h in order to establish the threshold limit value fortoxicity. For those SLNs that demonstrated any cytotoxicity usingMTTand NR at 500 μg/mL concentration, IC50 was further establishedusing both methods. F2, F3, F7 and F8 did not demonstrate any toxic-ity (data not shown) and all further experiments were performedwith 500 μg/mL. However, F1, F4 and F5 were hazardous to thecells. Table 2 shows the IC50 of these SLNs for both Vero and MDCKcells. The IC50 of F1 and F4 were higher than that observed for F5,demonstrating less cytotoxicity. Despite of the fact that F1 and F4showed some toxicity to cells, all further experiments were con-ducted using 500 μg/mL, due to their high IC50 values found. HoweverF5 showed cytotoxicity for Vero cells using MTT assay decreasingstrongly cell viability (IC50=247 μg/mL), nonetheless this low valuewas not observed in NR assay.

Cytotoxicity of surfactants Tween 80® and SDSwas also conductedin both cell lineages. It was observed that Tween 80® did not showcytotoxicity in the concentration tested including that used in SLNspreparation. However, SDS demonstrated toxicity in all concentra-tions tested (Supplementary data).

3.3. Mitochondrial potential measurement

The membrane-permeant JC-1 dye is widely used in apoptosisstudies to monitor mitochondrial health. Fig. 2 shows the resultsrelated to the effect of SLNs on mitochondrial membrane potential.It was observed that only the nanoparticles F4 and F5 decreased themitochondrial membrane potential in Vero and MDCK cell lineages,respectively.

3.4. Cell cycle analysis

To investigate the effects of nanoparticles on cell cycle distribu-tion, Vero and MDCK cells were treated with 500 μg/mL of SLNs for24 h, except for F5 (247 μg/mL) in Vero cells. Results are summarized

Fig. 1. TEM images of SLN. The name of the formulations is indicated in the figure. The samples were prepared according to the method described in Material and methods.

385A.H. Silva et al. / Science of the Total Environment 432 (2012) 382–388

in Figs. 3 and 4. Treatment with F5 decreased the population of MDCKcells in G0/G1 and G2/M cell phases, which was accompanied by amarkedly increase in Sub/G1 population, and as expected, untreatedcells showed a typical distribution of cell cycle phases. These resultsindicate that F5 induced cell death of MDCK lineage due to DNA frag-mentation, suggesting apoptosis.

Fig. 2. Effect of nanoparticles on mitochondrial membrane potential in Vero (A) andMDCK (B) cell lineages. The membrane potential was determined using JC-1 stainingafter 24 h of incubation of SLN at 500 μg/mL, with the exception of F5 at Vero cells(247 μg/mL). The decrease of red/green ratio indicates a decrease in mitochondrialmembrane potential, and the results are expressed as percentage control cells (100%),*pb0.05.

3.5. ROS generation

Fig. 5 shows the results related to the cell generation of ROS in-duced by SLNs. As observed, all formulations induced oxidative stressin both cell lineages, however the effect of F5 seems to be more rele-vant. Additionally, the ability of Tween 80® to increase ROS genera-tion was evaluated. The results obtained indicate that increase ofROS generation induced by Tween 80® was pronounced, but theoxidative stress induced by SLNs was much greater (Fig. 5). In orderto evaluate the relationship between cell death and ROS increment,antioxidants were added to cells and viability assay (MTT) was per-formed with indicated concentrations of SLN. The results are shownin Fig. 6. Interestingly, the presence of catalase and N-acetylcysteineprotected cells against the oxidative stress caused by F5 SLN, observedby the increase in cell viability. However this protection was observedonly for Vero cells.

4. Discussion

In the current study, we reported that SLNs prepared with distinctsolid lipids had equal influence on fibroblast cells regarding cell viabilityand oxidative stress. But in general, SLNs were biocompatible due totheir high values of IC50, and those formulations that triggered a de-crease in cell viability, ROS generation was involved in the cytotoxicity.

Parameters as size, shape, charge and material composition of SLNare decisive for nanoparticle entrance into the cell, and thus, crucialfor the activity as a nanocarrier (Sahay et al., 2010). Moreover, parti-cles of less than 200 nm have been stated as prospective nanocarrierssince nanoparticles smaller than 30 nm are more susceptible tophagocytosis (Moghimi et al., 2005). Our results demonstrated thatthe mean diameter of SLN was 150 nm and seems to be useful to useas nanocarriers.

