Analysis of oxidative stress in SK-N-MC neurons exposed to styrene-7,8-oxide

Toxicology in Vitro 19 (2005) 11–20

www.elsevier.com/locate/toxinvit

Analysis of oxidative stress in SK-N-MC neuronsexposed to styrene-7,8-oxide

M.V. Vettori a,*, A. Caglieri b, M. Goldoni a,b, A.F. Castoldi c,E. Dar�e d, R. Alinovi b, S. Ceccatelli d, A. Mutti b

a ISPESL Research Center at the University of Parma, via Gramsci 14, Parma 43100, Italyb Laboratory of Industrial Toxicology, Department of Clinical Medicine, Nephrology and Health Sciences, University of Parma, Parma, Italy

c Toxicology Division, Research Center, Salvatore Maugeri Foundation IRCCS, Institute of Pavia, Italyd Institute of Environmental Medicine, Division of Toxicology and Neurotoxicology, Karolinska Institutet, Box 210, Stockholm S-171 77, Sweden

Received 15 December 2003; accepted 28 April 2004

Abstract

Styrene-7,8-oxide (SO) is the main metabolite of styrene, a neurotoxic volatile organic compound used industrially. Here we

report the novel observation that several markers of oxidative stress were affected in SK-N-MC cells exposed for 16 h to con-

centrations of SO that induce apoptotic cell death. The production of Thiobarbituric Acid Reactive Substances (TBARS), rose from

69.1± 15.7 nmol/g protein (control) to 119.3± 39.2 and 102.0± 17.3 nmol/g protein after exposure to 0.3 and 1 mM SO, respectively.

Carbonyl group levels were significantly enhanced by SO at both concentrations. The lower dose also decreased sulphydryl groups.

SO caused a marked oxidative DNA damage, as shown by a fivefold increase in 8-hydroxy-20-deoxyguanosine (8-OHdG). In

addition, SO exposure resulted in alterations of scavenging abilities, given the reduction of both glutathione (GSH) and glutathione-

S-transferase (GST) activity. Induction of expression of the oxidative stress response gene heme-oxygenase-1 (HO-1) and an in-

creased HO-1 activity were also observed. These data provide compelling evidence that oxidative stress significantly contributes to

SO toxicity in neuronal cells.

� 2004 Elsevier Ltd. All rights reserved.

Keywords: Oxidative stress; Styrene oxide; Neurotoxicity; SK-N-MC neurons

1. Introduction

Styrene-7,8-oxide (SO) is an industrial chemical used

in the manufacture of epoxy resins, as an intermediate in

the preparation of various agricultural chemicals, cos-

metics, and plastics and in the processing of textiles andfibers. SO is also the main electrophilic intermediate

metabolite of styrene, a chemical used in several indus-

trial applications, e.g. manufacture of glass fiber rein-

Abbreviations: 1-Chloro-2,4-dinitrobenzene (CDNB); Dinitro-

phenylhydrazine (DNPH); Fetal bovine serum (FBS); Glutathione

(GSH); Glutathione-S-transferase (GST); Heme-oxygenase-1 (HO-1);

8-Hydroxy-20-deoxyguanosine (8-OHdG); Reactive oxygen species

(ROS); Sodium dodecyl sulfate (SDS); Styrene-7,8-oxide (SO); Thio-

barbituric acid (TBA)*Corresponding author. Tel.: +39-0521-033092; fax: +39-0521-

033076.

E-mail address: [email protected] (M.V. Vettori).

0887-2333/$ - see front matter � 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.tiv.2004.04.015

forced plastics and production of styrene monomer,

polyester resins and synthetic rubber. As an epoxide, SO

is able to undergo reactions with various nucleophilic

groups in tissue components, causing effects such as

growth inhibition, cytotoxicity and mutagenicity. Fur-

thermore, the most probable risk of SO is related to itsgenotoxicity. In particular, SO has been shown to

induce DNA adducts, chromosomal aberrations, sister

chromatide exchanges (SCE) micronuclei and DNA

damage in several in vitro systems (Vodicka et al., 2002;

Laffon et al., 2001, 2002). SO is a proven animal car-

cinogen and is classified as a possible human carcinogen

(group 2B) by IARC (1994). The role of SO in the

neurotoxic effects reported in workers exposed to sty-rene is still controversial. Alternative explanations to

styrene neurotoxicity have been put forward by our

group years ago. Brain dopamine depletion has been

reported upon exposure to styrene or styrene metabo-

lites in experimental studies (Mutti et al., 1984b, 1988),

12 M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20

and the dopaminergic system has been suggested as a

specific target for styrene neurotoxicity both in humans

(Mutti et al., 1984a; Arfini et al., 1987; Mutti and

Franchini, 1987) and laboratory animals (Mutti et al.,1984b, 1988). SO at high doses has been shown to be

neurotoxic in animals and using in vitro models (Katoh

et al., 1989; Trenga et al., 1991; Kohn et al., 1995; Bei-

swanger et al., 1993; Chakrabarti, 1999) and the mech-

anisms by which it acts are still unclear. SO exposure

induces depletion of cells in rat brain and is cytotoxic to

neurons in vitro (Kohn et al., 1995; Dypbukt et al.,

1992). In a recent work (Dar�e et al., 2002) we haveshown that exposure of primary cerebellar granule

neurons and human neuroblastoma SK-N-MC cells to

SO (0.3–1 mM) induces apoptosis. Cell shrinkage,

chromatin condensation and DNA cleavage into high

molecular weight fragments of regular size were

accompanied by the activation of class II caspases. In

addition, the presence of the 150-kDa cleavage product

of alpha-fodrin suggested a possible activation of cal-pains. Apoptosis can be triggered by oxidative stress

(Fleury et al., 2002). The ability to increase the synthesis

of reactive oxygen species (ROS) is a common factor in

the toxicity of several neurotoxic compounds including

metals (Naarala et al., 1995; Flora and Seth, 2000;

Olivieri et al., 2002; Belletti et al., 2002; Migheli et al.,

1999) and solvents (Nordmann, 1987; Uysal et al.,

1989). Several lines of evidence suggest that styrene andits metabolite SO can induce oxidative stress. Depletion

of GSH, the primary cellular defense against oxidative

stress, was observed in different brain regions of rats

exposed to SO (Trenga et al., 1991). Lower levels of

GSH were found to be associated with increased lipid

peroxidation in the liver of rodents exposed subchroni-

cally either to styrene or to SO (Katoh et al., 1989). In

addition, increased levels of DNA oxidation productswere detected in blood cells exposed to styrene oxide

in vitro (Marczynski et al., 1997).

