Io

AEa

b

a

ARRAA

KCIAQCN

1

tCwimripintw

h1

Mutation Research 770 (2014) 66–71

Contents lists available at ScienceDirect

Mutation Research/Genetic Toxicology andEnvironmental Mutagenesis

jo ur nal homep ag e: www.elsev ier .com/ locate /gentoxComm u ni t y add ress : www.elsev ier .com/ locate /mutres

n-vivo study of genotoxic and inflammatory effects of thergano-modified Montmorillonite Cloisite® 30B

.K. Sharmaa,∗, A. Mortensena, B. Schmidtb, H. Frandsenb, N. Hadrupa,.H. Larsenb, M.-L. Binderupa

Technical University of Denmark, National Food Institute, Division of Toxicology & Risk Assessment, Mørkhøj Bygade 19, 2860 Søborg, DenmarkTechnical University of Denmark, National Food Institute, Division of Food Chemistry, Mørkhøj Bygade 26, 2860 Søborg, Denmark

r t i c l e i n f o

rticle history:eceived 29 October 2013eceived in revised form 7 March 2014ccepted 4 April 2014vailable online 2 June 2014

eywords:omet assay

nflammationbsorptionuaternary ammonium compoundlayanocomposite

a b s t r a c t

Because of the increasing use of clays and organoclays in industrial applications it is of importance toconsider the toxicity of these materials. Recently it was reported that the commercially available Mont-morillonite clay, Cloisite® 30B, which is surface-modified by organic quaternary ammonium compounds,was genotoxic in vitro. In the present study the in-vivo genotoxic and inflammatory potential of Cloisite®

30B was investigated as a follow-up of the in-vitro studies. Wistar rats were exposed to Cloisite® 30Btwice 24 h apart by oral gavage, at doses ranging from 250 to 1000 mg/kg body weight [indicate durationof treatment; Ed.]. There was no induction of DNA strand-breaks in colon, liver and kidney cells andthere was no increase in inflammatory cytokine markers in blood-plasma samples. In order to verify thepossible absorption of Cloisite® 30B from the gastrointestinal tract, inductively coupled plasma mass-spectrometry (ICP-MS) analysis was performed on samples of liver, kidney and faeces, with aluminiumas a tracer element characteristic to clay. The results showed that aluminium could be detected in faeces,but not in the liver or kidneys. This indicated that there was no systemic exposure to clay particles from

®

Cloisite 30B. Detection and identification of free quaternary ammonium modifier in the highest doseof Cloisite® 30B was carried out by high-performance liquid chromatography coupled with quadrupoletime-of-flight mass spectrometry (HPLC-Q-TOF-MS). This analysis revealed a mixture of three quaternaryammonium analogues. The detected concentration of the organomodifier corresponded to an exposureof rats to about 5 mg quaternary ammonium analogues/kg body weight.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

Clay minerals have a number of potential applications due toheir large specific surface areas and ion-exchange properties [1].lay materials have important applications in ceramics, oil drilling,aste isolation, metal and paper industries, adsorbents, decolour-

zing agents and nanocomposite materials, including food-contactaterials [1–3]. The development and use of nanocomposites rep-

esents a new strategy to improve physical properties of polymers,ncluding mechanical strength, thermal stability, and gas-barrierroperties. Nanocomposites are a relatively new class of compos-

tes where at least one dimension of the dispersed particles is in the

anometer range. Among the potential nanocomposite precursors,hose based on clay and layered silicates have been studied mostidely, because the starting clay materials are easily available

∗ Corresponding author.E-mail address: aksha@food.dtu.dk (A.K. Sharma).

ttp://dx.doi.org/10.1016/j.mrgentox.2014.04.023383-5718/© 2014 Elsevier B.V. All rights reserved.

and their intercalation chemistry has been studied extensively.Moreover, nanocomposites have unique properties such as largesurface areas and large aspect ratios as well as improved mechan-ical, thermal and optical properties [4–6]. The most promisingnanometer-sized fillers are Montmorillonite and kaolinite clays [7].Montmorillonite is the main constituent of Bentonite, the weath-ering product of volcanic ash. A limitation to the use of clays isthe incompatibility between the hydrophilic clay and hydrophobicpolymers that could cause agglomeration of clay-mineral polymermatrices [6,8]. Therefore, to enhance the intercalation/exfoliationprocess in a polymer matrix, clay surfaces are chemically modifiedby treatment with organic compounds, so that the clays becomehydrophobic and thereby more compatible with polymers [9].Such modified clays are referred to as organoclays.

Natural and synthetic clay nanofillers in biopolymer nanocom-

posites were investigated in a project named NanoPack, which wasfunded by the Danish Strategic Research Council. Characterization,identification and migration studies of different nanofillers in thisproject [10,11] and the synthesis of biopolymer nanocomposites

ion Re

[tirnmiidtofctmtiadstpim

2

2

tns9mpT9pdo

2

4Tf8Tui(paaw

2

sbcaa3na23cmce

A.K. Sharma et al. / Mutat

12–14] were reported. A specific aim in the project was to considerhe potential toxicity of nanofillers. Generally, toxicological stud-es of clays and organo-modified clays including Montmorilloniteemain very sparse. In the NanoPack project, we recently testedatural Montmorillonite sodium salt (Cloisite® Na+) and organo-odified Montmorillonite (Cloisite® 30B) for genotoxic potential