Table 2The IC50 of SLN using MTT and NR viability assays.

SLN IC50 (μg/m)(MTT)

IC50 (μg/m)(NR)

Vero MDCK Vero MDCK

F1 489±4.9 >1000 >1000 >1000F4 682±6.7 >1000 694±9.3 >1000F5 247±7.2 603±2.7 643±2.4 586±15.0

Fig. 3. Effects of nanoparticles on cell cycle distribution, onVero cell lineage. (A)Histogramsshow the number of the cell channel (vertical axis) vs. DNA content (horizontal axis).(B) Cells were treated with 500 μg/mL of the formulations for 24 h, with the exception ofF5 at Vero cell lineage, forwhich the IC50 (247 μg/mL) forMTT assaywas used, andanalyzedby DNA flow cytometry. G0/G1, G2/M and S indicate the cell phase, and sub-G1 DNAcontent refers to the proportion of apoptotic cells. Each phase was calculated by using theWinMDI 2.9 program.

Fig. 4. Effects of nanoparticles on cell cycle distribution, on MDCK cell lineage.(A) Histograms show the number of cell channel (vertical axis) vs. DNA content (horizontalaxis). (B) Cells were treatedwith 500 μg/mL of nanoparticles for 24 h,with the exception ofF5 for which the IC50 (247 μg/ml) for MTT assay was used, and analyzed by DNA flowcytometry. G0/G1, G2/M and S indicate the cell phase, and sub-G1 DNA content refers tothe proportion of apoptotic cells. Each phase was calculated by using the WinMDI 2.9program.*pb0.05 and ***pb0.001.

Fig. 5. Effect of nanoparticles on free radical generation in Vero (A) and MDCK (B) celllineages. Free radical formation was followed using DCFH-DA as described in Materialand methods. The cells were incubated with the nanoparticles at 500 μg/mL for 24 h,with the exception of F5 at Vero cell lineage, for which the IC50 (247 μg/mL) for MTTassay was used. The results were expressed as the percentage of fluorescent cells incomparison to control samples (zero % of fluorescence), normalized with the resultsby the protein concentration. *pb0.05, **pb0.01 and ***pb0.001.

386 A.H. Silva et al. / Science of the Total Environment 432 (2012) 382–388

The PDI and zeta potential are also important parameters that in-dicate the stability of nanoparticles (Martins et al., 2012). The resultspresented here showed a low polydispersion index, demonstratingmonodisperse population of SLNs evaluated. Also, values of zeta po-tential varied from −18 to −41 indicating a good stability (Wu etal., 2011).

To determine whether cell toxicity occurs, firstly we applied dis-tinct in vitro viability and cytotoxicity assays using Vero and MDCKcell lineages. Vero cells have been used extensively due to easy avail-ability and fast growth (Bouaziz et al., 2006; Vaucher et al., 2010).Also they have been suggested as suitable model to study nephrotox-icity (Dias et al., 2009) as well as the MDCK cell lineage (Rezzani et al.,2002). These models can be used to study the cytotoxicity in vitro topredict potential toxicity to kidneys.

More than one type of test should be used to determine in vitrocell viability studies, because it increases the reliability of the results.Accordingly, we used MTT salt assay and neutral red assay. NeutralRed is based on the ability of viable cells to incorporate and bind neu-tral red dye into the lysosomes, thereby predicting the cytotoxicitythrough lysosomal activity (Repetto et al., 2008). NR uptake by cellsoccurs through a process requiring energy, being sensitive to sub-stances that interfere with cell membrane and lysosomal permeabili-ty as well as the process of energy-dependent endocytosis (Repettoet al., 2008). Our results demonstrated that nanoparticles F4 andF5 were harmful to Vero cell lineage and F5 to MDCK cell lineage,suggesting that these nanoparticles interfere in mitochondrial metab-olism of cells. We also demonstrated that Vero cells are more sensitiveto SLN than MDCK lineage. Moreover, F4 and F5 showed cytotoxicityfor both lineages assessed by NR assay, probably by interference incellular lysosomal activity. Comparing these two methods of cell via-bility, F4 and F5 seem to promote changes in mitochondrial activity