In the present study, we have further explored the

mechanisms of SO toxicity and tested, evaluating the

hypothesis that SO increases the levels of ROS in neu-

rons. To clarify intracellular events involved in neuronal

cells death, we have used the human neuroblastoma SK-

N-MC cells. This cell line is a well established experi-mental model to study mechanisms of neurotoxicity in

vitro (Hyun et al., 2002; Dar�e et al., 2002; Sherer et al.,2003). Since apoptotic chromatin rearrangements and

increased caspase activity were observed at 0.3 and 1

mM SO in SK-N-MC cells (Dar�e et al., 2002), theseconcentrations were chosen here to examine different

markers of oxidative stress. Cellular responses have been

studied at the 16 h time point, based on initial timecourse experiments (Dar�e et al., in press). The followingmarkers have been analyzed: (1) Thiobarbituric Acid

Reactive Substances (TBARS) as indicators of oxidative

damage to cell membranes; (2) protein sulphydryl

groups and carbonyl groups as markers of protein oxi-

dation; (3) 8-hydroxy-20-deoxyguanosine (8-OHdG) as

an index of DNA oxidation; (4) expression and activity

of heme-oxigenase-1 (HO-1) as a stress response tooxidative damage mediated by gene regulation. Experi-

ments were also performed to assess the scavenging

potential of cells (GSH intracellular levels and gluta-

thione-S-transferase (GST) activity) as an indirect

mechanism possibly accounting for oxidative stress

associated with SO exposure.

2. Materials and methods

2.1. Chemicals

1,1,3,3-Tetraethoxypropane (Malondialdehyde) was

obtained from Fluka Chemie (Buchs, CH-9471), while

Bioxytech GSH-400 Assay Kit from Oxis International,

Inc. (Portland, OR, USA). DNA isolation Kit was

bought from Gentra System (Minneapolis, MN, USA).Styrene-7,8-oxide (97%) was purchased from Sigma–

Aldrich (Stockholm, Sweden). Dinitrophenylhydrazine

(DNPH) and n-butanol were obtained from Aldrich

(Steinheim, Germany and Milwakee, USA, respec-

tively). Trichloroacetic acid and ethanol were bought

from Analar-BDH (England) and from Carlo Erba

(Milano, Italy), respectively. Ethyl acetate was bought

from Lab-Scan (Dublin, Ireland). Sodium dodecyl sul-fate (SDS) was obtained from Gibco (Paisley, Scotland).

The polyclonal anti-HO-1 antibody was obtained from

Stressgen (Victoria, BC, Canada), the goat antirabbit

peroxidase conjugated antibody was from Pierce

(Rockford, USA), and 30,30-diaminobenzidine substrate

was from Boehringer Mannheim Biochimica (Milan,

Italy).

All the other chemicals were obtained from SigmaChemical Co. (St Louis, MO, USA).

2.2. Cell cultures

The human neuroblastoma SK-N-MC clonal cell line

was purchased from the American Type Culture Col-

lection. Cells were routinely seeded at the density of

40,000 cells/cm2 in DMEM containing 10% fetal bovine

serum (FBS) and 50 units/ml of penicillin and 50 lg/mlof streptomycin. Cells were maintained at 37�C in a 5%CO2 humidified incubator and sub-cultured twice a

week. Treatments with SO were carried out using dilu-tions in medium without serum of a stock solution

prepared in DMSO (final DMSO concentration in the

cell exposure medium¼ 0.05%) and were started 24 hafter seeding, the density of seeding being always 4 · 104cells/cm2. For harvesting, the floating cells, collected

from the medium by centrifugation, were pooled

together with the cells detached by scraping or with

M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20 13

trypsin. In all experiments, the protein concentration

was determined by the BCA method (Pierce, Rockford,

USA), using bovine serum albumin as a standard. Mean

values of SO within the incubation period were extrap-olated from a time-dependent curve, similarly to the

method described in Dar�e et al. (2002) but using theexperimental conditions described above. Briefly, after

0, 8, 16 and 24 h of incubation, 1 ml of supernatant was

placed in 4 ml SPME vials to measure SO concentration

in medium by SPME-GC/MS (Poli et al., 2004). For

each exposure time, samples were collected in triplicate.

Styrene-d5 was added to samples as Internal Standard (2ll of the stock solution, final concentration 2 · 10�6 M).Samples were stored at )80 �C until analysis.

2.3. Detection of TBARS

Cellular TBARS were measured according to the

method of Jentzsch et al. (1996), adapted for cells.Briefly, after three cycles of freezing and thawing ()80�C to +37 �C), control and treated cells were centrifugedat 3000g for 5 min. Two hundred ll of supernatant werediluted with 200 ll of 0.2 M orthophosphoric acid. After

vortexing, 25 ll of a 0.11 M thiobarbituric acid (TBA)

solution prepared in 0.1 M NaOH were added and the

vortexed solution was incubated at 95 �C for 45 min.TBARS were extracted adding 500 ll of n-butanol and50 ll of a saturated solution of NaCl. After a vigorousmixing, the reaction mixture was centrifuged at 3000g

for 10 min and the upper solution was collected. TBARS

concentrations were measured using the difference in

absorption between 530 and 650 nm (the zero sig-

nal). Malondialdehyde was used as a standard for

the calibration curve (range 0–10 lM). TBARSconcentrations were normalized for the total proteincontent.

2.4. Quantification of protein carbonyl groups

Cellular carbonyl groups of proteins were measured

according to the method of Reznick and Packer (1994)adapted for cells. Briefly, after three cycles of freezing

and thawing ()80 �C to +37 �C), control and treatedcells were centrifuged at 3000g for 5 min. DNA was

precipitated from the supernatant adding an amount of

a 10% (w/v) solution of streptomycin sulphate prepared

in PBS, equal to 1/10 of the volume. After 10 min at

room temperature, the supernatant was centrifuged at

3000g for 5 min and the pellet was removed. Twohundred and fifty ll of the supernatant were added to600 ll of a solution of 15 mM DNPH prepared in 2.5 M

HCl. The mixture was left at room temperature for 60

min in the dark, and vortexed every 15 min. Six hundred

ll of a 20% (w/v) trichloroacetic acid solution were

added to precipitate proteins. After 10 min, samples

were centrifuged at 5000g for 5 min and the supernatant

was discarded. This procedure was repeated twice.