n Caco-2 cells and found that organo-modified Montmorillonitenduced DNA damage in the comet assay in a concentration-ependent manner [15]. The aim of the present in-vivo study waso follow up on the in-vitro results. The genotoxic potential of thergano-modified clay was studied in liver, kidney and colon of ratsollowing oral administration. These three organs or tissues werehosen as potential targets of adverse effects, because the liver ishe main organ for metabolic activation after absorption of the test

aterial; the kidney is relevant in case of urinary excretion of theest substance; and the colon is relevant for excretion via the faecesn particular if the substance is not absorbed or only absorbed to

limited extent. In order to illustrate the possible absorption andistribution of Cloisite® 30B, aluminium, which is a common con-tituent of clay, was used as a marker element and was quantified inhe liver, kidney and faeces of the exposed rats by inductively cou-led plasma-mass spectrometry (ICP-MS). To investigate potential

mmunotoxic effects of the clays, markers for inflammation wereeasured in the blood.

. Material and methods

.1. Clay material

Cloisite® 30B (Southern Clay products, Gonzales, Texas, USA) was used inhis study. Cloisite® 30B is a natural Montmorillonite modified with a quater-ary ammonium salt (containing methyl, bis-2-hydroxyethyl, and C14–C18 alkylide-chains, also commonly known as tallow). The modifier concentration is0 milliequivalents/100 g clay, corresponding to about 32% (w/w). According to theanufacturer’s data sheet, the proportion of the three quaternary ammonium com-

ounds (QACs) originating from the tallow was ∼65% C18, ∼30% C16 and ∼5% C14.he typical sizes of the dry particles by volume were 10% <2 �m, 50% <6 �m and0% <13 �m (manufacturer’s data sheet). The possible content of nanometer-sizedarticles was not indicated. The specific surface area of Cloisite® 30B was 7.5 m2/getermined by the multi-point BET (Brunauer, Emmett and Teller) method basedn nitrogen adsorption [15]. The aluminium content in Cloisite® 30B is 6.7 w/w %.

.2. Animals and animal husbandry

For the main study, 21 male and 21 female Wistar Hannover Galas rats (age, weeks) with specific-pathogen-free (SPF) health status, were purchased fromaconic MB, DK-4623 Lille Skensved, Denmark. Animals were allowed to acclimatiseor 1 week. The weight (mean ± SD) of the male and female rats was 106 ± 12 g and3 ± 16 g, respectively. The rats were housed two per cage (Macrolon type III high,echniplast Gazzada S ar. L., Buguggiate, Italy) with wood bedding (Tapvei, Finland)nder controlled environmental conditions (temperature 22 ± 1 ◦C, relative humid-

ty 55 ± 5%, 12-h light/dark cycle, air change, 10 times/h) and had free access to feedAltromin 1324, Lage, Germany) and tap water acidified with citric acid, pH 3.5 (torevent growth of microorganisms). During the acclimatisation and study periodsll rats were observed at least twice daily for any abnormalities in clinical appear-nce. In a dose range-finding study, two males and two females were used and theyere acclimatised and treated as the rats used in the main study.

.3. Experimental design

In the dose range-finding study, two female and two male rats received twoingle doses of 1000 mg/kg body weight (bw) of Cloisite® 30B suspended in water,y gavage 24 h apart. This dose was the maximum feasible dose based on physico-hemical properties of the substance in water and cell-culture medium. The controlnimals (two per sex) received water. No clinical signs of toxicity were observed inny of the animals during 30 h after the first dosing. Therefore, 1000 mg Cloisite®

0B/kg bw was used as the highest dose in the main experiment. In the main study,on-fasted rats of each sex were randomly divided into dose groups (three femalend three male animals per group) and received by oral gavage two single doses,4 h apart, of either water (vehicle control) or 250, 500 or 1000 mg/kg bw of Cloisite®

0B suspended in water. In addition, three male and three female rats also receivedell-culture medium or 1000 mg/kg bw of Cloisite® 30B suspended in cell-cultureedium (DMEM medium, Life Technologies Europe, Nærum, Denmark). A positive

ontrol group of three female and three male animals received 250 mg/kg bw ofthylmethanesulfonate (EMS) suspended in water. The suspensions were freshly

search 770 (2014) 66–71 67

prepared prior to each dosing occasion to give the desired concentrations in a volumeof 1 ml/100 g bw. In order to administer a homogeneous sample, the suspensionswere placed on a magnetic stirrer until dosing. Having received the treatment (withwater or Cloisite® 30B suspended in water), the rats were placed individually inmetabolism cages, with free access to water but no feed. Faeces samples were col-lected during 24 h on dry ice and kept at −80 ◦C until ICP-MS analysis. Three hoursafter the second dosing the animals were anaesthetized in CO2/O2, and decapi-tated. Whole blood (about 0.5 ml) from the neck was collected in tubes coated withsodium-heparin and kept on ice until centrifugation at 1000 × g (0–4 ◦C, 10 min).Plasma was kept at −80 ◦C until cytokine analysis. After macroscopic examination,liver, kidneys and colon were dissected. Liver and kidneys were weighed. Two slicesfrom each kidney and four slices from the liver were cut and placed in eight differentcryotubes. The colon was opened longitudinally, rinsed in 0.9% NaCl and cells wereharvested by scraping the lumen with an object glass, and placed in a cryotube. Thecryotubes with liver, kidney and colon cells were immediately frozen in liquid nitro-gen and kept at −80 ◦C until analysis for DNA-damage in the alkaline comet assay.Most of the published studies on the in-vivo comet assay have used fresh tissues andcells processed immediately after collection. However, studies comparing fresh andfrozen samples have shown similar dose–response curves, suggesting that frozensamples can be analysed [16,17]. The animal study was conducted under conditionsapproved by The Danish Agency for Protection of Experimental Animals and thein-house Animal Welfare Committee.

