This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Toxicology Letters 192 (2010) 101–107

Contents lists available at ScienceDirect

Toxicology Letters

journa l homepage: www.e lsev ier .com/ locate / tox le t

A comparative study of leukaemia inhibitory factor and interleukin-1�intracellular content in a human keratinocyte cell line after exposureto cosmetic fragrances and sodium dodecyl sulphate

Alessandro Parodia,b,∗,1, Roberta Sanguineti a,c,1, Mariafrancesca Catalanoa, Susanna Pencoa,Maria Adelaide Pronzatoa, Chiara Scanarotti a, Anna Maria Bassi a

a Department of Experimental Medicine, Via Leon Battista Alberti, 2, 16132, Genova, Italyb Advanced Biotechnology Centre, L.go R. Benzi, 10, 16132, Genova, Italyc Department of Internal Medicine and Medical Specialties, Viale Benedetto XV, 6, 16132, Genova, Italy

a r t i c l e i n f o

Article history:Received 25 August 2009Received in revised form 10 October 2009Accepted 12 October 2009Available online 28 October 2009

Keywords:Skin irritationIn vitroHuman keratinocytesIL-1�LIF

a b s t r a c t

According to European laws animal testing in cosmetic industry will be prohibited in a few years and itwill be replaced by alternative methods based on cell and tissue culture. Many ingredients of cosmeticformulations are potentially causes of skin inflammation and sensibilization. Since cytotoxicity is known,among other factors, to trigger irritation, in an alternative model for evaluation of skin irritation, it canbe considered also the precocious release of inflammatory mediators, i.e. cytokines, originating mainlyfrom keratinocytes.

In this in vitro study we have analysed some parameters directly or indirectly related to irrita-tion/inflammation, in NCTC 2544 human keratinocytes during short-time exposure to some potentialirritants cosmetic fragrances, included in the European Laws 2003/15/EEC. IIC50 was extrapolated by MTTand NRU viability indexes after exposure of cell ultures to Geraniol Limonene and Benzylic Alcohol for 1, 3and 6 h. NCTC cells were then exposed to sub-toxic doses of selected compounds and interleukin-1� (IL-1�) and leukaemia inhibitory factor (LIF) expressions were analysed as early proinflammatory cytokines.

To our knowledge our findings demonstrated for the first time that NCTC cells synthesize and modulateLIF after exposure to selected irritating stimuli. Moreover, our results give evidence on LIF role as in vitroprecocious endpoint for the assessment of the risk in cosmetic field, because its response under irritationstimuli is very quick and comparable to IL-1�.

© 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

All protocols capable of reducing, refining and replacing in vivotesting can be considered as alternative methods (Bulger, 1987;Russell, 1995). The development of alternative methods is, besidesbeing an ethical dogma, also a scientific duty. Some evidenceshowed that the animal model is often too far from the humanmodel to have an exact and direct transferability (Eun and Suth,2000).

The validation of an alternative method is a long and complexwork aimed at constant refinement of in vitro procedures so as tomake them unassailable in terms of prediction power and repro-ducibility.

∗ Corresponding author at: University of Genova, Department of ExperimentalMedicine, Via L.B. Alberti, n.2, 16132 Genova, Italy. Tel.: +39 010 3538821;fax: +39 010 3538836.

E-mail address: aleparodi1978@libero.it (A. Parodi).1 These authors contributed equally to this work.

Recently, the development of new alternative protocols hasacquired new vigor, owing also to farsighted European laws thatseverely restricted the trade of substances tested on animals.The development of chemicals is faster than the developmentof new alternative toxicological methods, and in particularin cosmetic formulations there are several ingredients thatcould trigger skin inflammation: alcohols (Chen, 2005), sur-factants (Martinez et al., 2006) and preservatives (Lee et al.,2007).

The European Directive 2003/15/EEC aims to introduce, withinthe European Union, the final and permanent ban of animal testingfor the “cosmetic finished products as well as for their ingredi-ents”. This directive designates the year 2009 as the deadline forbanning the marketing of cosmetic products, which have been ani-mal tested; the deadline for repeated dose toxicity, reproductivetoxicity and toxicokinetics is 2013 because of lack of appropriatealternative tests.

Moreover the 2003/15/EEC directive also forces the cosmeticindustry to include 26 fragrances, suspected of sensibilization orirritation, in the list of product ingredients.

0378-4274/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.toxlet.2009.10.013

Author's personal copy

102 A. Parodi et al. / Toxicology Letters 192 (2010) 101–107

Until now, under the supervision of the European Centre for theValidation of Alternative Methods (ECVAM), many in vitro methods,for the assessment of skin irritation, were considered worthwhileinserting into the validation process, but only few of them werevalidated (Botham, 2004; Spielmann et al., 2007).

In response to physical or chemical stress keratinocytes produceand release inflammatory cytokines, growth factors and immunomediators. So the analysis of these parameters may be useful toidentify the human risk of exposure to potential irritant com-pound(s).

Two recently validated methods, EPISKIN® and EPIDERM®

assays, resembling in vivo human skin, assess cellular toxicity byMTT test and a semi-quantitative evaluation of interleukin-1 alpha(IL-1�) secretion (see: Method B.40 bis. in Annex of CommissionRegulation 440/2008/EC, 2000, 2008).