Finally, the pellet was washed three times with 1 ml of

an ethanol–ethyl acetate solution (1:1, v/v) to remove

the residue of free DNPH and lipid contaminants. Theprecipitates were dissolved in 1 ml of an 8 M guanidine

hydrochloride solution and were left at 37 � C for 10min. No insoluble material was observed thereafter.

Carbonyl content was calculated by the peak absor-

bance at 365 nm, subtracting the zero signal, using an

extinction coefficient of 22,000 M�1 cm�1. The concen-

tration of carbonyls was normalized for the total protein

content.

2.5. Measurement of protein sulphydryl groups

Cellular protein sulphydryl groups were measured

according to the method of Inayama et al. (1996) with

some modifications. Briefly, after three cycles of freezing

and thawing ()80 �C to +37 �C), control and treatedcells were centrifuged at 3000g for 5 min. Two hundred

ll of supernatant were diluted with 600 ll of water andultra-filtered with a 5-kDa cut-off filter at 4000g for 15

min. The procedure was repeated three times to elimi-

nate free sulphydryl groups. After centrifugation, no

protein aggregation was observed. The upper solution in

the filter, corrected with water to restore the initial

volume of 200 ll, was mixed with 475 ll of a solutioncontaining 0.12 M Tris–HCl, 15 mM EDTA, 3% SDS

and 1 mM DTNB and incubated for 15 min at room

temperature. Sulphydryl content was calculated by the

peak absorbance at 412 nm, subtracting the zero signal.

LL-cysteine was used as a standard for the calibration

curve (range 0–200 lM). The concentration of sul-phydryl groups was normalized for the total non-filtered

protein content.

2.6. Quantification of 8-OHdG in DNA samples

DNA was extracted from cells exposed to SO and

controls using a commercial kit (Gentra System). The

DNA samples were digested with nuclease P1 for 30 minat 37 �C, and then incubated with alkaline phosphatasefor 60 min. To measure hydroxyl adducts to DNA, the

amount of 8-OHdG present in the DNA was measured

by HPLC with electrochemical detector as previously

described by Kim and Lee (1997).

2.7. Evaluation of HO-1 activity

HO-1 activity was measured by bilirubin generation.

The cells were exposed to SO for 16 h, then collected

from the flasks and centrifuged at 1000g at 4 �C for 10min. The pellet was resuspended in 500 ll of phosphatebuffer (100 mM KH2 PO4 and 2 mM MgCl2, pH 7.4)

and frozen–thawed ()80 to +37 �C) three times. After

Fig. 1. Time-course of SO concentration in the medium during

exposure of SK-N-MC cells. Cells were treated with the initial con-

centration of 0.3 mM SO, and SO was measured in aliquots of the

medium at different time points (8, 16 and 24 h). Values are the

means±S.D. of three determinations. (j¼ experimental data points;continuous line¼fitting curve; dotted line¼fitting curve reported byDar�e et al., 2002).

14 M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20

sonication on ice and centrifugation at 18,000g for 10

min at 4 �C, the reaction was started by adding thesupernatant to the NADPH-generating system con-

taining 0.8 mM NADPH, 2 mM glucose-6-phosphate,0.2 U glucose-6-phosphate dehydrogenase, 3 mg protein

of rat liver cytosol (prepared from 105,000g supernatant

fraction as source of biliverdin reductase), potassium

phosphate buffer (100 mM, pH 7.4), and hemin (20 lM)in a final volume of 1 ml. The reaction was incubated for

1 h at 37 �C in the dark and was terminated by additionof 1 ml chloroform. The solution was vortexed for 30 s,

centrifuged for 5 min at 4000g and the lower phase wascollected. The extracted bilirubin was calculated by the

difference in OD units recorded at 464 and 530 nm using

a quartz cuvette (extinction coefficient, 40 mM�1 cm�1

for bilirubin). HO-1 activity was expressed as picomol of

bilirubin/mg of protein/h.

2.8. Detection of HO-1 by immunocytochemistry

SK-N-MC cells grown on coverslips were exposed to

SO, then fixed with 80% methanol at 20 �C for 10 minand with acetone at 20 �C for 10 s. Since the 1 mM dose

induced cell detachment from the glass slide, only sam-

ples exposed to 0.3 mM SO and controls were analyzed

by immunocytochemistry. After three washes with PBS,

inhibition of endogenous peroxidases (1% H2O2 for 10min) and blocking with FBS (30 min at room tempera-

ture), the coverslips were incubated with a rabbit anti-

HO-1 polyclonal antibody (1:500) at 37 �C for 45 min.The cells were then rinsed with PBS and incubated for

1 h at room temperature with a goat peroxidase conju-

gated secondary antibody (1:1000). After washing with

PBS, the coverslips were incubated with 30,30-diam-

inobenzidine substrate for 15 min at room temperature,rinsed in water and then counterstained with methyl-

green (6 min at room temperature). The cells were then

examined at the light microscope (Olympus) and images

were collected with a digital camera.

2.9. Detection of intracellular glutathione-S-transferase

(GST) activity

For GST activity assays the cells were collected by

centrifugation (1100 rpm for 10 min at 4 �C). The pelletswere sonicated in cold phosphate buffered saline and

then centrifuged at 10,000g for 15 min at 4 �C. Thesupernatants were transferred into new tubes and stored

on ice. GST activity was determined in 100 mM potas-sium phosphate buffer pH 6.5 by measuring the conju-

gation of 1-chloro-2,4-dinitrobenzene (CDNB) with

reduced glutathione, which was accompanied by an

increase in absorbance at 340 nm (Habig et al., 1974).

The rate of product formation was monitored by mea-

suring the increase in absorbance at 340 nm in a Beck-

man DU 640 UV/visible spectrophotometer for 10 min.

GST activity was expressed as mU/mg protein. The di-

rect effect of SO on pure GST enzyme from human

placenta (Sigma–Aldrich, St Louis, MO, USA) was also

investigated. The reaction mixture was prepared byadding 30 lg of pure GST enzyme to different solutionsof SO (0.3 and 1 mM) in a 100 mM phosphate buffer pH

6.5, and incubated at 30 �C for different periods of time(1, 2, 4, 8 and 16 h). At the end of the incubation an

aliquot of the mixture was analyzed to determine the

GST activity, using the above-described method.