2.4. Comet assay

The single-cell gel electrophoresis assay (comet assay) was performed accord-ing to [18] following the recommendations of [19], with some minor modificationsaccording to the manufacturer of the CometAssay® Kit (Trevigen, Gaithersburg,Maryland). Liver and kidney cells from each rat were isolated after homogeniza-tion of the tissues in a Downs homogenizer containing ice-cold mincing solution,i.e. Hank’s balanced salt solution (Ca2+- and Mg2+-free) containing 20 mM EDTA and10% dimethyl sulfoxide. The colon scrapings from each rat were mixed with ice-cold mincing solution. Suspensions containing 2000–4000 liver, kidney or coloncells, measured by means of a Nucleocounter (Chemotec, type 900-0004, Allerød,Denmark), were mixed with 150 �l molten CometAssayTM LMAgarose. Fifty micro-liter of this mixture was applied onto one sample area of two duplicate slidesconsisting of two gels (CometSlideTM HT, Trevigen, Gaithersburg, Maryland, USA).After solidification of the agarose, the embedded cells were lysed in a cold alkalinelysis buffer (sodium chloride 2.5 M, EDTA 0.1 M and Trizma® base 10 mM, pH 10)for 60 min. For DNA unwinding the slides were placed in the alkaline electrophore-sis solution (0.3 M sodium hydroxide, 1 mM EDTA and Millipore water, pH > 13) inthe electrophoresis jar at 4 ◦C for 40 min, and electrophoresis was run in the samebuffer for 30 min at 4 ◦C (1 V/cm and 270 mA). After neutralisation, fixation in 96%ethanol, and DNA-staining with 10 �l SYBR Green, a drop of anti-fade solution wasadded onto each gel to avoid fading. Negative (culture medium) and positive (Caco-2 cells exposed to 0.05% ethylmethane sulfonate for 2 h) controls were included,each with two gels per slide for each electrophoresis run. Fully automated cometscoring was performed by use of the PathfinderTM Cellscan Comet-imaging system(IMSTAR, Paris, France) described in detail in [20]. Tail intensity (% tail DNA) of eachcomet was used for assessment of DNA damage. The comet distributions of thesamples showed non-normal distributions, therefore median values for each organfrom each rat were used, and the mean of the median values were evaluated. Thenumber of cells scored depended on the cell density in the two gels. The numberof cells scored on the slides (sum of two gels) in the experiment where Cloisite®

30B was suspended in water was: 1656–2127 cells for the liver, 851–1324 cells forthe kidney, and 488–750 cells for the colon. In the experiment where Cloisite® 30Bwas suspended in cell-culture medium these numbers were: 893–1462 cells for theliver, 433–555 cells for the kidney, and 797–1091 cells for the colon. There were nostatistically significant differences between the numbers of cells scored in the differ-ent dose groups within each organ (one-way ANOVA analysis with Tukey’s multiplecomparison test). The number of cells on the slides was acceptable; the frequencyof overlapping cells was low.

2.5. Cytokine measurements

Samples were thawed, diluted three-fold, and assayed in duplicate. Plasmaconcentrations of interleukin 1� (IL-1�), monocyte chemotactic protein-1 (MCP1),interleukin 6 (IL-6) and tumour necrosis factor-� (TNF�) were determined withthe Milliplex Map Rat-Serum Adipokine Panel Kit (Cat. No. RATPK-81K, Millipore,Billerica, MA, USA) and a Luminex 100 reader (Bio-Rad, Hercules, CA, USA).

2.6. Chemical analyses

2.6.1. Detection of quaternary ammonium compounds (QAC) in test suspensionsHPLC-Q-TOF-MS analysis was performed as described in [15] to detect the

amount of the free QAC in the top dose of Cloisite® 30B (1000 mg/kg bw) suspendedin water or in cell-culture medium.

6 ion Re

2m

CiI

aewdd

saStPv(lcraaaktcK

atpCidctiF

omib

2

lw

Fsbf1

8 A.K. Sharma et al. / Mutat

.6.2. Detection of clay in liver, kidney and faeces by use of aluminium as aarker element

Samples of liver, kidney and faeces from rats exposed to the highest dose ofloisite® 30B (1000 mg/kg bw) suspended in water and from rats in the correspond-

ng control group were analysed for aluminium as a marker element for clay byCP-MS.

Samples of livers and kidneys were homogenized with ultra-pure water from Milli-Q apparatus (Millipore A/S, Hellerup, Denmark) in an Ultra Turrax homog-nizer (Ultra Turrax T25, Labequip, Ontario, Canada). The ratio between tissue andater was 1:1 (w/w) for all samples. Samples of faeces were dried in an oven and thery matter content determined. The samples were homogenized in a mortar afterrying.

