cytotoxicity of peptide-coated silver nanoparticles on the human intestinal cell line caco-2
TRANSCRIPT
Arch Toxicol (2012) 86:1107–1115
DOI 10.1007/s00204-012-0840-4INORGANIC COMPOUNDS
Cytotoxicity of peptide-coated silver nanoparticles on the human intestinal cell line Caco-2
Linda Böhmert · Birgit Niemann · Andreas F. Thünemann · Alfonso Lampen
Received: 15 November 2011 / Accepted: 1 March 2012 / Published online: 15 March 2012© Springer-Verlag 2012
Abstract Silver nanoparticles are used in a wide range ofconsumer products such as clothing, cosmetics, householdgoods, articles of daily use and pesticides. Moreover, theuse of a nanoscaled silver hydrosol has been requested inthe European Union for even nutritional purposes. How-ever, despite the wide applications of silver nanoparticles,there is a lack of information concerning their impact onhuman health. In order to investigate the eVects of silvernanoparticles on human intestinal cells, we used the Caco-2cell line and peptide-coated silver nanoparticles with deW-ned colloidal, structural and interfacial properties. The par-ticles display core diameter of 20 and 40 nm and werecoated with the small peptide L-cysteine L-lysine L-lysine.Cell viability and proliferation were measured using Prom-egas CellTiter-Blue® Cell Viability assay, DAPI stainingand impedance measurements. Apoptosis was determinedby Annexin-V/7AAD staining and FACS analysis, mem-brane damage with Promegas LDH assay and reactive oxy-gen species by dichloroXuorescein assay. Exposure ofproliferating Caco-2 cells to silver nanoparticle induceddecreasing adherence capacity and cytotoxicity, wherebythe formation of reactive oxygen species could be the modeof action. The eVects were dependent on particle size (20,
40 nm), doses (5–100 �g/mL) and time of incubation (4–48 h).Apoptosis or membrane damage was not detected.
Keywords Oral uptake · Intestinal cells · Peptide-coated silver nanoparticles · Cytotoxicity
Introduction
In March 2011, an online nanotechnology consumer prod-ucts inventory contained 1,317 products or product lines,which were identiWed by manufactures themselves as con-taining nanotechnology. The most common material men-tioned in the product descriptions was silver with 313products (Project on emerging nanotechnologies). Silvernanoparticles are increasingly used in a wide range of com-mercial products, such as electronics, paints, clothing, cos-metics, household goods, articles of daily use, pesticidesand food packaging owing to the wide spectrum of physi-cal, chemical and antimicrobial activity.
By the reason of inhibiting eVects on microorganismgrowth, silver nanoparticles have been introduced into vari-ous food-contacting materials, such as plastic food contain-ers, refrigerators surfaces, storage bags and choppingboards (Chaudhry et al. 2008). Moreover, the use of a nano-scaled silver hydrosol for nutritional purposes has beenrequested in the European Union (EFSA 2008).
However, despite the wide applications of silver nano-particles, there is a lack of information concerning theimpact of manufactured silver nanomaterials on humanhealth. Among the dermal and the pulmonary exposure, theexposure of digestive organs via oral intake is of toxicologicalinterest. Recent studies demonstrated that silver nanoparti-cles were distributed to all organs after oral ingestion (Kimet al. 2008, 2009a, b). However, there is little information
This article is published as a part of the Special Issue “Nanotoxicology II” on the ECETOC Satellite workshop, Dresden 2010 (Innovation through Nanotechnology and Nanomaterials + Current Aspects of Safety Assessment and Regulation).
L. Böhmert (&) · B. Niemann · A. LampenDepartment Food Safety, Federal Institute for Risk Assessment, Max-Dohrn-Str. 8-10, 10589 Berlin, Germanye-mail: [email protected]
A. F. ThünemannFederal Institute for Material Research and Testing, Richard-Willstaetter-Str. 11, 12489 Berlin, Germany
123
1108 Arch Toxicol (2012) 86:1107–1115
available regarding the eVects of silver nanoparticles oncells representing intestinal functions.