2.10. Measurement of intracellular GSH

Intracellular levels of GSH were determined in freshsamples using a commercial colorimetric kit (Oxis

International, Portland, USA). Briefly, at the end of

the exposure period SK-N-MC cells were pelleted by

centrifugation at 1200 rpm for 5 min at 4 �C, and thenwashed in 5% cold metaphosphoric acid; the samples

were centrifuged at 4500g for 10 min at 4 �C and

the supernatants were immediately used to measure the

GSH content according to the protocol indicated in thekit. The sample GSH concentration was extrapolated

from a freshly prepared standard curve (0–60 lM) andnormalized to the protein content.

3. Results

3.1. Measurement of the SO concentration in the

medium

Fig. 1 (continuous line) shows the time-decay of 0.3

mM SO in the medium containing SK-N-MC cells. The

M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20 15

1 mM dose showed a similar trend (data not shown).

Using the same algorithm for the decay of SO given by

Dar�e et al. (2002) (Fig. 1, dotted line), the nominalconcentrations of 0.3 and 1 mM SO were shown tocorrespond to average concentrations of 0.15 and 0.51

mM, respectively (during the 16 h of incubation). To

simplify the presentation of the results, the concentra-

tions of SO given throughout this paper refer to the

initial levels in the medium.

3.2. Lipid peroxidation in SO-exposed neurons

To evaluate the effect of SO on lipid peroxidation the

concentration of TBARS was measured in lysates of

SK-N-MC cells exposed to 0.3 or 1 mM SO for 16 h.

Both doses caused a statistically significant enhancementof MDA levels relatively to control values (Fig. 2A).

The production of TBARS increased from 69.08± 15.7

nmol/g protein in control samples to 119.32 ± 39.25 and

102.05 ± 17.27 in samples exposed to 0.3 and 1 mM SO,

respectively.

3.3. Protein modifications induced by SO

As a measure of protein oxidation, the levels of both

carbonyl and sulphydryl groups were measured in SK-

Fig. 2. Effects of SO on TBARS (A), 8-OHdG levels (B), carbonyl group

means±S.D. of at least four determinations. The concentrations of SO given

medium and correspond to an extrapolated average of SO concentration o

performed by one-way ANOVA, followed by the post-hoc Dunnett’s test. � S

different from untreated control (p < 0:01); ��� Significantly different from u

N-MC cell lysates following 16 h exposure to SO. Car-

bonyl groups were significantly increased in cells ex-

posed to either 0.3 or 1 mM SO, as compared with

control levels (Fig. 2C) (4.71 ± 0.46 nmol/mg protein at0.3 mM and 4.73± 0.48 nmol/mg protein at 1 mM, vs.

4.17 ± 0.12 nmol/mg protein of controls). With respect

to protein sulphydryl groups, the latter were decreased

by about 26% in cell lysates treated with 0.3 mM

SO (Fig. 2D). In contrast, no significant changes in the

level of sulphydryl groups were observed in proteins

extracted from cells exposed to 1 mM SO for 16 h

(Fig. 2D).

3.4. Effects of SO on DNA oxidation

Measurements of hydroxyl adducts to DNA (8-

OHdG) were performed to evaluate SO-induced DNA

oxidation. Both 0.3 and 1 mM SO caused a marked

oxidative DNA damage after 16 h, as indicated by the

more than fivefold increase in 8-OHdG levels in SO-exposed cell lysates as compared to controls (Fig. 2B).

In basal conditions samples contained 0.6 8-OHdG/105

dG. Upon treatment with 0.3 and 1 mM SO, the levels

of 8-OHdG/105 dG were enhanced to 3.5 ± 0.5 and

3.75± 0.8, respectively, as compared to untreated con-

trols (0.6 ± 0.4).

s (C), and sulphydryl groups (D) in SK-N-MC cells. Values are the

in all the figures refer to the initial levels of SO (0.3 and 1 mM) in the

f 0.15 and 0.51 mM within 16 h, respectively. Statistical analysis was

ignificantly different from untreated control (p < 0:05); �� Significantly

ntreated control ðp < 0:001Þ.

Fig. 3. HO-1 activity (A) and expression (B and C) in SK-N-MC neurons exposed to SO for 16 h. Values are the means±S.D. of at least four

determinations. The concentrations of SO given in all the figures refer to the initial levels of SO (0.3 and 1 mM) in the medium and correspond to an

extrapolated average of SO concentration of 0.15 and 0.51 mM within 16 h, respectively. Statistical analysis was performed by one-way ANOVA,

followed by the post-hoc Dunnett’s test. (��� Significantly different from untreated control, p < 0:001). The expression of HO-1 was evaluated in

neurons exposed to 0.3 mM for 16 h (B) and in unexposed control cells (C) using immunocytochemistry (bar is 10 lm).

16 M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20

3.5. SO-induced changes in HO-1 activity and

expression

HO-1 activity was measured in SK-N-MC cells ly-

sates by bilirubin generation. A significant increase ðp <0:001Þ in HO-1 activity was seen in cells exposed to 1mM SO (Fig. 3A). HO-1 activity showed an increasing

trend also in cells treated with 0.3 mM SO, although

such changes did not reach statistical significance. Be-

sides, this lower dose of SO increased the expression

HO-1 in comparison with control cultures, as shown byimmunocytochemical staining (Fig. 3B and C). Since the

1 mM dose induced cell detachment from the glass slide,

these samples were not analyzed for this end-point.

3.6. Alterations of GST activity and GSH levels

in SO-exposed cells

A dose-dependent decrease in the activity of GST was

detected in SK-N-MC cell lysates exposed to 0.3 and 1

mM SO for 16 h. In particular, the extent to which the

enzyme activity was inhibited by SO increased from

about 54% upon exposure to 0.3 mM to approximately

70% upon treatment with 1 mM (Table 1). To assesswhether the inhibition of GST activity by SO exposure

occurred through a direct mechanism, we studied the

effects of SO on GST enzymes purified from human

Table 1

GST activity and GSH levels in SK-N-MC cells exposed to either 0.3

or 1 mM initial concentrations of SO for 16 h

Control 0.3 mM SO 1 mM SO

GST (mU/mg prot)a 83.88± 13.71 38.38± 14� 25.46± 8.58�

GSH (nmoles/mg

prot)a16.11± 1.28 18.47± 5.21 7.66± 1.39�

The concentrations of SO given refer to the initial levels of SO (0.3 and

1 mM) in the medium and correspond to an extrapolated average of

SO concentration of 0.15 and 0.51 mM within 16 h, respectively.* Significantly different from untreated control ðp < 0:02Þ.a Values are the means of six determinations± S.D. Statistical

analysis was performed by one-way ANOVA.

placenta. The results of these measurements showed the

inability of SO to affect directly the GST enzyme activity

(data not shown).Additional experiments were performed to assess the

GSH level in cell lysates. Statistically significant de-

creases in GSH were observed in neurons exposed to

1 mM SO for 16 h as compared to controls (Table 1).