Fig. 6. Protective effect of catalase and NAC against cytotoxicity of F1, F2, F4 and F5nanoparticles in Vero (A) and F5 in MDCK (B) cells, by MTT assay. Cells were pretreatedwith NAC (1 mM) for 1 h followed by SLN treatment at 500 μg/mL for 24 h (the catalasetreatments – 20.000 UI/mL – were done concomitantly with the SLN). Optical densityof untreated cells was taken as 100% of cell viability. *pb0.05, and ***pb0.001.

387A.H. Silva et al. / Science of the Total Environment 432 (2012) 382–388

and interference in lysosomal activity. It is well-known that the re-lease of lysosomal contents into the cytosol can initiate the cascadeof apoptosis, also leading to the release of pro-apoptotic factors bymitochondria (Zhao et al., 2003).

Mitochondria have a crucial function in controlling the cell deathprocess, besides playing a crucial role in maintaining the bioenergeticstatus of cells (Szabo et al., 2011). The impairment of mitochondrialfunction by nanoparticles can have drastic consequences, upsettingthe balance and cellular functions (Teodoro et al., 2011). In thisregard, our results showed that only ROS generation by F4 and F5seemed to involve mitochondria in the process of cell death; thesenanoparticles decrease the mitochondrial membrane potential inVero and in MDCK cell lineages. The differences in cytotoxicity in-duced by the nanoparticles between VERO and MDCK cell lineagesare probably due to the differences in the characteristic of the celllineages to respond to, for example, an oxidative stress. For thisreason we tested the formulations in two cell lineages and assayedthe cell viability monitoring different cell functions.

By flow cytometry, we confirmed the cytotoxicity found with cellviability assays and determined the type of cell death. Cell exposedto F5 decreased the population of MDCK cells in G0/G1 and G2/Mcell phases, which was accompanied by a marked increase of a Sub/G1 population at 24 h. G0/G1, G2/M and S indicate the cell phase,and Sub-G1 DNA content refers to the proportion of apoptotic cells(Ormerod, 2002; Yang et al., 2007; Yang et al., 2007).

To start understanding the mechanisms underlying the toxicityof SLN, we investigated whether oxidative stress was involved innanoparticle-induced cell death. The generation of ROS was evaluatedby DCFH assay, and except F1 in Vero lineage, all the nanoparticles in-duced oxidative stress, which could explain the cytotoxicity of somenanoparticles to cells. As one of the toxic mechanisms of nanoparticles,

the generation of ROS seems to be most widely recurrent and conse-quently widely studied (Park et al., 2008).

Recent reports showed that ROS production has been found incells exposed to different kinds and concentrations of nanoparticles,such as TiO2 (10 μg/mL) (Park et al., 2008; Shukla et al., 2011), carbonblack (30 μg/mL) (Foucaud et al., 2010), ZnO (5 μg/cm2) (De Berardiset al., 2010), and SiO2 (25 μg/mL) (Wang et al., 2009), among others.Oxidative stress has been proposed as a common mechanism of celldamage induced by many types of nanoparticles (Park et al., 2008).However, it is important to highlight that there is a complex systemof enzymatic and non-enzymatic antioxidants that protect cellsagainst harmful pro-oxidants (Reuter et al., 2010). Also, the increasein ROS production, onlywithout a proportional increase in the produc-tion of antioxidants, can induce mitochondrial membrane permeabil-ity, damage to the respiratory chain which may trigger the apoptosisprocess (Valko et al., 2006). To test this hypothesis, antioxidants(catalase or N-acetylcysteine) were incubated concomitantly with thenanoparticles F2, F4 and F5 and cell viability by MTT was performed.The results demonstrated that catalase protected cells increasingviability. However, Vero cells treated with F5 were protected inthe presence of both antioxidants (catalase and N-acetylcysteine),confirming the relationship between ROS and cell death in this case.