All acid digestions were carried out in high-pressure fluoropolymer LF100 ves-els (Anton Parr, Graz, Austria) with microwave-assisted acid digestion of siliceousnd organically based matrices according to Method 3052 of the US EPA [21].ix-hundred �l liver or 100 �l kidney homogenate or 100 mg dried faeces wereransferred to a pressure vessel. Then 4.5 ml of concentrated nitric acid (HNO3,lasmaPURE 67–69%, SCP Science, Champlain, NY, USA) was added to the openessels. After the immediate reaction had ceased, 0.5 ml of hydrogen peroxideH2O2, Suprapur 30%, Merck, Darmstadt, Germany) was added. The vessels wereeft overnight for additional chemical reaction to take place. Then 1 ml of con-entrated hydrochloric acid was added (PlasmaPURE 34–37%, SCP Science). Aftereaction 1.5 ml of hydrofluoric acid (HF, Suprapur 40%, Merck, Denmark) was addednd the bombs were left standing for 30 min. The pressure vessels were then closednd subjected to microwave-assisted digestion according to the following temper-ture and time program: increasing from room temperature for 5.5 min to 180 ◦C,eeping constant for 9.5 min, and cooling for 30 min. After digestion and coolinghe content of each vessel was transferred quantitatively to a 50-ml Sarstedt tubeontaining 10 ml of 6% boric acid (Puratronic 99.9995%, Alfa Aesar GmbH & Co GK,arlsruhe, Germany) in ultra-pure water and taken to a 25-ml volume with water.

Chemical blanks were produced by addition of digestion chemicals only. Finally, range of masses (20 mg, 40 mg, 60 mg or 100 mg) of Cloisite® 30B were takenhrough the digestion procedure as well. The digested samples were diluted in ultra-ure water before ICP-MS analysis (Agilent 7500ce, Agilent Technologies, Santalara, CA, USA). In this study aluminium was used as a tracer element character-

stic to clay. To determine the aluminium content, 0.1 ml of the acid digest wasiluted to 10 ml using ultra-pure water, which corresponds to a concentration oflay at about 10 ng/ml. Scandium (Sc) was used as internal standard that was addedo all samples, standards and blanks. External calibration standards, with the samenternal standard as for the samples, were analysed at the start of each sequence.or aluminium, the calibration standards ranged from 0 to 500 ng/ml.

The ICP-MS was operated in spectrum-analysis mode with an integration timef 0.3 s per mass and three repetitions. Helium was used as collision-cell gas for alleasurements. Furthermore, the ICP-MS was cleaned between samples with three

njections of HCl and HNO3, alone and in a mixture, in order to avoid carry-overetween samples.

.7. Statistics

The comet-assay data and the results of the cytokine measurements were ana-ysed by means of one-way ANOVA with Dunnett’s test to compare the dosed groups

ith the control group. The positive control group (exposed to EMS) was analysed

ig. 1. Results of the comet assay with various tissues from Wistar rats (n = 6) exposed

uspended in cell-culture medium (1000 mg/kg body weight). EMS (ethylmethane sulfonaody weight. Data from liver, kidney and colon cells of the EMS-exposed group were statrom the values in the corresponding control group (two-tailed, unpaired t-test). Internal s6 slides, % tail DNA, mean ± S.D., 21.6 ± 6.6; negative controls (untreated Caco-2 cells): 1

search 770 (2014) 66–71

against the control dose group by use of a two-tailed, unpaired t-test. Grad Pad Prism5.0 was used as statistical software.

3. Results

3.1. Genotoxicity and inflammation

All animals survived throughout the study. No abnormalitiesin clinical appearance were observed in the rats of the differenttreatment groups. Fig. 1 shows the results for DNA damage in liver,kidney, and colon cells after administration of Cloisite® 30B sus-pended in water or cell-culture medium, as measured with thecomet assay. The results from male and female rats in the dif-ferent dose groups are pooled, because there was no statisticallysignificant difference in DNA damage between male and femaleanimals. There were no statistically significant differences in % tailDNA between the negative controls and the different dose groupsin either of the organs tested for DNA damage. For the liver samples,the results were 3.5 and 4.2% tail DNA for treatment with Cloisite®

30B suspended in water or in cell-culture medium, respectively.The corresponding control values were 2.8 and 4.4% tail DNA. Theresults for the kidney samples were 3.8 and 8.7% tail DNA afterthe two different treatments, respectively, compared with controlvalues of 4.4 and 7.0%. For the colon samples the results were 7.8and 18.5% tail DNA for the two treatment groups, and 13.4 and18.3% tail DNA for the controls. There was a statistically signifi-cantly increase in tail DNA in the liver, kidney and colon cells of theEMS-exposed animals (positive control) compared with the vehiclecontrols, demonstrating the sensitivity of the assay.

Blood-plasma concentrations of IL-1�, MCP1, IL-6 and TNF�were measured to investigate the inflammatory effects of Cloisite®

30B. Table 2 shows the results. None of the plasma concentrationsof the cytokines tested were altered after exposure to Cloisite®

30B suspended in either water or cell culture medium. The TNF�and IL-1� concentrations in blood plasma of the EMS-exposedanimals (positive control) were statistically significantly differentfrom those in the unexposed control animals.

3.2. Chemical analyses

The analyses of the QAC in the Cloisite® 30B suspended in waterfor dosing at 1000 mg/kg bw showed that the three QACs, C14, C16and C18 could not be detected, indicating that they were not present

to Cloisite® 30B suspended in water (250, 500 and 1000 mg/kg body weight) andte) suspended in water was the positive control: six rats were exposed to 250 mg/kgistically significantly different (p < 0.001, p < 0.001: *** and p < 0.05: *), respectively,tandards: positive controls (Caco-2 cells exposed to 0.05% ethylmethane sulfonate):6 slides, % tail DNA, mean ± S.D., 1.8 ± 0.6.