IL-1� is a cytokine universally known as an early inflammationmarker (Dinarello, 2000) whose biological active isoform is storedalso in keratinocytes (Martinez et al., 2006). For these reasons thiscytokine is usually used as a reference positive control during theexperimental evaluation of acute skin inflammation tests (Martinezet al., 2006) and during the assessment of chemical irritation risk(Cotovio et al., 2005), according to the ECVAM Scientific AdvisoryCommittee (ESAC) no. 26 on the statement on the validity of in vitrotests for skin irritation (2007).

EPISKIN® and EPIDERM® assays evidence an inter-batch varia-tion for IL-1� release and it is impossible to use a standard % release.According to ESAC the results must be expressed in amount of IL-1� release, and the limit or irritation potential is more than 60 pgof cytokine/ml.

So the identification of new inflammation biomarkers is a use-ful step for the improvement and the validation of new in vitromethods.

Leukaemia inhibitory factor (LIF) is a cytokine generally asso-ciated to cell differentiation process, above all during stem cellmaturation. LIF induces differentiation on the murine leukaemiacell line and is involved in many inflammatory diseases (Tomidaet al., 1984; Baumann et al., 1984; Mori et al., 1991; Villiger et al.,1992; Waring et al., 1992; Lotz et al., 1992; Gearing, 1993; Pagliaet al., 1996).

The role of LIF in skin diseases is unclear and it is still a controver-sial matter of discussion by the scientific community; nevertheless,its pro-inflammatory role seems to be the most accredited (Zhuet al., 2001; Banner et al., 1998). LIF is secreted by keratinocytesin response to irritating stimuli, and it was associated to the ini-tial events that characterize contact dermatitis (Szepietowski etal., 1997; McKenzie and Szepietowski, 2001). LIF secretion wasreciprocally associated to the secretion of other pro-inflammatorycytokines, such as IL-1, IL-6 and TNF� (Villiger et al., 1992), and itsinvolvement in the first steps of contact dermatitis is supported byits great reactivity to inflammation stimuli. For these evidences LIFcould be suggested as ideal biomarker for designing and developingnew alternative toxicological methods for skin diseases.

Several literature data report that NCTC 2544 keratinocytesexpress several cytokines such as IL-1� IL-8, IL-18, Il-6, Il-10, Tumornecrosis factor �, TNF� (Grandjean-Laquerriere et al., 2003, 2007,2005; Martinez et al., 2006; Sanchez et al., 2006).

The aim of this work is to evaluate the intracellular modulationof LIF in this non-differentiated human keratinocyte cell line afterbrief exposure to Geraniol, Limonene and Benzylic Alcohol, threecommon fragrances present in the list of the 26 aromatic fragrances(2003/15/EEC) and to a surfactant such as sodium dodecyl sulphate(SDS), which is commonly used as positive control in skin toxicityand inflammation experiments (Sanchez et al., 2004; Burlando etal., 2008). In our study we verified the effects of the experimen-tal treatments on cell viability by MTT and NRU cytotoxicity testsand, as more specific irritant endpoints, we further analysed intra-

cellular levels of IL-1� and LIF, in order to assess if also the lattercytokine might be considered as a predictive biomarker for a skinirritating potential compound.

2. Materials and methods

2.1. Plastic-ware and reagents

Cell culture flasks and plates were purchased from Costar (Cambridge, MA, USA).Standard culture medium, foetal bovine serum (FBS), SDS, and glutamine were pur-chased from Sigma–Aldrich (Milan, Italy). Primary and secondary antibodies werefrom Santa Cruz Biotechnology (Santa Cruz, CA, USA). PVDF membranes (Hybond-P),Enhanced Western Blotting Analysis System (ECL) and polyclonal secondary anti-bodies were supplied by GE Healthcare (Amersham, Bucks, U.K.) and Sigma. TripLEExpress was purchased from Gibco (Invitrogen Life Technologies, Milan, Italy). Brad-ford reagent dye concentrate was obtained from Bio-Rad (Hercules, CA). Geraniol,Limonene and Benzylic Alcohol were kindly provided by Prof. E. Mariani (Depart-ment of Pharmaceutical Sciences, University of Genoa, Italy). All other chemicalsand reagents were of the highest purity available and were purchased from Merck(Darmstadt, Germany).

2.2. Cell cultures

Normal human undifferentiated keratinocyte, NCTC 2544, cell line was providedby Interlab Cell Line Collection (Genoa, Italy), and was maintained in DMEM mediumsupplemented with 10% FBS, at 37 ◦C in a humidified atmosphere containing 5%CO2. The medium was changed every 2–3 days. Cells were sub-cultured by TripLEExpress treatment when the original flask was approximately 80% confluent. All cellcultures were found to be mycoplasma-free on regular checks with the Reagent SetMycoplasma Euroclone (Celbio, Pero, Milano, Italy).