The reason of the higher cytotoxicity of the formulation F5 will befurther studied, however we have some evidence that it is caused bythe presence of SDS (sodium dodecyl sulfate). Recently, Caon et al.(2010) reported that nanoparticles produced with SDS were morecytotoxic when compared with nanoparticles produced with othersurfactants. The SDS has shown an irritant potential in compatibilitytests carried out on cell culture particularly due to its solubilizingability of lipid membranes (Spiekstra et al., 2009).

The SLNs studied here were shown to be biocompatible since onlysome of them showed a mild cytotoxicity. Even the formulation F5that was shown to bemore toxic among SLNs studied, seems to inducecell deathwith very high concentrations, if compared to the cytotoxic-ity induced by metallic nanoparticles such as TiO2, which with 10 μgwere shown to be cytotoxic to mitochondria (Freyre-Fonseca et al.,2011). Additionally, even for the formulation F4 that was also shownto be cytotoxic (although less than the F5), the IC50 for cell viabilitywas very high, ~700 μg/mL for both cell lineages. Furthermore, thereare some aspects that determine the content of nanoparticles andtheir biocompatibility. It would depend on the nature and concentra-tion of the molecules incorporated into the particles, the capability ofthe nanostructured systems to release the drugs, the site and mecha-nism of action of the drugs, as well as the sickness being treated. Theamount of nanoparticles would be particular to each case. Particleswith the combined materials used in the formulation F5 should beavoided.

However, it should be remarked that cytotoxic effects of particu-late carrier systems differ, depending on the cell lines used, due tometabolic abilities (e.g. presence of specific enzymes) and defensecapabilities of these cell lines. Further studies are required to eluci-date the in vivo toxicity of these nanoscale structures, in order todetermine the connotation to the health and to the environment oftheir increased and widespread use.

Acknowledgments

This study was supported by grants from CNPq (ConselhoNacional de Desenvolvimento Científico e Tecnológico), CAPES(Coordenação de Aperfeiçoamento de Pessoal de Nível Superior)and FAPESC (Fundação de Amparo à Pesquisa de Santa Catarina) aswell as for REUNI/MEC Program for the Pharmacy doctoral degree fel-lowship to Adny Henrique Silva. The group wishes to thank theLAMEB — Laboratório Multi Usuário de Estudos em Biologia-CCB/UFSC and LCME — Laboratório Central de Microscopia Eletrônica—UFSC, for the use of its facilities.

388 A.H. Silva et al. / Science of the Total Environment 432 (2012) 382–388

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.scitotenv.2012.06.018.

References

Ahamed M. Toxic response of nickel nanoparticles in human lung epithelial A549 cells.Toxicol In Vitro 2011;25:930–6.

Bernardi P. Mitochondrial transport of cations: channels, exchangers, and permeabilitytransition. Physiol Rev 1999;79:1127–55.

Bouaziz C, Abid-Essefi S, Bouslimi A, El Golli E, Bacha H. Cytotoxicity and related effectsof T-2 toxin on cultured Vero cells. Toxicon 2006;48:343–52.

Caon T, Zanetti-Ramos BG, Lemos-Senna E, Cloutet E, Cramail H, Borsali R, et al. Evalua-tion of DNA damage and cytotoxicity of polyurethane-based nano- and microparti-cles as promising biomaterials for drug delivery systems. J Nanopart Res 2010;12:10.

Chattopadhyay P, Shekunov BY, Yim D, Cipolla D, Boyd B, Farr S. Production of solidlipid nanoparticle suspensions using supercritical fluid extraction of emulsions(SFEE) for pulmonary delivery using the AERx system. Adv Drug Deliv Rev 2007;59:444–53.

De Berardis B, Civitelli G, Condello M, Lista P, Pozzi R, Arancia G, Meschini S. Exposureto ZnO nanoparticles induces oxidative stress and cytotoxicity in human colon car-cinoma cells. Toxicol Appl Pharmacol 2010;242:116–27.

Dias E, Andrade M, Alverca E, Pereira P, Batoréu MC, Jordan P, et al. Comparative studyof the cytotoxic effect of microcistin-LR and purified extracts from Microcystisaeruginosa on a kidney cell line. Toxicon 2009;53:487–95.

Elzoghby AO, Samy WM, Elgindy NA. Albumin-based nanoparticles as potentialcontrolled release drug delivery systems. J Control Release 2012;157:168–82.