A.K. Sharma et al. / Mutation Research 770 (2014) 66–71 69

Table 1Detection of aluminium content (�g aluminium/g tissue) in control rats and in rats exposed to 1000 mg/kg bw of Cloisite® 30B suspended in water. Detection by ICP-MS.Data are mean ± SD (n = 6).

Feces Liver Kidney

Controls Cloisite® 30B Controls Cloisite® 30B Controls Cloisite® 30B

<

aiawrots

nbmbtaetc

4

ncbtcsct

smspariekoh

TI

�g

<LOD 1818 ± 199 <LOD

LOD: Below level of detection. LOD (�g/g): liver = 50, kidney = 290, feces = 1150.

s free substances in the suspension. The detection of the QACsn the Cloisite® 30B suspended in cell-culture medium for dosingt 1000 mg/kg bw showed that the three QACs, C14, C16 and C18,ere indeed present, at concentrations of 55, 210 and 255 �g/ml,

espectively. This adds up to a total concentration of the mixturef the three QAC of 520 �g/ml corresponding to an exposure ofhe rats to 5 mg QAC as free substance/kg bw. In the control-doseuspension (only cell-culture medium) no QAC could be detected.

The presence of clay or clay minerals in samples of liver, kid-ey and faeces from the rats in the high-dose group (1000 mg/kgw) was measured and the results are shown in Table 1. The alu-inium dose originating from the Cloisite® 30B was calculated to

e 67 mg/kg bw, based on the information of the aluminium con-ent in the product. The tracer element aluminium was detectednd quantified in the faecal samples, but it was undetectable in liv-rs and kidneys. These results indicated that the clay remained inhe faeces and that it was not absorbed as particles or as dissolvedlay minerals.

. Discussion

Recently we tested the organo-modified Cloisite® 30B and theon-modified Cloisite® Na+ in the comet assay in vitro with Caco-2ells [15]. That study showed a genotoxic effect of Cloisite® 30B,ut not of Cloisite® Na+ and it was demonstrated that the geno-oxic effect was most likely due to the free quaternary ammoniumompounds (QAC) in Cloisite® 30B. In contrast, the in-vivo findingshowed no induction of DNA strand-breaks in liver, kidney or colonells from rats given Cloisite® 30B by oral administration, indicatinghat the test material was not genotoxic in vivo.

The rats received Cloisite® 30B by gavage, and therefore the tis-ue of the lumen of the gastrointestinal tract was exposed to the testaterial. The release of chemical elements in the clay and their sub-

equent absorption from the gastrointestinal tract could have takenlace [22,23]. In the present study, the ICP-MS data with aluminiums a tracer element characteristic of clays showed that clay mate-ial was present in the faeces, but no aluminium could be detectedn the liver and kidneys. This indicated that there was no systemic

xposure to clay material, because the presence of aluminium inidneys and liver would have been expected, considering the datan aluminium distribution after oral exposure [24,25]. On the otherand, only a minor fraction of the aluminium administered orally

able 2nflammatory cytokine plasma-concentrations after exposure of Wistar rats to Cloisite® 3

Dose (mg/kg body weight)

Controls 250

TNF� (pg/ml)˛ 8.4 ± 1.7 7.7 ± 1.0

TNF� (pg/ml)¤ <LOD

IL-1� (pg/ml)˛ 15.7 ± 5.2 12.2 ± 3.1

IL-1� (pg/ml)¤ 21.1 ± 14.6

IL-6 (pg/ml)˛ 208 ± 140 333 ± 166

IL-6 (pg/ml)¤ 420 ± 31

MCP1 (pg/ml)˛ 207 ± 64 256 ± 77

MCP1 (pg/ml)¤ 1009 ± 227

: Suspended in water, ¤: suspended in cell-culture medium, <LOD: below level of deteroup, six rats exposed to 250 mg/kg body weight. LOD (pg/ml); TNF� = 2.0, IL-1� = 1.0, IL** p < 0.01 (two-tailed, unpaired t-test).

<LOD <LOD <LOD

is available for absorption [25]. That the content of aluminium inthe faeces was lower than expected could be due to the fact thatthe rats were not fasted prior to dosing and, therefore, the collected24-h faecal samples may not contain all of the administered doseof aluminium. The negative results observed in the comet assay oncolon cells suggested the absence of a genotoxic effect of Cloisite®

30B in the colon of the rats. The lack of any genotoxic response inliver and kidney cells also indicated that the QAC surface-coatingof Cloisite® 30B did not exert any systemic genotoxic effect.

Absorption and distribution studies of clay particles in rodentsafter oral exposure are sparse. Absorption and organ distributionof chemical elements from bentonite clays have been reportedwith concentrations in the order: kidney as the major organ, fol-lowed by liver, heart, and brain [22]. Another study investigatedthe calcium and sodium salts of Montmorillonite clay at doses upto 1000 mg/kg bw/day that were administered in a balanced diet toSprague–Dawley rats during pregnancy [26]. The results indicatedthat there were no minerals originating from clay particles in any ofthe organs investigated, including liver and kidney on gestation day16. In another study, sediment consisting of 15% Montmorilloniteclay spiked with radiolabelled [14C]-hexachlorbenzene was givenby oral gavage to rats, and after 24 h several organs were analysedfor radioactivity [27]. About 3–4% was detected in the gastroin-testinal tract. Most of the 14C-hexachlorbenzene was detected inthe faeces, carcass and skin [27]. In the present study, the organo-modified Montmorillonite clay Cloisite® 30B was tested and thebioavailability may differ from that of unmodified clay. To addressthis issue more detailed ADME studies are needed.