No changes were detected with the 0.3 mM dose after

16 h (Table 1). The same trend in GSH level was also

observed at earlier time points (after 8 h, data not

shown).

4. Discussion

Although oxidative stress has been shown to con-

tribute significantly to the noxious effects of many neu-

rotoxic agents, so far little attention has been paid tothe role of ROS in the mechanism of SO neurotoxicity

in vitro. In this study, we report the novel observation

that several markers of oxidative stress are affected by

SO exposure in neuronal cells. Since at 0.3 and 1 mM

apoptotic chromatin rearrangements and increased

caspase activity were observed (Dar�e et al., 2002), thesedoses were chosen to perform our study. Notably, here

we refer to initial concentrations prepared in the med-ium, although mean values of SO within the incubation

period of 16 h were 0.15 and 0.51 mM corresponding to

0.3 and 1 mM, respectively. Significant increases in lipid

peroxidation (TBARS content), protein and DNA oxi-

dation were observed in SK-N-MC cells exposed to ei-

ther 0.3 or 1 mM SO for 16 h. Both SO concentrations

markedly influenced the detoxifying capabilities of cells,

as they strongly decreased GST activity. The higher dosewas also effective in reducing the cellular GSH content.

In addition, SO induced HO-1 expression and increased

HO-1 activity. Taken together, these data suggest that

an excess of oxidative reactions is an important mech-

anism contributing to SO neurotoxicity in vitro and

point to a role of oxidative stress in the early phase of

SO-triggered apoptosis in neurons.

M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20 17

In the present study, a higher level of lipid peroxi-

dation was detected in neurons exposed to SO in vitro,

as compared to controls. The values measured in SK-N-

MC cells were in the same order of magnitude as thosereported in other studies on neurons (Naito et al., 1995).

The occurrence of lipid peroxidation has been demon-

strated in the liver but not in the brain of rats sub-

chronically exposed to styrene (300–500 mg/kg i.p., 3

times/week, 7 weeks) and styrene oxide (200–400 mg/kg)

(Katoh et al., 1989). These authors suggested that the

enhancement of hepatic lipid peroxidation resulted from

GSH depletion to certain critical levels and delayedrecovery of lipid peroxides (Katoh et al., 1989). Al-

though the same investigators also reported a decrease

in cerebral GSH levels in SO-exposed rats, they

hypothesized that such decrement was not critical en-

ough, in terms of degree and duration, to produce an

increase in brain lipid peroxidation (Katoh et al., 1989).

In our cell model, both 0.3 and 1 mM SO caused the

enhancement of lipid peroxidation at 16 h, but the lowerSO concentration did not affect GSH levels at the same

time point. The latter result may be ascribed to the lack

of SO effect on this target at the 0.3 mM dose. Although

the basal levels of GSH measured in our study are in

accordance with the ones reported in the literature

(Verity and Sarafian, 1991) no differences in GSH levels

were observed at 0.3 mM SO after 2, 4 and 8 h (data not

shown). The discrepancies with the in vivo data could bedue both to the different method of exposure and to SO

concentrations to which neuronal cells have been ex-

posed. The decrease in GST activity reflects a compro-

mised ability of the cell to scavenge electrophilic species.

GST is the enzyme that conjugates electrophilic species,

like SO, to GSH, and it is one of the essential cellular

antioxidant enzymes that can counteract the toxicity of

free radicals. Indeed, Xie et al. (2001) have demon-strated that over-expression of GST confers resistance

to oxidative stress in SY5Y neuroblastoma cells and that

GST activity decreases after exposure to 4-hydroxy-

nonenal, a compound formed during lipid peroxidation,

in a dose-dependent manner. Here we show that after

exposure to 1 mM SO for 16 h both GST activity and

GSH content were decreased. These concomitant alter-

ations were consistent with the presence of diffuse celldamage. In a previous study we have documented that 1

mM SO induced secondary necrosis in 57% of SK-N-

MC cells after 16 h (Dar�e et al., 2002). In agreementwith our findings, GSH depletion by 1 mM SO has been

previously reported in PC12 cells (Dypbukt et al., 1992).

A different pattern of changes has been observed in SK-

N-MC cells treated with 0.3 mM SO for 16 h, where

only 24% of the cells were necrotic (Dar�e et al., 2002). Inthis case GST activity was 50% lower than that of

controls in the absence of significant GSH changes. To

explain the reduced GST activity, we hypothesized a

direct action of SO on the catalytic site of the enzyme.

However, no changes were seen in the activity of GST

purified from human placenta in the presence of SO,

ruling out a direct interaction between the enzyme and

the toxic compound (data not shown). SO may insteadexert an indirect effect, increasing generation of ROS

that can cause enzyme modification. Moreover it is

possible a direct interaction of SO with GST genes and a

subsequent inactivation, SO being capable to form ad-

ducts with guanine and adenine (Vodicka et al., 2002).

The determination of carbonyl groups level has been

widely employed as a parameter of protein oxidation to

characterize tissue alterations occurring in neurodegen-erative diseases (Yatin et al., 1999; Aksenov et al., 2001;

Butterfield et al., 2002) and to unravel the mechanisms

of neurotoxicity in vitro (Keller et al., 2000; Kanski

et al., 2001; Choi et al., 2002). An increased level of

carbonyl groups has been related to a decrease in pro-

teasome activity in PC12 cells (Keller et al., 2000),

damage of mitochondrial proteins (Kim et al., 2001) and

to a general impairment of cellular functions (Deanet al., 1997). Under our experimental conditions, the

levels of carbonyl groups measured in SK-N-MC neu-

rons exposed to either 0.3 or 1 mM SO for 16 h are in

accordance with the values reported in the literature

(Lee et al., 2001) and were significantly increased as

compared to controls, providing evidence that SO

damages cellular proteins by oxidative reactions. In

SK-N-MC cells treated with 0.3 mM SO, a significantdecrease in protein sulphydryl groups was observed,

suggesting a higher number of oxidized cysteine and

methionine residues in the polypeptidic chains of SO

exposed neurons. Recent evidence indicates that the

oxidation of protein sulphydryl groups may result in

mitochondrial dysfunction, e.g. mitochondrial complex

I is highly vulnerable to inactivation through oxidation

(Sriram et al., 1998; Kenchappa et al., 2002). We foundthat SO at higher concentrations, namely 1 mM, was

devoid of significant effects toward the level of protein

sulphydryl groups in SK-N-MC cells. This difference in

protein oxidation observed at the two doses of SO after

16 h may relate to the fact that neurons have reached

different stages in the process of cell death, e.g. more

cells in secondary necrosis are detected at the higher SO

concentration (Dar�e et al., 2002). Disruption of disul-phide bridges and exposure of cysteine and methionine

residues, which are the measured in our assay, vary in

relation to the degree of protein denaturation and pro-

teolysis occurring in the cells.