2.3. Cell treatments

NCTC 2544 cells were seeded 24 h before treatment in 96-well plates at5 × 104 cells/well so as to obtain semi-confluent cultures. Stock solutions of Geran-iol, Limonene and Benzylic Alcohol in dimethyl sulfoxide (DMSO) were added to theculture medium at the indicated concentrations (final vehicle concentration <0.5%).Control cultures received the same vehicle concentration. For the Western blottinganalysis the cell cultures were seeded in 75 cm2 flasks.

2.4. MTT assay

Tetrazolium salt (MTT) was dissolved in PBS (5 mg/ml) and added to the cells(100 �l/ml medium without serum and phenol red) according to Mosman (1983).After a 3-h incubation at 37 ◦C, a solution of 1N hydrogen chloride–isopropanol(1:24, v/v) was added to each well, followed by mixing to dissolve the dark-blueformazan crystals formed. After a few minutes at room temperature, the plates wereread at 550 nm in a Uniskan II Microplate reader (Labsystem, Helsinki, Finland).

2.5. Neutral red uptake (NRU) assay

The cells were exposed to a medium (without serum and phenol red) containing50 �g/ml neutral red dye and processed according to the method of Borenfreund etal. (1988). After a 3-h incubation at 37 ◦C, the cells were washed twice in PBS andfixed according to the procedure described by Riddell et al. (1986). The plates werethen left at room temperature for 10 min and the absorbance of the released neutralred dye was read at 550 nm.

2.6. mRNA isolation and polymerase chain reaction with reverse transcription(RT-PCR)

Total RNA from NCTC 2544 was harvested in TRIzol reagent according tothe manufacturer’s instructions (Invitrogen), treated with DNase I (Promega),and reverse transcribed using oligo (dT) and reverse transcriptase (PromegaCorporation, Milan, Italy). Gene-specific primers for Il-1�, LIF and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, housekeeping gene) were designed accordingto the sequences available at the National Center for Biotechnology Information(http://www.ncbi.nlm.nih.gov/) in order to amplify specific cDNA sequences andavoid genomic DNA amplification. Primers were synthesized according to the pub-lished cDNA sequences, and were provided by TIB (TIB MOLBIOL, ABC, Genoa,Italy). For human IL-1� the forward primer is 5′-TGGCACAATCTCGGCTCACTG,and the reverse is 5′-CTAGGTGAACATGCGTTATTGC; for LIF the primers are 5′-TGGACGTGACCTACGGCCCTG and 5′-TGCAGGTCCAGCCATCAGACC, and for GAPDH5′-TGAAGGTCGGAGTCAACGGATTTGGT and 5′-CATGTGGGCCATGAGGTGCACCAC).

Equal amounts of total RNA were taken and reverse transcribed with GoTaq PCRCore Systems (Promega). PCR products were subjected to electrophoresis on a 2%agarose TBE gel and DNA was visualized by ethidium bromide staining under UV.Digital images of PCR products were analysed by means of BIORAD Geldoc 2000 (Bio-Rad Laboratories, Milan, Italy) and the bands quantified by Quantity One software(Bio-Rad Laboratories) and normalized to its own GAPDH.

The results were expressed as percentage of test cells versus control.

Author's personal copy

A. Parodi et al. / Toxicology Letters 192 (2010) 101–107 103

2.7. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis

SDS-PAGE was done under reducing conditions with a MiniPROTEAN® 3 Cell(Bio-Rad), using a 4% stacking gel and a 12% separating gel. Untreated and treatedcells were harvested by scraping into ice-cold PBS and centrifuging at 1000 × g for5 min at 4 ◦C. Cellular lysates, obtained by sonication at 50 W for 7 s, were denaturedin sample buffer for SDS-PAGE (Laemmli, 1970). All samples were heated for 5 min inboiling water, cooled to room temperature, and subjected to electrophoresis (20 �gof protein per well) at a constant voltage of 120 V until the dye front reached thebottom of the gel. At the end of the run the proteins were transferred to Hybond ECLmembrane, and processed using an ECL–Plus kit.

2.8. Western blotting

Rabbit anti-human IL-1� and anti-human LIF antibody, diluted at 1:1000, wereused as primary antibodies to detect the presence of the proteins, while horseradishperoxidase-labelled anti-rabbit IgG antibodies (GE, Healthcare), diluted 1:20,000,were used as secondary antibodies. Polyclonal anti-GAPDH antibody was used asprimary antibody to control for the amount of loaded protein.

The proteins were detected by chemiluminescence (ECL advance, Amersham-Pharmacia-Biotech), exposure to film and analysed using a BIORAD Geldoc 2000. Thedata presented were calculated after normalization with GAPDH. Densitometricaldata obtained from Quantity One software (Bio-Rad) were applied for statisticalanalysis.

2.9. Indirect immunofluorescence

Indirect immunofluorescence staining was used to detect LIF expression in NCTC2544 in basal cultures and in 3 h exposure to 0.5 mM Geraniol.

NCTC cells were seeded 24 h before treatment on 13 mm diameter slides locatedin 12-well plates at 5 × 104 cells/well. The culture were exposed or not to 0.5 mMGeraniol for the last 3 h and then were rinsed with phosphate buffered saline (PBS)pH 7.4 fixed for 30 min in 4% (w/v) p-formaldehyde in PBS at room temperature,quenched with 30 mM NH4Cl for 30 min, and permeabilised with 0.1% Triton X-100 for 15 min at room temperature. Incubation with primary monoclonal anti-LIFantibody was carried out for 1 h at room temperature, in a humid chamber. A slidewithout the primary antibody was used as a negative control.