Inflammation, if present, could induce oxidative stress that maylead to genotoxic effects [28]. However, in the present study, theconcentrations of inflammatory cytokines in the blood plasma ofthe rats showed no increase between the exposed groups and thecontrols. This could be due to lack of internal exposure resultingfrom no or negligible absorption of the test material. The posi-tive control substance EMS decreased the plasma concentrationsof IL-1� and TNF�. The relevance of this finding remains unclear,although it could indicate a reduced inflammatory response in therats upon exposure to EMS.

To our knowledge, only one paper has addressed the in-vivogenotoxic potential of Montmorillonite clay [29]. Nanometer-sized platelets of the sodium salt of Montmorillonite weretested for genotoxic effects in the micronucleus test with mouse

0B suspended in water or cell-culture medium. Data are mean ± SD (n = 6).

500 1000 EMS

7.2 ± 0.4 7.1 ± 1.4 4.8 ± 1.9**

<LOD14.0 ± 6.0 11.4 ± 6.2 4.2 ± 2.3**

21.8 ± 4.1200 ± 180 289 ± 108 99.6 ± 65.9

521 ± 553269 ± 51 255 ± 44 247.5 ± 83.4

1349 ± 277

ction. EMS (ethylmethane sulfonate) suspended in water was the positive control-6 = 15.4, MCP1 = 34.8.

7 ion Re

lbmdwrb

cMib(fwiHaNnm

cbdwaCddrbpQtonaoootwwad7wstbtsri

igttsb

Cal

[

[

[

[

[

[

[

[

0 A.K. Sharma et al. / Mutat

ymphocytes but did not show any effect at doses up to 500 mg/kgw. Tests involving hydrated calcium silicate were carried out inice, which received oral doses up to 1500 mg/kg bw. The clay

id not increase mutation frequencies in a host-mediated assayith the microorganisms Salmonella TA-1530 and G-46 and Saccha-

omyces D3. Furthermore, there was no induction of micronuclei inone marrow of rats at doses up to 5000 mg/kg bw [30].

Apart from genotoxic effects, information on other toxicologi-al effects of Montmorillonite clays in vivo is limited. Unmodifiedontmorillonite clay induced very weak haematological effects

n rats after oral exposure for 72 h at doses up to 143 mg/kgw [31]. A naturally occurring calcium salt of MontmorilloniteNovaSil) was given to rats in the diet at 300–1200 mg/kg bw/dayor 28 weeks. There was no effect on feed consumption, bodyeight, body-weight gain, and haematological and serum biochem-

cal parameters [32]. NovaSil clay has also been tested in humans.ealthy human subjects received a capsule once daily with either

low (1.5 g/day) or a high dose (3 g/day) for 2 weeks. No effects ofovaSil were observed on haematological and urinary parameters,or on liver and kidney function, electrolytes, vitamins A and E andineral composition of these tissues and body fluids [33].In our previous in-vitro study, Cloisite® 30B suspended in cell-

ulture medium gave a genotoxic effect that was most likely causedy the QAC [15]. Therefore, we also used Cloisite® 30B at the highestose (1000 mg/kg bw) suspended in cell-culture medium. The QACas quantified in Cloisite® 30B suspended in cell-culture medium

nd the ranking of the three QAC was C14 ∼ 11%, C16 ∼ 40% and18 ∼ 49%. This ranking is in accordance with the manufacturer’sata sheet and suggests that all types of alkyl side-chain becomeisplaced from the clay surface. The quantification of the QAC cor-esponded to a combined exposure of rats to about 5 mg QAC/kgw. In contrast, the free QAC could not be detected in the sus-ension of Cloisite® 30B in water. A possible reason was that theAC was bound to the clay. On the other hand, it may be assumed

hat the QAC, even when administered in an aqueous suspensionf the organo-modified clay, may be released in the gastrointesti-al tract of rats because of chemical reaction with proteins ormino acids or displacement caused by salts. However, the extentf release remains undetectable and uncertain. In the literature,nly a few studies have reported the absorption and distributionf QACs. In one study, rats received orally [14C]-labelled hexadecylrimethyl ammonium bromide [34]. About 80% of the radioactivityas found in the gastrointestinal tract 8 h after administration andithin 3 days 92% of the radioactivity was excreted via the feces

nd 1% via urine. Another study investigated radiolabelled didecyl-imethylammonium chloride [35]. The experiments showed that

days after oral exposure, 89–99% of the recovered radioactivityas found in the faeces, and less than 2.5% in urine. These two

tudies indicated very low absorption of QAC from the gastroin-estinal tract. Up to 8 h after exposure most of the QAC could stille found in the gastrointestinal tract. Even though the QACs fromhese two studies are different from the QAC used in the presenttudy, it may be assumed that at the time when the rats were sac-ificed (3 h after the second exposure) most of the QAC was presentn the gastrointestinal tract.

QACs have been used for decades and common applicationsnclude uses as disinfectants and sanitizers [36]. Whereas in-vivoenotoxicity studies of QACs are sparse, some studies have reportedhe genotoxic potential of QACs in vitro. One in-vivo study reportedhat didecyl-dimethylammonium chloride did not induce chromo-ome aberrations in rat bone-marrow cells at doses up to 600 mg/kgw [35].