The genotoxic potential of SO is well known and its

ability to induce DNA adducts and DNA strand breaks

both in vitro and in vivo has been reported (Bastlova

et al., 1995; Dypbukt et al., 1992; Horvath et al., 1994;Vodicka et al., 1993; Walles et al., 1993). Oxidative DNA

damage has also been implicated as a causative factor

in the DNA strand breaks detected in blood samples

exposed to SO in vitro (Marczynski et al., 1997). Our

18 M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20

study, showing basal levels of adducts in agreement with

the values reported in the literature (Lee et al., 2001),

provide evidence that also neurons are subjected to DNA

oxidation by SO, documenting the presence of increased8-OHdG levels in SK-N-MC cells treated with 0.3 or 1

mM SO.

When exposed to SO (1 mM) SK-N-MC neurons

increased the activity of HO, the isoenzymes involved in

the stepwise degradation of heme to iron, CO and bili-

verdin, which is converted by biliverdin reductase to

bilirubin. The HO proteins, which are rate-limiting en-

zymes in heme catabolism, are present both in a con-stitutive (HO-2) and inducible (HO-1) form. Even if the

mechanisms by which HO-1 can mediate its functions is

not clear, accumulating evidence suggests that the

endogenous induction of this enzyme provides potent

cytoprotective effects in various in vitro and in vivo

models of cellular and tissue injury (Hancock et al.,

1998; Otterbein et al., 1999; Soares et al., 1998). HO-1,

which is known also as heat shock protein 32 (HSP32),provides a remarkable example of gene regulation by

oxidative stress in mammalian cells, including neurons

(Tyrrell et al., 1993). For example, GSH depletion and

ROS generation have been associated to an increase

in HO-1 expression in motor neurons exposed to eth-

acrylic acid (Rizzardini et al., 2003). HO-1 is dramati-

cally up regulated by oxidative stress also in

dopaminergic neuronal cells (Yoo et al., 2003). We havefound that 0.3 mM SO induced the expression of HO-1

in SK-N-MC neurons. This alteration is a sign of a

cellular stressresponse to the increased generation of

ROS determined by the exposure to SO. The discrep-

ancy between the results obtained by immunocyto-

chemistry and HO-1 activity measurements at 0.3 mM

SO could be ascribed to the lower sensitivity of the

enzymatic method.In conclusion, we provide evidence that oxidative

stress is induced by SO exposure in neuronal cells at the

same concentrations that result in apoptotic cell death.

Further work is in progress to evaluate the contribution

of oxidative stress, to the ‘‘initiation phase’’ of SO-

induced apoptosis. In this respect, we have obtained

promising results showing that antioxidant pre-treat-

ment reduces activation of caspases and apoptotic deathinduced by 0.3 mM SO in SK-N-MC cells (Dar�e et al.,in press). Altogether, our finding point to a role of

oxidative stress in the early phases of SO-triggered

apoptosis in neurons.

Acknowledgements

This study was supported by the European Com-

mission (contract QLK4-1999-01356). The cooperation

of Dr. Giuseppina Folesani and Paola Mozzoni for the

technical assistance is gratefully acknowledged. We

thank Dr. Paola Manini for comments on the manu-

script.

References

Aksenov, M.Y., Aksenova, M.V., Butterfield, D.A., Geddes, J.W.,

Markesbery, W.R., 2001. Protein oxidation in the brain in

Alzheimer’s disease. Neuroscience 103 (2), 373–383.

Arfini, G., Mutti, A., Vescovi, F., Ferroni, C., Ferrari, M., Giaroli, C.,

Franchini, I., 1987. Impaired dopaminergic modulation of pituitary

secretion in workers occupationally exposed to styrene: further

evidence from PRL response to THR. Journal of Occupational and

Environmental Medicine 10, 826–830.

Bastlova, T., Vodicka, P., Peterkova, K., Hemminki, K., Lambert, B.,

1995. Styrene oxide-induced HPRT mutations, DNA adducts and

DNA strand breaks in cultured human lymphocytes. Carcinogen-

esis 16, 2357–2362.

Beiswanger, C.M., Mandella, R.D., Graessle, T.R., Reuhl, K.R.,

Lowndes, H.E., 1993. Synergistic neurotoxic effects of styrene

oxide and acrylamide: glutathione-independent necrosis of cere-

bellar granule cells. Toxicology and Applied Pharmacology 118,

233–244.

Belletti, S., Orlandini, G., Vettori, M.V., Mutti, A., Uggeri, J.,

Scandroglio, R., Alinovi, R., Gatti, R., 2002. Time corse assess-

ment of methylmercury effects on C6 glioma cells: submicromolar

concentrations induce oxidative DNA damage and apoptosis.

Journal of Neuroscience Research 70 (5), 703–711.

Butterfield, D.A., Castegna, A., Lauderback, C.M., Drake, J., 2002.

Evidence that amyloid beta-peptide-induced lipid peroxidation and

its sequelae in Alzheimer’s disease brain contribute to neuronal

death. Neurobiology of Aging 23 (5), 655–664.

Chakrabarti, S.K., 1999. Styrene and styrene oxide affect the transport

of dopamine in purified rat striatal synaptic vesicles. Biochemical

Biophysical Research Communications 255, 70–74.

Choi, J., Conrad, C.C., Malakowsky, C.A., Talent, J.M., Yuan, C.S.,

Gracy, R.W., 2002. Flavones from Scutellaria baicalensis Georgi

attenuate apoptosis and protein oxidation in neuronal cell lines.

Biochimica and Biophysica Acta 1571 (3), 201–210.

