Ecotoxicology and Environmental Safety 73 (2010) 1712–1719

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety

journal homepage: www.elsevier.com/locate/ecoenv

Uptake and biochemical responses of mussels Mytilus galloprovincialis

exposed to sublethal nickel concentrations
Hajer Attig a, Alessandro Dagnino b, Alessandro Negri b, Jamel Jebali a, Hamadi Boussetta a,
Aldo Viarengo b, Francesco Dondero b, Mohamed Banni a,n
a
Laboratory of Biochemistry and Environmental Toxicology, Higher Institute of Agronomy, ISA, 4042, Chott Mariem, Sousse, Tunisia
b
Department of Environmental and Life Sciences, University of Piemonte Orientale Amedeo Avogadro, Via Bellini 25 G, 15100 Alessandria, Italy

a r t i c l e in f o a b s t r a c t

Article history: In the present study, mussel (Mytilus galloprovincialis) digestive gland oxidative stress biomarkers and
Received 15 June 2010 detoxification responses to acute exposure to nickel (Ni) were investigated. Mussels were exposed to
Received in revised form two sublethal concentrations of Ni (135 mg/L per animal (2.5 mM) and 770 mg/L per animal (13 mM)) for
4 August 2010
24, 48, 72, 96 h and 8 days. Following biological responses were measured: (1) glutathione S-transferase
Accepted 6 August 2010
Available online 25 August 2010
(GST) activity as a phase II conjugation enzyme, (2) catalase activity as antioxidant response,
(3) malondialdehyde accumulation (MDA) as lipid peroxydation marker and metallothionein as specific
Keywords: response to metals exposure. The cholinergic system was evaluated using the acetylcholinesterase
Acute exposure activity (AChE). Moreover, Ni uptakes during the exposure periods were assessed and the uptake rate
Ni
constant determined. A correlation matrix (CM) between the investigated biomarkers and a principal
Toxicokinetic
component analysis (PCA) were achieved for the two tested concentrations. The Ni-uptake constant was
Toxicodynamic
higher in animals exposed to the lowest concentration. The CM and the PCA showed a time-dependent
effect of the Ni exposure on the investigated biomarkers being more pronounced in animals exposed to
the highest Ni concentration. While AChE showed a significant increase after 48 h and a further return
to control values in the lowest concentration, it was drastically maintained inhibited in the highest
concentration. Our data provided clues about the occurrence of different toxicokinetics and
toxicodynamics of two Ni sublethal concentrations in an ecologically relevant organism.
& 2010 Elsevier Inc. All rights reserved.

1. Introduction physiological status of the organism and factors that control metal
uptake from solution (Salinity and dissolved organic carbon
Nickel is a metal of high environmental importance that has concentration) (Pan and Wang, 2004).
been shown to exert long-term harmful effects to aquatic biota Bivalve molluscs, particularly marine mussels such as Mytilus
(Kienle et al., 2009). Nickel is often found in the coastal spp., have been used as indicator organisms in environmental
environment as a result of industrial discharges from electroplat- monitoring programmes (Banni et al., 2007; Viarengo et al., 2007)
ing, smelting, mining and refining operations and other industrial due to their wide distribution, sedentary lifestyle, tolerance to a
emissions (Vijayavel et al., 2009). Relative to other divalent large range of environmental conditions and because they are
metals, nickel has not been well studied in terms of toxicity to filter feeders with very low metabolism, which allows the
different species and mode of action. bioaccumulation of many chemicals in their tissues (Widdows
Over the past decade, a major advance in understanding metal et al., 2002).
ecotoxicology has been the development of the biokinetic or The use of biomarkers in studying the effects of environmental
biodynamic model, which has been instrumental in interpreting pollutants on mussels has been reported to be very informative
the total bioaccumulated metal concentrations in different about the organism’s stress response to toxicants (Banni et al.,
species of aquatic organisms (Wang and Rainbow, 2008). Never- 2005; Dagnino et al., 2007, Viarengo et al., 2007). Metallothio-
theless, there is a substantial need to understand the bioaccumu- neins (MTs) are a class of low molecular weight, cysteine rich,
lation and toxicity of some metals such as Ni in aquatic organisms inducible, cytosolic proteins well known for their high affinity to
that may depend essentially on the exposure concentration, the bind metals. MTs are implicated in homeostasis of essential
metals, detoxification of noxious metal cations (Viarengo et al.,
1999) and oxidative stress protection (Dondero et al., 2005).
n
Corresponding author. Fax: + 216 73 327591. Glutathione S-transferase (GST) is a phase II enzyme known to be
E-mail address: [email protected] (M. Banni). involved in the metabolism of lipophilic organic contaminants,