After thoroughly rinsing for three times in PBS/Triton X-100 (0.1%), the cellswere treated with FITC-labelled secondary goat anti-mouse antibody, and incubatedfor 1 h at room temperature. After extensive washing in PBS/Triton X-100 (0.1%) toremove excess secondary antibody, slides were mounted with Gelvatol according toLennet (1978), and protected by coverslips. The stained cells were observed undera Leica inverted epi-fluorescence microscope fitted with a Leica DC digital camera(Leica Microsystems Ltd., Wetzllar, D). Digital images were captured using IM 500software.

2.10. Statistics

All experiments were carried out in at least triplicate, using three wells for eachsample. The data were analysed using the Instat and Prism software package (Graph-Pad Software, Inc). The effect concentrations, IC50, and their 95% confidence intervalswere determined by using a downhill logistic dose–response curve.

3. Results

As first analysis, in order to have a comparative evaluation ofthe toxicity potential of each chemical compound, dose–responsecurves were extrapolated from NRU and MTT viability indexesobtained from NCTC 2544 cell cultures after 3 h exposure to increas-ing concentrations ranging between 0.5 and 18 mM of Geraniol,Limonene, Benzylic Alcohol, and SDS, used as positive control(Fig. 1).

In order to exclude possible effects induced by the solvent DMSOin the experiments with the selected three fragrances, the toxicityof DMSO was determined in parallel, finding a minimum effectivetoxicity (IC05) of 1.4%, i.e., from about 3- to 15-folds higher than theconcentrations of the vehicle used in our experiments (data notshown). IC50 evaluated by the two viability texts are quite sim-ilar, even if for all fragrances the IC50s derived from MTT wereinvariably lower than those derived from NRU (Table 1), and thissuggests a more aggressive cytotoxicity of the fragrances on cellmitochondrial compartment.

The ranking of IC50s shows that the positive control, SDS,resulted in the most cytotoxic compound, followed by Geraniol andLimonene. For Benzylic Alcohol it was not possible to determine a

Fig. 1. Concentration–response curves obtained by plotting the percent viability ofNCTC 2544 cells exposed to Geraniol, Limonene, Benzylic Alcohol and SDS. Valuesare extrapolated by NRU and MTT viability index (panels A and B, respectively) afterexposure of cultures to the experimental treatments for 3 h. Data are expressedas mean ± SD of three independent experiments, done in triplicate. Concentrationsare normalized as the logarithm of concentration. For each curve, the top has beenset at 100% and the bottom at 0%. (�) Geraniol; (�) Limonene; (♦) Benzylic Alcohol;(©) SDS.

real value of IC50 since at maximum tested dose (18 mM), allowedby solvent concentration (0.1% DMSO), cell viability did not dropbelow 50% level versus untreated control.

Martinez et al. (2006) have recently reported that the humanundifferentiated keratinocyte cell line NCTC 2544 was able toexpress IL-1� but data about LIF were lacking, so we first verifiedgene expression of IL-1� and LIF in our in vitro model by assess-ing their mRNA expression levels. Our results demonstrated that inbasal conditions NCTC 2544 cells express both mRNA (Fig. 2).

During exposure of NCTC 2544 cells to experimental com-pounds, mRNA levels of both cytokines have not revealed anysignificant changes, probably due to the short-time of experimentaldesign.

Therefore we have analysed only the intracellular protein lev-els of IL-1� and LIF by western blotting during 1–6 h exposure toaromatic cosmetic ingredients at sub-toxic doses (viability index>85%). In parallel we also have exposed NCTC 2544 to SDS, forexcluding interferences with death effects (Fig. 3).

Three hours exposure to Geraniol resulted in a remarkablemodulation of the intracellular levels of both cytokines tested. Atthe lowest dose (0.25 mM) Geraniol increased IL-1� levels up to

Table 1Cytotoxicity of Geraniol, Limonene, Benzilyc acid and SDS on NCTC 2544 cells.

Compound Viability Test

NRU MTT

Geraniol 1521 ± 1.05 1265 ± 1.12Limonene 8623 ± 1.02 6657 ± 1.00Benzylic Alcohol n.d. n.d.SDS 233 ± 1.05 132 ± 1.18

Data are expressed as IC50 (�M), extrapolated from NRU and MTT viability index,and are referred to three separate experiments run in triplicate (see Section 2.10);n.d.: not determined.

Author's personal copy

104 A. Parodi et al. / Toxicology Letters 192 (2010) 101–107

Fig. 2. Analysis of LIF and IL-1� mRNA levels in NCTC 2544 cells by RT-PCR. Withselected specific primers for LIF and IL-1�, the amplification predicted fragments of886 and 498 bp, respectively, are shown in length.bp: base pairs; Line M: size marker; Lines 1 and 2: LIF and IL-1�, respectively.

50%, while at higher doses (0.5 and 1 mM) cytokine intracellularexpression dropped to basal levels. Intracellular LIF expression wasreduced to fairly constant levels (≤40% versus untreated cultures)by Geraniol, in all conditions.