Dar�e, E., Tofighi, R., Vettori, M.V., Momoi, T., Poli, D., Saido, T.,Mutti, A., Ceccatelli, S., 2002. Styrene 7,8-oxide induces caspase

activation and regular DNA fragmentation in neuronal cells. Brain

Research 933, 12–22.

Dar�e, E., Tofighi, R., Nutt, L., Vettori, M.V., Emgard, M., Mutti, A.,Ceccatelli, S., 2004. Styrene, 8-oxide induces mitochondrial damage

and oxidative stress in neurons. Toxicology, in press.

Dean, R.T., Fu, S., Stocker, R., Davies, M.J., 1997. Biochemistry and

pathology of radical-mediated protein oxidation. Biochemical

Journal 324, 1–18.

Dypbukt, J.M., Costa, L., Manzo, L., Orrenius, S., Nicotera, P., 1992.

Cytotoxic and genotoxic. Carcinogenesis 13, 417–424.

Fleury, C., Mignotte, B., Vayssiere, J.L., 2002. Mitochondrial

reactive oxygen species in cell death signaling. Biochimie 84, 131–

141.

Flora, G.J., Seth, P.K., 2000. Alterations in some membrane proper-

ties in rat brain following exposure to lead. Cytobios 103 (403),

103–109.

Habig, W.H., Pabst, M.J., Jakoby, W.B., 1974. Glutathione-S-trans-

ferases. The first enzymatic step in mercapturic acid formation.

Journal of Biological Chemistry 249 (22), 7130–7139.

Hancock, W.W., Buelow, R., Sayegh, M.H., Turka, L.A., 1998.

Antibody-induced transplant arteriosclerosis is prevented by graft

expression of anti-oxidant and anti-apoptotic genes. Nature

Medicine 4, 1392–1396.

Horvath, E., Pongracz, K., Rappaport, S., Bodell, W.J., 1994. 32P-

post-labeling detection of DNA adducts in mononuclear cells of

M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20 19

workers occupationally exposed to styrene. Carcinogenesis 15,

1309–1315.

Hyun, D.H., Lee, M.H., Halliwell, B., Jenner, P., 2002. Proteasomal

dysfunction induced by 4-hydroxy-2,3-trans-nonenal, an end-

product of lipid peroxidation: a mechanism contributing to

neurodegeneration? Journal of Neurochemistry 83 (2), 360–

370.

IARC, 1994. Styrene-7,8-oxide. IARC Monographs on the Evaluation

of Carcinogenic Risks to Humans 60, 321–346.

Inayama, T., Kumagai, Y., Sakane, M., Saito, M., Matsuda, M., 1996.

Plasma protein-bound sulfhydryl group oxidation in humans

following a full marathon race. Life Sciences 59 (7), 573–

578.

Jentzsch, A.M., Bachmann, H., Furst, P., Biesalski, H.K., 1996.

Improved analysis of malondialdehyde in human blood fluids. Free

Radical Biology and Medicine 20 (2), 251–256.

Kanski, J., Drake, J., Aksenova, M., Engman, L., Butterfield, D.A.,

2001. Antioxidant activity of the organotellurium compound 3-[4-

(N,N-dimethylamino) benzenetellurenyl] propanesulfonic acid

against oxidative stress in synaptosomal membrane systems and

neuronal cultures. Brain Research 911 (1), 12–21.

Katoh, T., Higashi, K., Inoue, N., 1989. Sub-chronic effects of styrene

and styrene oxide on lipid peroxidation and the metabolism of

gluthathione in rat liver and brain. Journal of Toxicological

Sciences 14, 1–9.

Keller, J.N., Huang, F.F., Dimayuga, E.R., Maragos, W.F., 2000.

Dopamine induces proteasome inhibition in neural PC12 cell line.

Free Radical Biology and Medicine 29 (10), 1037–1042.

Kenchappa, R.S., Diwakar, L., Boyd, M.R., Ravindranath, V., 2002.

Thioltransferase (glutaredoxin) mediates recovery of motor neu-

rons from excitotoxic mitochondrial injury. Journal of Neurosci-

ence 22, 8402–8410.

Kim, K.B., Lee, B.M., 1997. Oxidative stress to DNA, protein, and

antioxidant enzymes (superoxide dismutase and catalase) in rats

treated with benzo(a)pyrene. Cancer Letters 113, 205–212.

Kim, D.H., Jang, Y.Y., Han, E.S., Lee, C.S., 2001. Protective effect of

harmaline and harmalol against dopamine- and 6-hydroxydop-

amine-induced oxidative damage of brain mitochondria and

synaptosomes, and viability loss of PC12 cells. European Journal

of Neuroscience 13 (10), 1861–1872.

Kohn, J., Minotti, S., Durham, H., 1995. Assessment of the neuro-

toxicity of styrene, styrene oxide, and styrene glycol in primary

cultures of motor and sensory neurons. Toxicology Letters 75, 29–

37.

Laffon, B., P�asaro, E., M�endez, J., 2001. Genotoxic effects of styrene-

7,8-oxide in human white blood cells: comet assay in relation to the

induction of sister-chromatid exchanges and micronuclei. Mutation

Research 491, 163–172.

Laffon, B., P�asaro, E., M�endez, J., 2002. DNA damage and repair in

human leukocytes exposed to styrene-7,8-oxide measured by the

comet assay. Toxicology Letters 126, 61–68.

Lee, M., Hyun, D.H., Marshall, K.A., Ellerby, L.M., Bredesen, D.E.,

Jenner, P., Halliwell, B., 2001. Effect of overexpression of BCL-2

on cellular oxidative damage, nitric oxide production, antioxidant

defenses, and the proteasome. Free Radical Biology and Medicine

31 (12), 1550–1559.

Marczynski, B., Peel, M., Baur, X., 1997. Changes in high molecular

weight DNA fragmentation following human blood exposure to

styrene-7,8-oxide. Toxicology 120, 111–117.

Migheli, R., Godani, C., Sciola, L., Delogu, M.R., Serra, P.A.,

Zangani, D., De Natale, G., Miele, E., Desole, M.S., 1999.

Enhancing effect of manganese on L-DOPA-induced apoptosis in

PC12 cells: role of oxidative stress. Journal of Neurochemistry 73

(3), 1155–1163.

Mutti, A., Mazzucchi, A., Rustichelli, P., Frigeri, G., Arfini, G.,

Franchini, I., 1984a. Exposure-effect and exposure-response rela-

tionships between occupational exposure to styrene and neuropsy-

chological functions. American Journal of Industrial Medicine 5,

275–286.