0147-6513/$ - see front matter & 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.ecoenv.2010.08.007
 H. Attig et al. / Ecotoxicology and Environmental Safety 73 (2010) 1712–1719 1713

but also plays a role in cellular protection against oxidative stress GST activity was measured in digestive gland cytosol by the method of Habig
et al. (1974) using 10 mg of cytosolic protein, 1 mM 1-chloro-2,4-dinitrobenzene
(Michel et al., 1998). Catalase (CAT) is a well-known anti-oxidant
(CDNB) (Sigma-Aldrich, Saint Louis, MO, USA) as substrate, and 4 mM glutathione
enzyme, its activity increasing in organisms submitted to reduced from GSH, in 100 mM sodium phosphate buffer, pH 7.5. GST activity was
oxidative stress (Durou et al., 2007). One of the well-known lipid determined by kinetic measurement at 20 1C using a Jenway 6105 spectro-
peroxidation products (LPO) is malondialdehyde (MDA); this photometer (l ¼ 340 nm). Results were expressed as nmoles GSH–CDNB produced
marker was usually used to evaluate the state of lipid peroxida- per min and per mg protein.
CAT was determined according to the Claiborne’s method (1985). Reaction
tion of the membrane (Alexandrova and Bochev, 2005). mixture (final volume of 1 ml) contained 0.78 ml 0.1 M phosphate buffer (pH 7.5)
Acetylcholinesterase (AChE) is an enzyme essential for the correct and 0.2 ml 0.5 mM H2O2. After 30 s pre-incubation, the reaction was started by the
transmission of nerve impulses. Its inhibition is directly linked addition of 0.02 ml of the (S9) solution containing CAT fractions. CAT activity was
with the mechanisms of toxic action of some pesticides and evaluated by kinetic measurement at 20 1C using a Jenway 6105 spectro-
photometer (l ¼ 240 nm). Results were expressed as mmoles hydrogen peroxide
metals (Tsangaris et al., 2007; Viarengo et al., 2007).
transformed per min and per mg protein.
The aim of the present study is to evaluate the toxicokinetic Lipid peroxidation was estimated in terms of thiobarbituric acid reactive
and toxicodynamics of two sublethal Ni concentrations on the species (TBARS) with use of 1,1,3,3-treaethyloxypropane as a standard. The
mussel Mytilus galloprovincialis. A set of biomarkers was applied reaction was assessed at 532 nm, using TBA reagent as described by Buege and
in order to assess the Ni-induced stress after 24, 48, 72, 96 h and Aust (1978). MDA content was expressed as nmoles equivalent MDA/mg proteins.
AChE was determined accordingly to Ellman et al. (1961). Reaction mixture
8 days of exposure.
(final volume of 1 ml) contained 0.85 ml 0.1 M phosphate buffer (pH 7.5), 0.05 ml
8 mM DNTB (Sigma-Aldrich, Saint Louis, MO, USA) and 0.05 ml of the (S9) fraction
containing AChE. After pre-incubation, the reaction was started by the addition of
2. Material and methods 0.05 ml 8.25 mM acetylthiocholine (Sigma-Aldrich, Saint Louis, MO, USA).
Acetylcholinesterase activity was determined by kinetic measurement at 20 1C
using a Jenway 6105 spectrophotometer (l ¼420 nm). Results were expressed as
2.1. Animals and treatments
mmoles thiocholine produced per min and per mg protein.