After 1 and 2 mM Limonene treatment, IL-1� levels decreasedsignificantly, and, on the contrary, at highest dose (4 mM) cytokinelevels increased up to 20% respect basal levels. An opposite trendwas evidenced in LIF intracellular content: 1 mM Limonene induceda significant increase of LIF (up to 50%, versus untreated cultures),while higher concentrations of Limonene failed down cytokine lev-els to 55% of basal values.

NCTC cells exposed to Benzylic Alcohol resulted in a parallelincrease of both cytokine intracellular levels in a dose dependentway.

SDS treatment decreased IL-1� and LIF expressions and at thehighest dose the reached levels of both cytokines resulted quitesimilar, even if only LIF showed a more marked significant decrease.

To verify if both cytokines have a time-dependent modulation,we investigated IL-1� and LIF protein levels after 1–3–6 h exposure(Fig. 4) with Geraniol (0.5 mM), Limonene (2 mM), Benzylic Alcohol(2 mM) and SDS (0.05 mM).

During 1 and 3 h exposure to 0.5 mM Geraniol, NCTC cells evi-denced a stable amount of IL-1� followed by a dramatic reduction(60% versus basal levels) at the end of treatment. A 20% decrease

Fig. 3. IL-1 � and LIF protein levels in NCTC 2544 cell lines, during exposure to Geraniol, Limonene, Benzylic Alcohol and SDS.NCTC 2544 were exposed to 0.25, 0.5, 1 mM Geraniol (A); 1, 2, 4 mM Limonene (B); 1, 2, 4 mM Benzylic Alcohol (C) and 25, 50, 100 �M SDS (D) for 3 h. Each immunoblotsshown originates from one representative experiment among five independent experiments. Immunoreactive bands were quantified and the arbitrary units, percentageversus untreated cultures of IL-1� and LIF densitometric analysis (� and �, respectively), normalized to GAPDH levels, are expressed as mean ± SD of five separate experiments.The statistical significance was calculated using one-way ANOVA followed by Dunnett’s test. *P < 0.05, **P < 0.01, ***P < 0.001.

Author's personal copy

A. Parodi et al. / Toxicology Letters 192 (2010) 101–107 105

Fig. 4. Time-course of IL-1� and LIF protein levels in NCTC 2544 cell line, during exposure to Geraniol, Limonene, Benzylic Alcohol and SDS for 1, 3 and 6 h. NCTC 2544 wereexposed to 0.5 mM Geraniol, 2 mM Limonene, 2 mM Benzylic Alcohol and 50 �M SDS for 1 h (�), 3 h ( ) and 6 h (�).Each immunoblots shown originate from one representative experiment among five independent experiments. Immunoreactive bands were quantified and the arbitraryunits, percentage versus untreated cultures, of IL-1� and LIF densitometric analysis, normalized to GAPDH levels, are expressed as mean ± SD of five separate experiments.The statistical significance was calculated using one way ANOVA followed by Dunnett’s test. *P < 0.05, **P < 0.01, ***P < 0.001, versus untreated cell cultures.

of intracellular LIF content was detected just after 1 h exposureand resulted significant after 3 h, then the cytokine levels showeda slight recovery.

NCTC cells exposed for 1 h to 2 mM Limonene showed oppositeslight changes in both cytokines. After 3 h IL-1� intracellular con-tent dropped down dramatically and then recovered to basal levels;otherwise LIF content showed a marked reduction, by reaching 50%of basal levels.

Benzylic Alcohol induced a marked increase of both cytokinesin a time-dependent way, and this effect is more evident for IL-1� at the end of experimental treatment exposure. Exposure to50 �M SDS evidenced a time-dependent mild decrease in bothcytokines.

Indirect immunofluorescence analysis on localisation ofcytokines during exposure to irritating stimuli was limited to LIF,since this study is the first report on its expression in NCTC 2544cell line (Fig. 5).

For this survey cell cultures were exposed for 3 h to 0.5 mMGeraniol, because in this experimental condition the intracellu-lar levels of LIF evidenced a marked decrease, in the time-courseanalysis. Untreated NCTC 2544 cells (Fig. 5, Panel A) resultedpositive to LIF and the cytokine compartmentalization was inblunt-ended vesicles, localised around the nucleus and the con-tours of the cells. After 3 h Geraniol-treated cells (Panel B) showeda diffuse positivity of LIF through all the cytoplasm, and thenuclei resulted covered by a widespread fluorescence and thecounters of the cells resulted more approximate. These observa-tions could lead to speculate that LIF spreads also outside cellularbody because the fluorescence was detectable also outside thecells.

4. Discussion

Skin irritation testing has been performed now for decades withthe “Draize Rabbit skin irritation test” (Gibbs, 2009). Besides the

discomfort and pain to the animals the results do not always matchthe human response (York et al., 1996) and moreover, one of themajor criticisms to in vitro system is that cultured cells lose manymarkers.