In order to investigate the eVects of silver nanoparticles onhuman intestinal cells, we used peptide-coated silver nano-particles with deWned colloidal, structural and interfacialproperties (Graf et al. 2009) and proliferating and diVerenti-ated Caco-2 cells. The nanoparticles consist of a 20 and40 nm silver core coated with the small peptide L-cysteineL-lysine L-lysine. The human Caco-2 cell line is derived froma colon adenocarcinoma, whereby diVerentiated Caco-2 cellsshow more similarity with intestinal cells reXecting manyfunctions of the intestine (Lampen et al. 1998).
Methods
Nanoparticles
Peptide L-cysteine L-lysine L-lysine (CKK)-coated silvernanoparticles (20, 40 nm) were obtained from Dr. Alexan-dre Mantion from the Federal Institute for MaterialResearch and Testing in Germany. The particle synthesisand characterization were described in Graf et al. (2009).Stock solution was generated by suspending the dried parti-cles in 0.015 N HCl at a concentration of 1 mg/mL fol-lowed by the soniWcation for a few seconds to ensure anoptimal dispersion.
Cell lines and culture conditions
The human colon adenocarcinoma cell line Caco-2, thehuman liver hepatocellular carcinoma cell line HepG2 and thechinese hamster lung Wbroblast V79 cells were obtained fromthe European Collection of Cell Cultures (ECACC, PortonDown, UK). Caco-2 cells were maintained in Dulbecco’smodiWed Eagle’s medium (DMEM, PAN Biotech, Aiden-bach, Germany), HepG2 cells in Roswell Park MemorialInstitute medium (RPMI, PAN Biotech, Aidenbach, Germany)and V79 cells in Minimum Essential Medium eagle (MEM,PAN Biotech, Aidenbach, Germany) at 37 °C in a humidiWedatmosphere of 5 % CO2. All media were supplemented with1 % (v/v) penicillin/streptomycin and 10 % (v/v) FCS (foetalcalf serum) during cultivation. Cells were maintained in tissueculture Xasks as a subconXuent monolayer for propagation(75 cm²). For the experiments, 1 % (v/v) ITS (insulin, transfer-rin, selenium) was added instead of FCS.
Cell viability
Real-time cell analyser
The Caco-2 cells were plated into 96-well E-plates at a den-sity of 5 £ 103 cells per well in 100 �L culture medium.
Proliferating Caco-2 cells were allowed to attach for 24 hbefore treatment. Subsequently, culture medium wasreplaced by 100 �L particle suspension at concentrations of4, 10, 25, 50 and 100 �g/mL, medium control and 5 �g/mLpeptide control.
CellTiter-Blue® cell viability assay (CTB)
The Caco-2 cells were plated into 96-well plates at a den-sity of 1 £ 104 cells per well in 100 �L culture medium.Proliferating Caco-2 cells were allowed to attach for 24 hbefore treatment. DiVerentiated Caco-2 cells were grown in96-well pates for 20 days, and medium was changed every2 days. Subsequently, culture medium was replaced by100 �L particle suspension at concentrations of 4, 10, 25,50 and 100 �g/mL, medium control and 5 �g/mL peptidecontrol (this corresponds to the amount of peptide bound to100 �g/mL silver nanoparticle), and the cells were exposedfor 4, 6, 8, 24, 30 and 48 h. Afterwards, 20 �L CTB(Promega) was added to each well, incubated for 2 h andmeasured at a microplate reader (Berthold Mithras LB940)with 540 nm excitation and 590 nm emission.
DAPI assay (Klenow et al. 2009)
Cells were Wxed and lysed with methanol, and the DNAwas stained with 100 �L 20 �M 4�,6-diamidino-2-phe-nylindole (DAPI, Sigma-Aldrich) for at least 30 min. Fluo-rescence was measured using a microplate reader (BertholdMithras LB940) with 380 nm excitation and 460 nm emis-sion.
Apoptosis (Annexin-V/7AAD staining, FACS)
Apoptosis was quantiWed by Annexin-V/7AAD (ImmunoTools GmbH, BD Biosciences) staining and followingFACS analysis. Cells were plated into 12-well plates at adensity of 1 £ 105 Caco-2 cells or 1.5 £ 105 HepG2 cellsper well in 1 mL culture medium and allowed to attach for24 h before treatment. Culture medium was replaced by1 mL 5 �g/mL particle suspension, medium control andpositive control 4 �M staurosporine (Sigma-Aldrich), andcells were exposed for 2, 4, 6, 8, 16, 18, 20, 22 and 24 h.Caco-2 cells were additionally treated with 2.5 �g/mLcycloheximid (Sigma-Aldrich). Treated cells were har-vested and washed with annexin buVer and stained for20 min with Annexin-V/7AAD before FACS analysis wasperformed using BD FACS Canto II.