In conclusion, our results indicate that the QAC-modifiedloisite® 30B did not induce DNA strand-breaks in liver, kidneynd colon cells of rats exposed orally by gavage. Tissues of theumen of the gastrointestinal system were exposed to Cloisite®

[

search 770 (2014) 66–71

30B, but no genotoxic effect was induced. Aluminium was usedas tracer element, characteristic of clay particles, but could not bedetected in the liver and kidney. This indicated that there was nodetectable internal exposure to the clay component of Cloisite®

30B. Furthermore, the negative results of the comet assay in thesetwo organs and the absence of any detectable effect on plasma con-centrations of inflammatory cytokines indicated that Cloisite® 30Bdid not cause any systemic effect to rats at doses up to 1000 mg/kgbw.

Conflict of interests

There are no conflicts of interests.

Acknowledgements

The study is part of the project NanoPack and is a national fundedproject by NABIIT Grant number 2106-06-0061 under the Dan-ish Research and Innovation Agency. Vivian Jørgensen and MajaDanielsen DTU-National Food Institute, Divison of Toxicology andRisk Assessment are acknowledged for their skillful technical assis-tance.

References

[1] J.T. Kloprogge, S. Komarneni, J.E. Amonette, Synthesis of smectite clay minerals:a critical review, Clays Clay Miner. 47 (1999) 529–554.

[2] Q. Chaudhry, M. Scotter, J. Blackburn, B. Ross, A. Boxall, L. Castle, R. Aitken, R.Watkins, Applications and implications of nanotechnologies for the food sector,Food Addit. Contam. 25 (2008) 241–258.

[3] J.T. Kloprogge, Synthesis of smectites and porous pillared clay catalysts: areview, J. Porous Mater. 5 (1998) 5–41.

[4] S. Pavlidou, C.D. Papaspyrides, A review on polymer-layered silicate nanocom-posites, Prog. Polym. Sci. 33 (2008) 1119–1198.

[5] L.A. Utracki, M.R. Kamal, Clay-containing polymeric nanocomposites, ArabianJ. Sci. Eng. 27 (2002) 43–67.

[6] M. Ellenbecker, S. Tsai., Engineered nanoparticles: safer substitutes for toxicmaterials, or a new hazard? J. Clean. Prod. 19 (2011) 483–487.

[7] A. Arora, G.W. Padua, Review: nanocomposites in food packaging, J. Food Sci.75 (2010) R43–R49.

[8] T. Deutschle, U. Porkert, R. Reiter, T. Keck, H. Riechelmann, In vitro genotox-icity and cytotoxicity of benzalkonium chloride, Toxicol. In Vitro 20 (2006)1472–1477.

[9] F. Chivrac, E. Pollet, L. Averous, Progress in nano-biocomposites based onpolysaccharides and nanoclays, Mater. Sci. Eng. R 67 (2009) 1–17.

10] B. Schmidt, J. Petersen, H.C. der Koch, D. Plackett, N. Johansen, R.V. Katiyar,E. Larsen H, Combining asymmetrical flow field-flow fractionation with light-scattering and inductively coupled plasma mass spectrometric detection forcharacterization of nanoclay used in biopolymer nanocomposites, Food Addit.Contam. Part A Chem. Anal. Control Expo. Risk Assess. 26 (2009) 1619–1627.

11] B. Schmidt, V. Katiyar, D. Plackett, E. Larsen, H.N. Gerds, C. Koch, J. Bender, H.Petersen, Migration of nanosized layered double hydroxide platelets from poly-lactide nanocomposite films, Food Addit. Contam. Part A Chem. Anal. ControlExpo. Risk Assess. 28 (2011) 956–966.

12] N. Gerds, V. Katiyar, C.B. Koch, J. Risbo, D. Plackett, H.C. Hansen, Synthesisand characterization of laurate-intercalated Mg-Al layered double hydroxideprepared by coprecipitation, Appl. Clay Sci. 65–66 (2012) 143–151.

13] N. Gerds, V. Katiyar, C.B. Koch, H.C.B. Hansen, D. Plackett, E.H. Larsen, J.Risbo, Degradation of l-polylactide during melt processing with layered doublehydroxides, Polym. Degrad. Stab. 97 (2012) 2002–2009.

14] V. Katiyar, N. Gerds, C.B. Koch, J. Risbo, H.C. Hansen, D. Plackett, Melt processingof poly(L-lactic acid) in the presence of organomodified anionic or cationicclays, J. Appl. Polym. Sci. 122 (2011) 112–125.

15] A.K. Sharma, B.r. Schmidt, H. Frandsen, N.R. Jacobsen, E.H. Larsen, M.L. Binderup,Genotoxicity of unmodified and organo-modified montmorillonite, Mutat.Res./Genet. Toxicol. Environ. Mutagen. 700 (2010) 18–25.

16] P. Jackson, L. Pedersen, Z. Kyjovska, N. Jacobsen, A. Saber, K. Hougaard, U. Vogel,W.H. Validation of freezing tissues and cells for analysis of DNA strand breaklevels by comet assay, Mutagenesis 28 (2013) 699–707.

17] L. Recio, G. Kissling, C. Hobbs, K. Witt., Comparison of Comet assaydose-response for ethyl methanesulfonate using freshly prepared versus cryo-

preserved tissues, Environ. Mol. Mutagen. 53 (2011) 101–113.