Fadeel B, Garcia-Bennett AE. Better safe than sorry: understanding the toxicologicalproperties of inorganic nanoparticles manufactured for biomedical applications.Adv Drug Deliv Rev 2010;62:362–74.

Foucaud L, Goulaouic S, Bennasroune A, Laval-Gilly P, Brown D, Stone V, et al. Oxidativestress induction by nanoparticles in THP-1 cells with 4-HNE production: stress bio-marker or oxidative stress signalling molecule? Toxicol In Vitro 2010;24:1512–20.

Freyre-Fonseca V, Delgado-Buenrostro NL, Gutiérrez-Cirlos EB, Calderón-Torres CM,Cabellos-Avelar T, Sánchez-Pérez Y, et al. Titanium dioxide nanoparticles impairlung mitochondrial function. Toxicol Lett 2011;202:111–9.

Hsiao IL, Huang YJ. Effects of various physicochemical characteristics on the toxicitiesof ZnO and TiO nanoparticles toward human lung epithelial cells. Sci Total Environ2011;409:1219–28.

Hu X, Cook S, Wang P, Hwang HM. In vitro evaluation of cytotoxicity of engineeredmetal oxide nanoparticles. Sci Total Environ 2009;407:3070–2.

Ivanova L, Uhlig S. A bioassay for the simultaneous measurement of metabolic activity,membrane integrity, and lysosomal activity in cell cultures. Anal Biochem 2008;379:16–9.

Joshi MD, Müller RH. Lipid nanoparticles for parenteral delivery of actives. Eur J PharmBiopharm 2009;71:161–72.

Kahru A, Dubourguier H-C. From ecotoxicology to nanoecotoxicology. Toxicology2010;269:105–19.

Kayser O, Lemke A, Hernández-Trejo N. The impact of nanobiotechnology on the de-velopment of new drug delivery systems. Curr Pharm Biotechnol 2005;6:3–5.

Kwon SM, Nam HY, Nam T, Park K, Lee S, Kim K, et al. In vivo time-dependent gene ex-pression of cationic lipid-based emulsion as a stable and biocompatible non-viralgene carrier. J Control Release 2008;128:89–97.

Liu J, Gong T, Fu H, Wang C, Wang X, Chen Q, et al. Solid lipid nanoparticles for pulmo-nary delivery of insulin. Int J Pharm 2008;356:333–44.

Martins S, Costa-Lima S, Carneiro T, Cordeiro-da-Silva A, Souto EB, Ferreira DC. Solidlipid nanoparticles as intracellular drug transporters: an investigation of theuptake mechanism and pathway. Int J Pharm 2012;430:216–27.

Moghimi SM, Hunter AC, Murray JC. Nanomedicine: current status and future prospects.FASEB J 2005;19:311–30.

Mosmann T. Rapid colorimetric assay for cellular growth and survival: application toproliferation and cytotoxicity assays. J Immunol Methods 1983;65:55–63.

Müller RH, Runge S, Ravelli V, Mehnert W, Thünemann AF, Souto EB. Oral bioavailabilityof cyclosporine: solid lipid nanoparticles (SLN) versus drug nanocrystals. Int J Pharm2006;317:82–9.

Nassimi M, Schleh C, Lauenstein HD, Hussein R, Hoymann HG, Koch W, et al. A toxico-logical evaluation of inhaled solid lipid nanoparticles used as a potential drugdelivery system for the lung. Eur J Pharm Biopharm 2010;75:107–16.

Nel A, Xia T, Mädler L, Li N. Toxic potential of materials at the nanolevel. Science2006;311:622–7.

Oberdörster G, Oberdörster E, Oberdörster J. Nanotoxicology: an emerging disciplineevolving from studies of ultrafine particles. Environ Health Perspect 2005;113:823–39.

Ormerod MG. Investigating the relationship between the cell cycle and apoptosis usingflow cytometry. J Immunol Methods 2002;265:73–80.

Pardeike J, Hommoss A, Müller RH. Lipid nanoparticles (SLN, NLC) in cosmetic andpharmaceutical dermal products. Int J Pharm 2009;366:170–84.

Park EJ, Yi J, Chung KH, Ryu DY, Choi J, Park K. Oxidative stress and apoptosis inducedby titanium dioxide nanoparticles in cultured BEAS-2B cells. Toxicol Lett 2008;180:222–9.