Mutti, A., Falzoi, M., Romanelli, A., Franchini, I., 1984b. Regional

alterations of brain catecholamines by styrene exposure in rabbits.

Archives of Toxicology 55, 173–177.

Mutti, A., Franchini, I., 1987. Toxicity of metabolites to dopaminergic

systems and the behavioural effects of organic solvents. British

Journal of Industrial Medicine 44, 721–723.

Mutti, A., Falzoi, M., Romanelli, A., Bocchi, M.C., Ferroni, C.,

Franchini, I., 1988. Brain dopamine as a target for solvent toxicity:

effects of some monocyclic aromatic hydrocarbons. Toxicology 49,

77–82.

Naarala, J.T., Loikkanen, J.J., Ruotsalainen, M.H., Savolainen, K.M.,

1995. Lead amplifies glutamate-induced oxidative stress. Free

Radical Biology and Medicine 19 (5), 689–693.

Naito, M., Umegaki, H., Iguchi, A., 1995. Protective effects of

probucol against glutamate-induced cytotoxicity in neuronal cell

line PC12. Neuroscience Letters 186 (2-3), 211–213.

Nordmann, R., 1987. Oxidative stress from alcohol in the brain.

Alcoholism Alcohol Suppl. 1, 75–82.

Olivieri, G., Novakovic, M., Savaskan, E., Meier, F., Baysang, G.,

Brockhaus, M., Muller-Spahn, F., 2002. The effects of beta-

estradiol on SHSY5Y neuroblastoma cells during heavy metal

induced oxidative stress, neurotoxicity and beta-amyloid secretion.

Neuroscience 113 (4), 849–855.

Otterbein, L.E., Mantell, L.L., Choi, A.M., 1999. Carbon monoxide

provides protection against hyperoxic lung injury. American

Journal of Physiology: Lung Cellular and Molecular Physiology

276, L688–L649.

Poli, D., Vettori, M.V., Andreoli, R., Manini, P., Alinovi, R.,

Ceccatelli, S., Mutti, A., 2004. A novel approach based on solid

phase micro extraction gas chromatography and mass spectrom-

etry to the determination of highly reactive organic compounds in

cell cultures: styrene-7,8-oxide. Chemical and Research in Toxicol-

ogy 17 (1), 104–109.

Reznick, A.Z., Packer, L., 1994. Oxidative damage to proteins:

spectrophotometric method for carbonyl assay. Methods in Enzy-

mology 233, 357–363.

Rizzardini, M., Lupi, M., Bernasconi, S., Mangolini, A., Cantoni, L.,

2003. Mitochondrial dysfunction and death in motor neurons

exposed to the glutathione-depleting agent ethacrylic acid. Journal

of the Neurological Sciences 8798, 1–8.

Sherer, T.B., Betarbet, R., Testa, C.M., Seo, B.B., Richardson, J.R.,

Kim, J.H., Miller, G.W., Yagi, T., Matsuno-Yagi, A., Greenamyre,

J.T., 2003. Mechanism of toxicity in rotenone models of Parkin-

son’s disease. Journal of Neuroscience 23 (34), 10756–10764.

Soares, M.P., Lin, Y., Anrather, J., Csizmadia, E., Takigami, K., Sato,

K., Grey, S.T., Colvin, R.B., Choi, A.M., Poss, K.D., Bach, F.H.,

1998. Expression of heme oxygenase-1 can determine cardiac

xenograft survival. Nature Medicine 4, 1073–1077.

Sriram, K., Shankar, S.K., Boyd, M.R., Ravindranath, V., 1998. Thiol

oxidation and loss of mitochondrial complex I precede excitatory

amino acid-mediated neurodegeneration. Journal of Neuroscience

18, 10287–10296.

Trenga, C.A., Kunkel, D.D., Eaton, D.L., Costa, L.G., 1991. Effect of

styrene oxide on rat brain glutathione. Neurotoxicology 12, 165–

178.

Tyrrell, R.M., Applegate, L.A., Tromvoukis, Y., 1993. The proximal

promoter region of the human heme oxygenase gene contains

elements involved in stimulation of transcriptional activity by a

variety of agents including oxidants. Carcinogenesis 14, 761–

765.

Uysal, M., Kutalp, G., Odzemirler, G., Ayka, G., 1989. Ethanol-

induced changes in lipid peroxidation and glutathione content in

rat brain. Drug and Alcohol Dependence 23, 227–230.

Verity, M.A., Sarafian, T., 1991. Role of oxidative damage in the

pathogenesis of methylmercury neurotoxicity. In: Suzuki, T.,

20 M.V. Vettori et al. / Toxicology in Vitro 19 (2005) 11–20

Imura, N., Clarkson, T.W. (Eds.), Advances in Mercury Toxicol-

ogy. Plenum Press, New York, pp. 209–222.

Vodicka, P., Koskinen, M., Arand, M., Oesch, F., Hemminki, K.,

2002. Spectrum of styrene-induced DNA adducts: the relationship

to other biomarkers and prospects in human biomonitoring.

Mutation Research 511 (3), 239–254.

Vodicka, P., Vodickova, L., Hemminki, K., 1993. 32P-postlabeling of

DNA adducts of styrene-exposed lamination workers. Carcino-

genesis 14, 2059–2061.

Walles, S.A., Edling, C., Anundi, H., Johanson, G., 1993. Exposure

dependent increase in DNA single strand breaks in leucocytes from

workers exposed to low concentrations of styrene. British Journal

of Industrial Medicine 50, 570–574.

Xie, C., Lovell, M.A., Xiong, S., Kindy, M.S., Guo, J., Xie, J.,

Amaranth, V., Montine, T.J., Markesbery, W.R., 2001. Expression

of glutathione-S-transferase isozyme in the SY5Y neuroblastoma

cell line increases resistance to oxidative stress. Free Radical

Biology and Medicine 31 (1), 73–81.

Yatin, S.M., Varadarajan, S., Link, C.D., Butterfield, D.A., 1999. In

vitro and in vivo oxidative stress associated with Alzheimer’s

amyloid beta-peptide (1–42). Neurobiology of Aging 20 (3), 325–

330.

Yoo, M.S., Chun, H.S., Son, J.J., Giorgio, L.A., Kim, D.J., Peng, C.,

Son, J.H., 2003. Oxidative stress regulated genes in nigral dopami-

nergic neuronal cells: correlation with the known pathology in

Parkinson’s Disease. Molecular Brain Research 110 (1), 76–84.