18] R.R. Tice, E. Agurell, D. Anderson, B. Burlinson, A. Hartmann, H. Kobayashi, Y.Miyamae, E. Rojas, J.-C. Ruy, Y.F. Sasaki, Single cell gel/comet assay: guidelinesfor in vitro and in vivo genetic toxicology testing, Environ. Mol. Mutagen. 35(2000) 206–221.

ion Re

[

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

[DDAC, Canadian Cataloguing in Publication Data, Henderson, N.D., 1992, pp.

A.K. Sharma et al. / Mutat

19] A. Hartmann, E. Agurell, C. Beevers, S. Brendler-Schwaab, B. Burlinson, P. Clay,A. Collins, A. Smith, G. Speit, V. Thybaud, R.R. Tice., Recommendations for con-ducting the in vivo alkaline Comet assay, Mutagenesis 18 (2003) 45–51.

20] A.K. Sharma, F. Soussaline, J. Sallette, M. Dybdahl, The influence of the numberof cells scored on the sensitivity in the comet assay, Mutat. Res./Genet. Toxicol.Environ. Mutagen. 749 (2012) 70–75.

21] US EPA, METHOD 3052- Microwave Assisted Acid Digestion of Siliceous andOrganically Based Matrices, US EPA, 1996, pp. 1–20, Ref type: report.

22] N. Mascolo, V. Summa, F. Tateo, In vivo experimental data on the mobility ofhazardous chemical elements from clays, Appl. Clay Sci. 25 (2004) 23–28.

23] F. Tateo, V. Summa, Element mobility in clays for healing use, Appl. Clay Sci. 36(2007) 64–76.

24] G. Garbossa, G. Galvez, G. Perez, J. Stripeikis, M. Tudino, A. Nesse, Oral alu-minium administration to rats with normal renal function. 2. Body distribution,Hum. Exp. Toxicol. 17 (1998) 318–322.

25] EFSA, Safety of aluminium from dietary intake, EFSA J. 754 (2008) 1–34.26] M. Wiles, H. Huebner, E. Afriyie-Gyawu, R. Taylor, G. Bratton, T. Phillips, Toxico-

logical evaluation and metal bioavailability in pregnant rats following exposureto clay minerals in the diet, J. Toxicol. Environ. Health A Curr. Issues 67 (2004)863–874.

27] S.A. Saghir, M.J. Bartels, R.A. Budinsky, J. Harris, E. Eric, A.J. Clark, J.L. Staley, Y.Chai, J.W. Davis, - Effect of organic carbon content, clay type, and aging on theoral bioavailability of hexachlorobenzene in rats, Environ. Toxicol. Chem. 26(2007) 2420–2429.

28] A.M. Knaapen, P.J. Borm, C. Albrecht, R.P. Schins, Inhaled particles and lungcancer. Part A: mechanisms, Int. J. Cancer 109 (2004) 799–809.

29] P.R. Li, J.C. Wei, Y.F. Chiu, H.L. Su, F.C. Peng, J.J. Lin., Evaluation on cytotoxicityand genotoxicity of the exfoliated silicate nanoclay, ACS Appl. Mater. Interfaces2 (2010) 1608–1613.

[

search 770 (2014) 66–71 71

30] A.R. Elmore, F.A. Andersen, Final report on the safety assessment of aluminiumsilicate, calcium silicate, magnesium aluminium silicate, magnesium silicate,magnesium trisilicate, sodium magnesium silicate, zirconium silicate, atta-pulgite, bentonite, Fuller’s earth, hectorite, kaolin, lithium magnesium silicate,lithium magnesium sodium silicate, montmorillonite, pyrophyllite, and zeolite,Int. J. Toxicol. 22 (2003) 37–102.

31] Y.-H. Lee, T.-F. Kuo, B.-Y. Chen, Y.-K. Feng, Y.-R. Wen, W.-C. Lin, L.F.H. Toxicityassessment of montmorillonite as a drug carrier for pharmaceutical appli-cations: yeast and rats model, Biomed. Eng. Appl. Basis Commun. 17 (2005)12–18.

32] E. Friyie-Gyawu, J. Mackie, B. Dash, M. Wiles, J. Taylor, H. Huebner, L.Tang, H. Guan, J.S. Wang, T. Phillips, Chronic toxicological evaluation ofdietary NovaSil Clay in Sprague-Dawley rats, Food Addit. Contam. 22 (2005)259–269.

33] J.S. Wang, H. Luo, M. Billam, Z. Wang, H. Guan, L. Tang, T. Goldston, E.friyie-Gyawu, C. Lovett, J. Griswold, B. Brattin, R.J. Taylor, H.J. Huebner,T.D. Phillips, Short-term safety evaluation of processed calcium mont-morillonite clay (NovaSil) in humans, Food Addit. Contam. 22 (2005)270–279.

34] B. Isomaa, J. Reuter, B. Djupsund, Subacute and chronic toxicity ofcetyltrimethylammonium bromide (Ctab). A cationic surfactant in ratauthor(S), Arch. Toxicol. 35 (1976) 91–96.

35] N.D. Henderson, A Review of the Environmental Impact and Toxic Effects of

1–45.36] A. Baker, Quaternary ammonium compound (QAC): case study of disinfect-

ant field application methodology and effectiveness during hurricane katrinaflooding, J. Occup. Environ. Hyg. 4 (2007) D95–D102.