Membrane leakage
The membrane leakage was measured using PromegasCytoTox-ONE™ Homogeneous Membrane Integrity Assay.
123
Arch Toxicol (2012) 86:1107–1115 1109
The Caco-2 cells were plated into 96-well plates at adensity of 1 £ 104 cells per well in 100 mL culture mediumand allowed to attach for 24 h before treatment. Afterwards,culture medium was replaced by 100 �L particle suspen-sion at concentrations of 4, 10, 25, 50 and 100 �g/mL andcells were exposed for 2, 4, 6, 8 and 24 h. The positive con-trol was added a few minutes before measurement. Theplates were treated with the assay reagents following themanufacturers’ instructions and measured using a microplatereader (Berthold Mithras LB940) with 540 nm excitationand 590 nm emission.
Reactive oxygen species detections
The Caco-2 cells were plated into 96-well plates at a den-sity of 1 £ 104 cells per well in 100 mL culture mediumand allowed to attach for 24 h before treatment. Afterwards,culture medium was replaced by 100 �L 100 �M DCF incell culture medium with 1 % ITS and incubated for 1 h.Then, the cells were washed with 200 �L PBS and exposedto a medium control or a particle suspension at concentra-tions of 1, 3, 5, 10, 25, 50 and 100 �g/mL for 24 h. Theplates were measured using a microplate reader (BertholdMithras LB940) with 485 nm excitation and 535 nm emis-sion.
Micronucleus test
Chromosome breakages were analysed using the micronu-cleus test with V79 cells as this test was standardized, vali-dated and implemented in the framework of the regulatorytoxicology (OECD-Guideline 487: In Vitro MicronucleusTest). Cells were plated on microscope slides at a density of6–8 £ 104 cells in culture medium and allowed to attach for24 h before treatment. Afterwards, culture medium was
replaced by particle suspension at concentrations of 5, 10,25 and 50 �g/mL and medium control and the cells wereexposed for 24 h. Microscope slides were washed in 1.5 %citrate solution for 5 min and 2 times in Wx solution (25 mLacetic acid, 75 mL ethanol, 1.25 mL 37 % formaldehyde)for 2 min. Subsequently, slides were dried, treated withMay Grünwald solution for 3 min, washed with WeisebuVer and treated with Giemsa solution (5.2 mL Giemsa in200 mL Weise buVer) for 20 min. Afterwards, slides wereagain washed and dried.
Results
Microscopic observation
Exposure of silver nanoparticles to Caco-2 cells leads to adecreasing adherence capacity of the cells as observed inoptical microscope image (Fig. 1). The cells detached andtake a spherical form.
Cell viability
Cell analyser
The xCELLigence system from Roche monitors cell cultureexperiments in real time by measuring electrical impedanceacross microelectrodes integrated on the bottom of cultureplates. The impedance represents the interaction of the cellswith the electrodes and is dependent from the cell number,cell viability, cell morphology and degree of adhesion ofthe cells. Impedance is displayed as dimensionless cellindex.
The silver nanoparticles alone without cells did not inXu-ence the impedance. As shown in Fig. 2, the peptide L-cysteine
Fig. 1 Silver nanoparticle exposure caused spherical cell morphology. Microscopically images (200-fold magniWcation) of proliferating Caco-2cells after 24 h of exposure to medium control (a) and 5 �g/mL 20 nm peptide-coated silver nanoparticles (b)
123
1110 Arch Toxicol (2012) 86:1107–1115
L-lysine L-lysine, which is coated to the nanoparticles,exhibited a slight inXuence on the Caco-2 cells. The cellindex rises more slowly under the inXuence of the peptidethan the medium control does. Both nanoparticle sizesdemonstrated strong concentration-dependent eVects on theimpedance. At Wrst, the cell index increased directly afterincubation with the nanoparticles, but then decreases con-tinuously with the greatest eVect after 6 h of incubation.