Some compounds can induce irritation without affect cell via-bility, and, as well known, cytokines play a key role in initiating andmaintaining pathophysiology as well as in inducing recovery in skininflammations and IL-1� and TNF- � are considered to be primaryinitiators of cutaneous inflammation (Ullrich, 1995; Asadullah etal., 1999; Bonifati and Ameglio, 1999; Corsini and Galli, 2000).

IL-1� is a marker already used by some scientists to predict theirritation power of the chemicals tested (Sanchez et al., 2006).

Similarly to other cytokines, IL- 1� is temperature-sensitive andenvironment sensitive and this is to be considered an important fac-tor of variability in longer incubation periods under classical cultureconditions. LIF cytokine plays an important role in skin inflamma-tory process, and it was reported that it can act as pro- or/andanti-inflammatory mediator (Banner et al., 1998; Heinrich et al.,2003; Gadient and Patterson, 1999; McKenzie and Szepietowski,2004).

Improved understanding of cytokine–cytokine interactions hasled to a consensus that simultaneous assessment of many cytokinesin a biological sample provides more comprehensive informationrather than assessing a single cytokine.

The aim of this investigation was to enhance and highlight theimportance of LIF as an early marker of irritation produced bykeratinocytes. Moreover the expression of IL-1� seems to be con-ditioned by LIF secretion (Wilmer et al., 1994; Fouladi-Nashta et al.,2008; Villiger et al., 1992).

So we have exposed our in vitro model (normal undifferentiatedhuman keratinocyte cell line NCTC 2544) to common cosmetic fra-grances, suspected of causing irritation and sensibilization, and toSDS as positive control. In our study we evidenced for the first time,to our knowledge, the expression of LIF in NCTC 2544 cells. Fur-thermore we reported an intracellular modulation of both cytokine

Author's personal copy

106 A. Parodi et al. / Toxicology Letters 192 (2010) 101–107

Fig. 5. The immunohistochemical localisation of LIF has been performed by theindirect IF. (A) Control cultures made of unexposed cells. (B) Cells exposedto 0.5 mM Geraniol, for 3 h. A sharp fluorescence is seen in the cytoplasm ofuntreated cells, mostly polarised (arrow) at one side with respect to the nucleusposition. In Geraniol-treated cultures the fluorescence is less intense than incontrols, and localised all around the nuclei towards the cortical region of thecytoplasm. The fluorescence is present only in a small number of flattenedcells.

after short-time exposure (1–6 h) especially after Geraniol andLimonene treatments.

As well known, in keratinocytes this cytokine is character-ized by the accumulation of its immature 30 kDa form, whichundergoes proteolytic cleavage to be secreted in its mature 20 kDaform. So the observed biphasic trend of IL-1�, during Geraniol andLimonene exposure, could be ascribed to its early secretion andits prompt recovery through the cleavage of its immature form bycalpain (Werman et al., 2004). On the contrary, LIF intracellularlevels showed only a decrement; in fact, from the point of viewof proteomic, for this cytokine, only the intracellular accumula-tion of mature form has been demonstrated and, according to ourresults, a biphasic intracellular modulation after brief treatmentwas not justified. Moreover indirect immunofluorescence analysis,performed on Geraniol-treated cells, confirmed the hypothesis thatthe NCTC 2544 keratinocytes respond to an irritating stimulus by

leakage LIF from its storage and probably subsequent release of thecytokine.

After Benzylic Alcohol exposure we observed only an increas-ing phase of intracellular levels of both cytokines. This observationcould be due to proteolytic cleavage of IL-1� in its immature formand to the translation of the m-RNA of LIF that normally is storedin keratinocytes (Paglia et al., 1996; Szepietowski et al., 1997).Treatments with Benzylic Alcohol were probably not sufficientlyaggressive to induce the secretion of LIF, and only evidenced anincrease of the cytokine within the cells. Besides in order to main-tain a comparable concentration range for all the fragrances, weused doses of Benzylic Alcohol very far from the IC50 of this aromaticcompound. These findings emphasize the importance of classic tox-icity tests as first level screening before performing more specificanalysis on the molecular pathway of action of a tested compound.

Furthermore, only a combination of several end points will beuseful for a reliable prediction of irritancy by an in vitro toxicitytest. Our study stated to enhance and highlight the role of LIF, likeIL-1�, as useful early biomarker in an in vitro test for evaluatingthe irritation potential of cosmetic ingredients and other chemicalsthat could accidentally came into contact with skin.

The research for new biomarkers of irritation is essential toimprove the predictive power and the reproducibility of alterna-tive methods, even because the effect of a single cytokine mustbe interpreted according to the secretion of all cytokines secretedin response to external agents. LIF plays a key role in the dif-ferentiation of immature cells (e.g., stem cells) and is involvedin accelerated maturation of keratinocytes and, consequently, itmight be a parameter of skin aging. Therefore, the role of LIF inthe cosmetic field takes on greater significance, because LIF couldalso be considered a biomarker of skin aging, which is a target ofmany cosmetic products.

Conflict of interest statement

The authors declare that there are no conflicts of interest.

Acknowledgments

This study was supported by Equivita (www.equivita.org), Ani-malisti Italiani and Movimento Ecologico Nazionale UNA. Theauthors gratefully acknowledge the technical support by Dr. SarahMangini.