Repetto G, del Peso A, Zurita JL. Neutral red uptake assay for the estimation of cellviability/cytotoxicity. Nat Protoc 2008;3:1125–31.

Reuter S, Gupta SC, Chaturvedi MM, Aggarwal BB. Oxidative stress, inflammation, andcancer: how are they linked? Free Radic Biol Med 2010;49:1603–16.

Rezzani R, Angoscini P, Borsani E, Rodella L, Bianchi R. Cyclosporine A-induced toxicityin two renal cell culture models (LLC-PK1 andMDCK). Histochem J 2002;34:27–33.

Rico CM, Majumdar S, Duarte-Gardea M, Peralta-Videa JR, Gardea-Torresdey JL. Inter-action of nanoparticles with edible plants and their possible implications in thefood chain. J Agric Food Chem 2011;59:3485–98.

Sahay G, Alakhova DY, Kabanov AV. Endocytosis of nanomedicines. J Control Release2010;145:182–95.

Sauer H, Wefer K, Vetrugno V, Pocchiari M, Gissel C, Sachinidis A, et al. Regulation ofintrinsic prion protein by growth factors and TNF-alpha: the role of intracellularreactive oxygen species. Free Radic Biol Med 2003;35:586–94.

Shukla RK, Sharma V, Pandey AK, Singh S, Sultana S, Dhawan A. ROS-mediatedgenotoxicity induced by titanium dioxide nanoparticles in human epidermal cells.Toxicol In Vitro 2011;25:231–41.

Spiekstra SW, Dos Santos GG, Scheper RJ, Gibbs S. Potential method to determine irri-tant potency in vitro— comparison of two reconstructed epidermal culture modelswith different barrier competency. Toxicol In Vitro 2009;23:349–55.

Szabo A, Balog M, Mark L, Montsko G, Turi Z, Gallyas F, et al. Induction of mitochondrialdestabilization and necrotic cell death by apolar mitochondria-directed SODmimetics. Mitochondrion 2011;11:476–87.

Teodoro JS, Simões AM, Duarte FV, Rolo AP, Murdoch RC, Hussain SM, et al. Assessmentof the toxicity of silver nanoparticles in vitro: a mitochondrial perspective. ToxicolIn Vitro 2011;25:664–70.

Valko M, Rhodes CJ, Moncol J, Izakovic M, Mazur M. Free radicals, metals and antioxi-dants in oxidative stress-induced cancer. Chem Biol Interact 2006;160:1-40.

Vaucher RA, Teixeira ML, Brandelli A. Investigation of the cytotoxicity of antimicrobialpeptide P40 on eukaryotic cells. Curr Microbiol 2010;60:1–5.

Videira M, Almeida AJ, Fabra À. Preclinical evaluation of a pulmonary deliveredpaclitaxel-loaded lipid nanocarrier antitumor effect. Nanomed Nanotechnol BiolMed in press. http://dx.doi.org/10.1016/j.nano.2012.12.007.

Wang M, Thanou M. Targeting nanoparticles to cancer. Pharmacol Res 2010;62:90–9.Wang F, Gao F, Lan M, Yuan H, Huang Y, Liu J. Oxidative stress contributes to silica

nanoparticle-induced cytotoxicity in human embryonic kidney cells. Toxicol In Vitro2009;23:808–15.

Wu L, Zhang J, Watanabe W. Physical and chemical stability of drug nanoparticles. AdvDrug Deliv Rev 2011;63:456–69.

Yang L, Wu S, Zhang Q, Liu F, Wu P. 23,24-Dihydrocucurbitacin B induces G2/Mcell-cycle arrest and mitochondria-dependent apoptosis in human breast cancercells (Bcap37). Cancer Lett 2007;256:267–78.

Yuan H, Zhang W, Du YZ, Hu FQ. Ternary nanoparticles of anionic lipidnanoparticles/protamine/DNA for gene delivery. Int J Pharm 2010;392:224–31.

Zhao M, Antunes F, Eaton JW, Brunk UT. Lysosomal enzymes promote mitochondrialoxidant production, cytochrome c release and apoptosis. Eur J Biochem 2003;270:3778–86.