CellTiter-Blue® cell viability and DAPI assay
The CellTiter-Blue® Cell Viability Assay provides a Xuo-rescent method for monitoring cell viability. The assay isbased on the ability of living cells to convert blue resazu-rine into pink resoruWne. Nonviable cells rapidly lose meta-bolic capacity and thus do not generate a Xuorescent signal.After the viability measurement, the same 96-well plateswere used for DAPI staining. DAPI (4�,6-diamidino-2-phe-nylindole) is a Xuorescent stain that binds to double-stranded DNA. The amount of DNA is related to thenumber of cells. The combination of CTB and DAPI stain-ing in the same 96-well plate allows a good discriminationbetween the metabolic capacity and the restriction of thecell growth.
The exposure of proliferating and diVerentiated Caco-2cells to the peptide L-cysteine L-lysine L-lysine alone didnot inXuence cell viability and growth (Figs. 3, 4). How-ever, the silver nanoparticles and silver nitrate demonstratea time-, concentration- and particle size-dependent eVect oncell viability and cell number at both cell stages. After24–48 h of incubation with both particle sizes at diVerentconcentrations, the cell number and viability decreased sig-niWcantly. With increasing concentration, the smaller nano-particles (20 nm) exhibit more toxic eVects than the 40 nmparticles. However, soluble silver ions showed the strongestimpact on the cells. Moreover, diVerentiated Caco-2 cellsare less sensitive to the nanoparticles and silver ions thanthe proliferating cells.
Apoptosis (Annexin-V/7AAD staining, FACS)
Annexin-V/7AAD staining is based on the observation thatat the beginning of apoptosis, cells translocate the mem-brane phosphatidylserine from the inner face of the plasmamembrane to the cell surface. Once on the cell surface,phosphatidylserine can be stained with a Xuorescent conju-gate of Annexin-V, a protein with a high aYnity for phos-phatidylserine. To diVerentiate apoptosis and necrosis, cells
Fig. 2 Impedance measurement with the real-time cell analyser. The cell index is a measure for the cell number, cell viability, cell morphology and adhesion degree of the cells. Proliferating Caco-2 cells were incubated with medium, 5 �g/mL of the peptide L-cysteine L-lysine L-lysine (CKK), that is coating the nanoparticles and 5–100 �g/mL 20 and 40 nm silver nanoparticles for up to 46 h. Impedance was measured every 15 min. Means were cal-culated from three wells, and the time of incubation was set to 0 h
123
Arch Toxicol (2012) 86:1107–1115 1111
were stained with 7AAD. 7AAD does not pass throughintact cell membranes and stains therefore only death cells.Cells were analysed by Xow cytometry (FACS). Addition-ally, Caco-2 cells were treated with 2.5 �g/mL cyclohexi-mide in order to prevent the expression of anti-apoptoticproteins.
As shown in Fig. 5, the positive control staurosporineshowed an increase in apoptosis up to 24 % around 20 h ofexposure for proliferating Caco-2 cells under cyclohexi-mide inXuence, whereas the silver nanoparticle showed noapoptosis. Since the Caco-2 cells are quite resistant toapoptosis inducing signals, we veriWed these results in themore sensitive cell line HepG2, which also do not showapoptosis.
Membrane leakage
The CytoTox-ONE™ Homogeneous Membrane IntegrityAssay (Promega) is a Xuorometric method for estimatingthe cytotoxicity on the basis of membrane integrity damagein multiwell plates. It measures the release of lactate dehy-drogenase (LDH) from cells with a damaged membrane.LDH release from cells in the culture medium is measuredwith a coupled enzymatic assay that results in the conver-
sion of resazurin into a Xuorescent resoruWn product. TheXuorescence signal is proportional to the number of lysedcells.
Figure 6 shows the results of the LDH assay. Up to 8 hof exposure, membrane leakage in proliferating Caco-2cells could not be detected. A concentration-dependentincrease in LDH leakage could be observed after 24 h,probably due to the cytotoxic eVect.