References

Asadullah, K., Döcke, W.-D., Volk, H.-D., Sterry, W., 1999. The pathophysiological roleof cytokines in psoriasis. Drugs Today 35, 913.

Banner, L.R., Patterson, P.H., Allchorne, A., Poole, S., Woolf, C.J., 1998. Leukemiainhibitory factor is an anti-inflammatory and analgesic cytokine. J. Neurosci.15; 18 (14), 5456–5462.

Baumann, H., Jahreis, G.P., Sauder, D.N., Koi, A., 1984. Human keratinocytes andmonocytes release factors which regulate the synthesis of major acute phaseplasma proteins in hepatic cell from man, rat and mouse. J. Biol. Chem. 259,7331–7342.

Bonifati, C., Ameglio, F., 1999. Cytokines in psoriasis. Int. J. Dermatol. 38, 241.Borenfreund, E., Babik, H., Martin-Alguacil, N., 1988. Comparison of two in vitro

cytotoxicity assays—the neutral red (NR) and tetrazolium MTT tests. Toxicol. InVitro 2, 1–6.

Botham, P.A., 2004. The validation of in vitro methods for skin irritation. Toxicol.Lett. 149, 387–390.

Bulger, R.E., 1987. Use of animals in experimental research: a scientist’s perspective.Anat. Rec. 219, 215–220.

Burlando, B., Parodi, A., Volante, A., Bassi, A.M., 2008. Comparison of the irritationpotentials of Boswellia serrata gum resin and of acetyl-11-keto-beta-boswellicacid by in vitro cytotoxicity tests on human skin-derived cell lines. Toxicol. Lett.177, 144–149.

Chen, M., 2005. Amended final report of the safety assessment of t-Butyl Alcohol asused in cosmetics. Int. J. Toxicol. 24 (Suppl. 2), 1–20.

Corsini, E., Galli, C., 2000. Epidermal cytokines in experimental contact dermatitis.Toxicol. 142, 203–211.

Author's personal copy

A. Parodi et al. / Toxicology Letters 192 (2010) 101–107 107

Cotovio, J., Grandidier, M.H., Portes, P., Roguet, R., Rubinstenn, G., 2005. The in vitroacute skin irritation of chemicals: optimisation of the EPISKIN prediction modelwithin the framework of the ECVAM validation process. Altern. Lab. Anim. 33,329–349.

Dinarello, C.A., 2000. Proinflammatory cytokine. Chest 118, 503–508.Eun, H.C., Suth, D.H., 2000. Comprehensive outlook of in vitro tests for assessing skin

irritancy as alternatives to Draize tests. J. Dermatol. Sci. 24, 77–91.Fouladi-Nashta, A.A., Mohamet, L., Heath, J.K., Kimber, S.J., 2008. Interleukin 1 sig-

naling is regulated by leukemia inhibitory factor (LIF) and is aberrant in Lif−/−

mouse uterus. Biol. Reprod. 79, 142–153.Gadient, R., Patterson, P.H., 1999. Leukemia inhibitory factor, interleukin 6, and

other cytokines using the GP130 transducing receptor: roles in inflammationand injury. Stem Cells 17, 127–137.

Gearing, D.P., 1993. The Leukaemia inhibitory and its receptor. Adv. Immunol. 53,51–58.

Gibbs, S., 2009. In vitro irritation models and immune reactions. Skin Pharmacol.Physiol. 22, 103–113.

Grandjean-Laquerriere, A., Antonicelli, F., Gangloff, S.C., Guenounou, M., Le Naour,R., 2007. UVB-induced IL-18 production in human keratinocyte cell line NCTC2544 through NF-kappaB activation. Cytokine 37, 76–83.

Grandjean-Laquerriere, A., Le Naour, R., Gangloff, S.C., Guenounou, M., 2003.Differential regulation of TNF-alpha, IL-6 and IL-10 in UVB-irradiatedhuman keratinocytes via cyclic AMP/protein kinase A pathway. Cytokine 23,138–149.

Grandjean-Laquerriere, A., Le Naour, R., Gangloff, S.C., Guenounou, M., 2005. Con-tribution of protein kinase A and protein kinase C pathways in ultravioletB-induced IL-8 expression by human keratinocytes. Cytokine 29, 197–207.

Heinrich, P.C., Behrmann, I., Haan, S., Hermanns, H.M., Müller-Newen, G., Schaper, F.,2003. Principles of interleukin (IL)-6-type cytokine signalling and its regulation.Biochem. J. 374 (Pt 1), 1–20.

Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the headbacteriophage T4. Nature 227, 680–685.

Lee, E., An, S., Choi, D., Moon, S., Chang, I., 2007. Comparison of objective andsensory skin irritations of several cosmetic preservatives. Contact Dermat. 56,131–136.

Lennet, E.D.A., 1978. An improved mounting medium for immunofluorescencemicroscopy. Am. J. Clin. Pathol. 69, 647–648.

Lotz, M., Moats, T., Villiger, P.M., 1992. Leukaemia inhibitory factor s expressed incartilage and synovium and can contribute to the pathogenesis of arthritis. J.Clin. Invest. 90, 888–896.