Reactive oxygen species detections
The dichlorodihydroXuorescein assay detects reactive oxy-gen species and is representative for oxidative stress incells. 2�,7�-dichlorodihydroXuorescein diacetate (DCF)diVuses passively into cells and can be hydrolysed by cellu-lar esterases to 2�,7�-dichlorodihydroXuorescein. Subse-quently, it can be oxidized to 2�,7�-dichloroXuorescein byreactive oxygen species, which can be measured as Xuores-cence signal.
As shown in Fig. 7, a concentration-dependent eVect ofthe silver nanoparticles on the oxidation of 2�,7�-dichlo-rodihydroXuorescein to 2�,7�-dichloroXuorescein could beobserved. Similar eVects were caused by the positive con-trol.
Fig. 3 Results of the CellTiter-Blue® Cell Viability Assay and DAPIstaining for proliferating Caco-2 cells. The cells were exposed for 24,30 and 48 h to 20 and 40 nm peptide-coated silver nanoparticles(Ag-CKK) at concentrations of 5–100 �g/mL and 5 �g/mL of the peptide
L-cysteine L-lysine L-lysine (CKK) that is coating the nanoparticles.The medium control was set to 100 %. Means and standard deviationswere calculated from at least three independent experiments. DiVerenceswere calculated by f test and t test using Microsoft Excel® (*p < 0.05)
123
1112 Arch Toxicol (2012) 86:1107–1115
Micronucleus test
The micronucleus test quantiWes chromosomal defects likechromosome aberration and damage at the spindle Wbres.These damages lead to the forming of micronuclei duringthe anaphase of mitosis or meiosis. The test was performedup to a maximal concentration of 25 �g/mL for 20 nmAgCKK and up to 50 �g/mL for 40 nm AgCKK, sincehigher concentrations interfered the microscopical countingby forming agglomerates. The cells showed no formation ofmicronuclei after treatment with the nanoparticles.
Discussion
So far, most investigations about the toxicity of silverfocused on silver ions, because the bulk silver is estimatedto be a nonhazardous material. Silver ions can cause agyria,a permanent blush grey pigmentation of the skin (Kim et al.2009a, b). In contrast to the bulk silver and silver ions, sil-ver nanoparticles exhibit special characteristics like aninhibiting eVect on a broad spectrum of bacteria and fungi(Wright et al. 1999, 2002). This makes them interesting forconsumer products, food technologies, textiles and medical
applications. Although the antimicrobial eVect of silverions has been studied, the impact of silver nanoparticles onmicroorganisms is only partially understood, as well asthere eVects on the human health.
Since the amount of food contact materials increasescontinuously, the greater oral uptake of silver nanoparticlesshould also be considered. In mice and rat studies, an oraladministration of 13 and 56 nm silver nanoparticles showeda dose-dependent increase in the silver concentration in theintestinal villi and in the blood (Cha et al. 2008; Kim et al.2010). That indicates that the orally adsorbed silver fromnanoparticles is able to overcome the intestinal barrier.With the blood circulation, silver can be distributed to allorgans. The endothelial cells in the gastrointestinal tractrepresent the Wrst barrier, which comes in contact withorally ingested nanoparticles.
We used the human colon carcinoma cell line Caco-2 tostudy the eVect of peptide-coated silver nanoparticles onthe intestinal barrier. The nanoparticles were 20 and 40 nmin size and were stabilized with the small peptide L-cysteineL-lysine L-lysine. They are approximately spherical withsome distortions. The colloidal properties of the nanoparti-cles depend on the pH of the dispersant. At low pH, single,well-dispersed nanoparticles are present in the solution.
Fig. 4 Results of the CellTiter-Blue® Cell Viability Assay and DAPIstaining for diVerentiated Caco-2 cells. The cells were exposed for 24, 30and 48 h to 20 and 40 nm peptide-coated silver nanoparticles (Ag-CKK)at concentrations of 5–100 �g/mL and 5 �g/mL of the peptide L-cysteine
L-lysine L-lysine (CKK) that is coating the nanoparticles. The mediumcontrol was set to 100 %. Means and standard deviations were calculatedfrom at least three independent experiments. DiVerences were calculatedby f test and t test using Microsoft Excel® (*p < 0.05)
123
Arch Toxicol (2012) 86:1107–1115 1113
With increasing pH, greater aggregates were formed (Grafet al. 2009). Our results showed a strong cytotoxic eVect ofthe silver nanoparticles on proliferating and diVerentiatedCaco-2 cells. This eVect becomes more evident withincreasing particle doses, time of incubation and decreasingparticle size. Beginning toxicity of the nanoparticles to thecells is represented by a changed morphology and a reduc-tion in the adherence capacity, which was detected with thereal-time cell analyser. This early eVect may indicate areaction of the cytoskeleton proteins to the nanoparticles.The metabolic activity measured by the CellTiter-Blue®
Cell Viability assay is Wrst inXuenced after longer incuba-tion times. Treatment with nanoparticles up to 24 or 48 hled to cell death. The toxic pathway seems to be necrosisrather than apoptosis as conWrmed by Annexin-V/7AADstaining and FACS analysis. The LDH assay showed nomembrane leakage up to an exposure of 8 h. After 24 h ofincubation, a concentration-dependent amount of LDHcould be measured in the medium, which is probablycaused by dying cells rather than by membrane damage. Apossible reason for the necrosis could be the production ofreactive oxygen species, as detected by the dichloroXuores-cein assay.
There are very little comparable toxicological studies,which use Caco-2 cells. Bouwmeester et al. (2011) tested20, 34, 61 and 110 nm silver nanoparticles for 24 h on pro-liferating Caco-2 cells, but in contrast to our results, therewas almost no inXuence on the metabolic activity up to50 �g/mL. Moreover, they used the M-cell model, which isa co-culture system with diVerentiated Caco-2 cells, and dida whole genome analysis, which demonstrated changes of
Fig. 5 Annexin-V/7AAD staining and FACS analysis to quantify theapoptosis. Proliferating Caco-2 and HepG2 cells were exposed for 2,4, 6, 8, 16, 18, 20, 22 and 24 h to medium control, 4 �M staurosporinepositive control and 5 �g/mL 20 and 40 nm peptide-coated silver nano-particles (Ag-CKK). The Caco-2 cells were additionally treated with2.5 �g/mL cycloheximide, to improve the apoptosis induction. Afterthe incubation, time cells were harvested, stained and measured usingXow cytometer
Fig. 6 Results of the CytoTox-ONE™ homogeneous membraneintegrity assay (Promega) to detect the membrane damage. Proliferat-ing Caco-2 cells were exposed to medium control and 20 and 40 nmsilver nanoparticles at a concentration of 5–100 �g/mL for 2, 4, 6, 8
and 24 h. The producer provided positive control was added short timebefore measurement. Medium control was set to 100 %. Means andstandard deviations were calculated from at least two independentexperiments
123
1114 Arch Toxicol (2012) 86:1107–1115
gene expression in a wide range of stress response genes,including oxidative stress genes, endoplasmatic stressresponse genes and genes involved in apoptosis. Due of thevery similar results for silver ions and nanoparticles, theyassumed that the eVect is just caused by released silver ionsand not by the particles itself.
Haase et al. (2011) used the same peptide-coated silvernanoparticle and could demonstrate an uptake in macro-phages. The particles were less toxic in these cells than inour study but also caused oxidative stress.
Nowrouzi et al. (2010) observed a concentration-depen-dent toxic eVect of 5–10 nm silver nanoparticles in HepG2cells by XTT and MTT assay. The microscopical analysisrevealed shrunken and rounded cells, which is comparableto our observations. In contrast to our results, they detecteda signiWcant higher amount of LDH in the culture mediacompared to the medium control. However, they also foundhints for nanoparticle-induced oxidative stress, like adecrease in the GSH content and superperoxide dismutaseactivity and an increase in the level of lipid peroxidationand the content of cytocrome c.
Another study with HepG2 cells treated with 10 nm sil-ver nanoparticles demonstrated also a dose-dependent sig-niWcant eVect with the MTT and Alamar Blue assay. Andthey used the dicloroXuorescein assay to show an inductionof oxidative stress by the silver nanoparticles. In contrast toour results, they detected DNA damage using gamma-H2AX phosphorylation (Kim et al. 2009a, b).