Martinez, V., Corsini, E., Mitjans, M., Pinazo, A., Vinardell, M.P., 2006. Evaluationof eye and skin irritation of arginine-derivative surfactants using differentin vitro endpoints as alternatives to the in vivo assays. Toxicol. Lett. 164,259–267.

McKenzie, R.C., Szepietowski, J.C., 2001. Leukaemia inhibitory factor (LIF) expressionin skin from amyotrophic lateral sclerosis patients compared with skin fromnormal individuals: what is the function of LIF in the skin? J. Invest. Dermatol.116, 476–478.

McKenzie, R.C., Szepietowski, J., 2004. Cutaneous leukemia inhibitory factor andits potential role in the development of skin tumors. Dermatol. Surg. 30,279–290.

Mori, M., Yamaguchi, K., Honda, S., Nagasaki, K., Veda, M., Abe, O., Abe, K., 1991.Cancer cachexia syndrome developed in nude mice bearing melanoma cellsproducing Leukaemia inhibitory factor. Cancer Res. 51, 6656–6659.

Mosman, T., 1983. Rapid colorimetric assay for cellular growth and survival:application to proliferation and cytotoxicity assays. J. Immunol. Methods 63,55–63.

Paglia, D., Kondo, S., Ng, K.M., Sauder, D.N., McKenzie, R.C., 1996. Leukaemiainhibitory factor is expressed by normal human keratinocytes in vitro and invivo. Br. J. Dermatol. 134, 817–823.

Riddell, R.J., Clothier, R.H., Balls, M., 1986. An evaluation of three in vitro cytotoxicityassays. Food Chem. Toxicol. 24, 469–471.

Russell, W.M., 1995. The development of the three Rs concept. Altern. Lab Anim. 23(3), 298–304.

Sanchez, L., Mitjans, M., Infante, M.R., Vinardell, M.P., 2004. Assessment of the poten-tial skin irritation of lysine derivate anionic surfactants using mouse fibroblastsand human keratinocytes as an alternative to animal testing. Pharm. Res. 21,1637–1641.

Sanchez, L., Mitjans, M., Infante, M.R., Vinardell, M.P., 2006. Determination ofinterleukin-1alpha in human NCTC 2544 keratinocyte cells as a predictor of skinirritation from lysine-based surfactants. Toxicol. Lett. 167, 40–46.

Spielmann, H., Hoffmann, S., Liebsch, M., Botham, P., Fentem, J., Eskes, C., Roguet,R., Cotovió, J., Cole, T., Worth, A., Heylings, J., Jones, P., Robles, C., Kandárová,H., Gamer, A., Remmele, M., Curren, R., Raabe, H., Cockshott, A., Gerner, I.,Zuang, V., 2007. The ECVAM international validation study on in vitro testsfor acute skin irritation: report on the validity of the EPISKIN and Epi-Derm Assays and on the Skin Integrity Function Test. Altern. Lab. Anim. 35,559–601.

Szepietowski, J.C., McKenzie, R.C., Keohane, S.G., Walker, C., Aldrgige, R.D., Hunter,J.A.A., 1997. Leukaemia inhibitory factor induction in early phase of allergiccontact dermatitis. Contact Dermat. 36, 21–25.

Tomida, M., Yamamoto-Yamaguchi, Y., Hozumi, M., 1984. Purification of a fac-tor inducing differentiation of mouse myeloid leukemic M1 cells fromconditioned medium of mouse fibroblasts L929 cells. J. Biol. Chem. 259,10978–10982.

Ullrich, S.E., 1995. The role of epidermal cytokines in the generation of cutaneousimmune-reactions and ultraviolet radiation-induced immune suppression. Pho-tochem. Photobiol. 62, 389.

Villiger, P.M., Geng, Y., Lotz, M., 1992. Induction of cytokine expression by leukaemiainhibitory factor. J. Clin. Invest. 90, 888–896.

Waring, P., Wycherley, K., Cary, D., Nicola, N., Metcalf, D., 1992. Leukaemia inhibitoryfactor levels are elevated in septic shock and various inflammatory body fluids.J. Clin. Invest. 90, 2031–2037.

Werman, A., Werman-Venkert, R., White, R., Lee, J.K., Werman, B., Krelin, Y.,Voronov, E., Dinarello, C.A., Apte, R.N., 2004. The precursor form of IL-1(alpha) is an intracrine proinflammatory activator of transcription. PNAS 101,2434–2439.

Wilmer, J.L., Burleson, F.G., Kayama, F., Kanno, J., Luster, M.I., 1994. Cytokine induc-tion in human epidermal keratinocytes exposed to contact irritants and itsrelation to chemical-induced inflammation in mouse skin. J. Invest. Dermatol.102, 915–922.

York, M., Griffiths, H.A., Whittle, E., Basketter, D.A., 1996. Evaluation of a humanpatch test for the identification and classification of skin irritation potential.Contact Dermat., 204–212.

Zhu, M., Oishi, K., Lee, S.C., Patterson, P.H., 2001. Studies using leukemia inhibitoryfactor (LIF) knockout mice and a LIF adenoviral vector demonstrate a key anti-inflammatory role for this cytokine in cutaneous inflammation. J. Immunol. 1(166), 2049–2054.