Kawata et al. (2009) investigated his polyethylamine sta-bilized 7–10 nm silver nanoparticles in HepG2 cells. Simi-lar to our results, they demonstrated an increase in cellviability at low concentrations and a signiWcant cytotoxiceVect at higher doses. Moreover, they also observed thatincubation with silver nanoparticles led to an abnormal
shape of HepG2 cells. In contrast to our data, they found aremarkably increased frequency of micronucleus forma-tion, indicating DNA damage and chromosome aberration.Interpreting there microarray results, they assumed in con-trast to Bouwmeester et al. (2011) that the eVect is due tothe released silver ions and the nanoparticles itself.
In summary, all cited studies found a concentration-dependent toxic eVect of diVerent silver nanoparticles indiVerent cell lines and test systems. Nearly, all of theauthors reported morphological changes in the investigatedcells. Several authors showed membrane damage of thecells after silver nanoparticle treatment. In contrast to ourstudy, some groups found a DNA damage. In our experi-ments, the cytotoxicity of the particles may overlay a possi-ble genotoxicity since we could not detect the formation ofmicronuclei after treatment of V79 Wbroblasts with silvernanoparticle. Like our study, all other investigations dis-cuss a nanoparticle-induced reactive oxygen species-depen-dent mechanism. To analyse, if the observed eVects of ournanoparticles are caused by released silver ions or the nano-particles themselves, we tested the medium supernatant of24 h pre-incubated 20 nm nanoparticles for 48 h in the CTBand DAPI assay on proliferating Caco-2 cells (data notshown). A slight eVect was only detected in the DAPI assayat a concentration of the 100 �g/mL. This indicates that ourpeptide-coated silver nanoparticles did not release enoughsilver ions within 24 h in medium to explain the toxiceVects. Hence, the nanoparticles itself are important.
Our investigations and the other mentioned studiesabove show clearly that nanoparticles have many proper-ties, e.g. chemical composition, size, surface area, surfacechemistry, water and liquid solubility and coagulation oraggregation state, which could responsible for the toxicity. Sofar, it is not completely known which properties inXuences
Fig. 7 Detection of reactive oxygen species using 2�,7�-dichlorodihy-droXuorescein diacetate. Proliferating Caco-2 cells were exposed for24 h to diVerent concentrations of peptide-coated silver nanoparticles(Ag-CKK) and positive control (FeSO4) as indicated in the Wgure. The
oxidized product 2�,7�-dichloroXuorescein was quantiWed by Xuores-cence measurement. Medium control was set to 100 %. Means andstandard deviations were calculated from at least three independentexperiments
123
Arch Toxicol (2012) 86:1107–1115 1115
mostly cell cytotoxicity. Probably, the toxicity of silvernanoparticle is a combination of diVerent properties, e.g.smaller particles exhibit a greater surface to volume ratio(Carlson et al. 2008). But also the coating, surroundingmedium and state of agglomeration inXuences the interac-tion of the nanoparticles with the exposed cells (Ahamedet al. 2008).
Acknowledgments We thank Alexandre Mantion for providing thepeptide-coated silver nanoparticles.
ConXict of interest I work at the Federal Institute for Risk Assess-ment (BfR). The BfR is independent in its scientiWc assessment andresearch work. I declare that I have no conXict of interest.
References
Ahamed M, Karns M, Goodson M, Rowe J, Hussain SM, Schlager JJ,Hong Y (2008) DNA damage response to diVerent surface chem-istry of silver nanoparticles in mammalian cells. Toxicol ApplPharmacol 233(3):404–410 (Epub 2008 Sep 27)
Bouwmeester H, Poortman J, Peters RJ, Wijma E, Kramer E, MakamaS, Puspitaninganindita K, Marvin HJP, Peijnenburg AACM, Hen-driksen PJM (2011) Characterization of translocation of silvernanoparticles and eVects on whole-genome gene expression usingan in vitro intestinal epithelium coculture model. ACS Nano5(5):4091–4103 (Epub 2011 Apr 20)
Carlson C, Hussain SM, Schrand AM, Braydich-Stolle LK, Hess KL,Jones RL, Schlager JJ (2008) Unique cellular interaction of silvernanoparticles: size-dependent generation of reactive oxygen spe-cies. J Phys B 112:13608–23619
Cha K, Hong HW, Choi G, Lee MJ, Park JH, Chae HK, Ryu G, MyungH (2008) Comparison of acute responses of mice livers to short-term exposure to nano-sized or micro-sized silver particles. Bio-technol Lett 30(11):1893–1899 (Epub 2008 Jul 5)
Chaudhry Q, Schotter M, Blackburn J, Ross B, Boxall A, Castle L, AitkenR, Watkins R (2008) Applications and implications of nanotechnolo-gies for the food sector. Food Addit Contam 25(3):241–258
EFSA (2008) Inability to assess the safety of a silver hydrosol addedfor nutritional purposes as a source of silver in food supplementsand the bioavailability of solver from this source based on the sup-porting dossier. EFSA J 884:1–3
Graf P, Mantion A, Foelske A, Shkilnyy A, Masic A, Thünemann AF,Taubert A (2009) Peptode-coated silver nanoparticles: synthesis,
surface chemistry, and pH-triggered, reversible assembly intoparticle assemblies. Chem Eur J 15(23):5831–5844
Haase A, Arlinghaus HF, Tentschert J, Jungnickel H, Graf P, MantionA, Draude F, Galla S, Plendl J, Goetz ME, Masic A, Meier W,Thünemann AF, Taubert A, Luch A (2011) Application of laserpostionization secondary neutral mass spectrometry/time-of-Xight secondary ion mass spectrometry in nanotoxicology: visu-alization of nanosilver in human macrophages and cellularresponses. ACS Nano 5(4):3059–3068 (Epub 2011 Apr 2)
Kawata K, Osawa M, Okabe S (2009) In vitro toxicity of silver nano-particles at noncytotoxic doses to HepG2 human hepatoma cells.Environ Sci Technol 43(15):6046–6051
Kim YS, Kim JS, Cho HS, Rha DS, Kim JM, Park JD, Choi BS, LimR, Chang HK, Chung YH, Kwon IH, Jeong J, Han BS, Yu IJ(2008) Twenty-eight-day oral toxicity, genotoxicity, and gender-related tissue distribution of silver nanoparticles in Sprague-Daw-ley rats. Inhal Toxicol 20(6):575–583
Kim WY, Kim J, Park JD, Ryu HY, Yu IJ (2009a) Histological studyof gender diVerences in accumulation of silver nanoparticles inkidneys of Fischer 344 rats. J Toxicol Environ Health A 72(21–22):1279–1284
Kim Y, Suh HS, Cha HJ, Kim SH, Jeong KS, Kim DH (2009b) A caseof generalized argyria after ingestion of colloidal silver solution.Am J Ind Med 52(3):246–250
Kim YS, Song MY, Park JD, Song KS, Ryu HR, Chung YH, ChangHK, Lee JH, Oh KH, Kelman BJ, Hwang IK, Yu IJ (2010) Sub-chronic oral toxicity of silver nanoparticles. Part Fibre Toxicol7:20
Klenow S, Pool-Zobel LB, Glei M (2009) InXuence of inorganic andorganic iron compounds on parameters of cell growth and sur-vival in human colon cells. Toxicol In Vitro 23:400–407
Lampen A, Bader A, Bestmann T, Winkler M, Witte L, Borlak JT(1998) Catalytic activities, protein- and mRNA-expression ofcytochrome P450 isoenzymes in intestinal cell lines. Xenobiotica28(5):429–441
Nowrouzi A, Meghrazi K, Golmohammadi T, Golestani A, AhmadianS, ShaWezadeh M, Shajary Z, Khaghani S, Amiri AN (2010)Cytotoxicity of subtoxic AgNP in human hepatoma cell line(HepG2) after long-term exposure. Iran Biomed J 14(1–2):23–32;(Project on emerging nanotechnologies) 2010 Analyses available:http://www.nanotechproject.org
Wright JB, Lam K, Hansen D, Burrell RE (1999) EYcacy of topicalsilver against fungal burn wound pathogens. Am J Infect Control27(4):344–350
Wright JB, Lam K, Buret AG, Olson ME, Burrell RE (2002) Earlyhealing events in a porcine model of contaminated wounds:eVects of nanocrystalline silver on matrix metalloproteinases, cellapoptosis, and healing. Wound Repair Regen 10(3):141–151
123