Specimens of M. galloprovincialis (Lam) 4–5 cm shell length were purchased

from an aquaculture farm in Bizerta Lagoon (Tunisia) and further acclimatised to 2.4. Statistical analysis
aerated and daily renewed clean natural seawater in an aquarium for 15 days at
16 1C (35% salinity, 1.5 L/animal). For the exposure experiments, mussels were
The toxicokinetic model was computed using Sigma SYSTAT 10 from the
separated into different groups and transferred to 40 L glass aquaria filled with
following equation (Wang and Rainbow, 2008):
clean natural seawater to a volume of 1.5 L per mussel (3 aquaria per experimental
condition). Mussels were then treated for 24, 48, 72, 96 h and 8 days under natural Cin ¼ kCwt
light and without any food source. Three different conditions were analyzed: clean
water (control reference), 2.5 mM Ni (daily theoretical nominal dose of 135 mg/L where ‘‘Cin’’ is Ni concentration in mussel tissues (mg/g dry weight), ‘‘k’’ the uptake
per animal) and 13 mM Ni (daily theoretical nominal dose of 770 mg/L per animal). constant (g dry weight/L/day), ‘‘Cw’’ the Ni concentration in saltwater (mg/L) and
Water was daily renewed. The exposure levels represent the EC25 and EC50 for the ‘‘t’’ the exposure time (day).
effect on digestive gland lysosomal membrane stability (LMS) (mussels were Statistical analyses were performed with SP SS/PC (SP SS, Microsoft, and
exposed to a range of 0.01–15 mg/L for 96 h) (Dondero et al., 2008). Redmond, WA). Significant differences between means were determined using
After treatments, a first set of mussel’s digestive gland tissue (20 individual one-way ANOVAs and the Tukey’s test for multiple range comparison with
fractions per exposure period) were rapidly dissected out, washed into artificial significance level established at po 0.05.
seawater buffered with 20 mM Hepes, pH 7.4, flash frozen into liquid nitrogen and Spearman Correlation matrix was also calculated to study the relationships
stored at  80 1C until biomarker analysis. A second set of tissues (10 individual between the different biomarkers measured (Statistica Soft Inc.). Factor analysis of
digestive glands) were flash frozen into liquid nitrogen and stored at  80 1C until the variables analyzed was carried out by means of the principal component
Ni analysis. analysis (PCA) method with orthogonal rotation (Varimax) using the Systat
11 software (SYSTAT Software Inc.). (Varimax minimizes the number of variables
that have high loadings on each factor.) The PCA analysis was used as an effective
2.2. Nickel analysis technique simplifying the correlation structure through linear transformation of
the original variables (Jolliffe, 1986).

The content of Ni in the digestive gland fractions was determined by atomic

absorbtion spectrophotometry (AAS) after acid digestion of 0.5 g dry tissues with
4 ml of 65% HNO3 and 1 ml of 70% HCIO4 for 24 h at 80 1C (Amiard et al., 1987). The 3. Results
acid was removed from the samples by evaporation and the residues were diluted
in 10 ml of 1 N HNO3. Ni was determined by flame and flameless AAS using a
varian spectrophotometer Vectra 250 Plus with ZEEMAN correction. Blanks and The Ni retrieval rates were in the range of 101.8–103.4% of
reference materials (dogfish liver DOLT-2, NRCC) were assessed through the the nominal concentrations (see Table 1).The content of Ni in
procedure in the same way as the sample. Our results (21.0 71.1 mg/g dry weight, M. galloprovincialis digestive gland after the exposure periods is
n¼ 3) were in good agreement with certified values (20.87 0.5 mg/g dry weight).
Levels of Ni are given relative to the dry weight of tissue. Analytical confirmation
reported in Fig. 1. In animal exposed for 24 h to 2.5 mM and 13 mM
of nickel concentrations in reconstituted water was performed by flame atomic Ni, data show an accumulation up to 0.2870.1 and 0.5970.1 mg/g
absorption spectroscopy. dry weight, respectively, in comparisons with a value of 0.08 mg/g
dry weight in control animals. These amounts increase over time
to reach 5.78 70.13 mg/g dry weight in 2.5 mM-exposed animals
2.3. Biochemical analysis
and 8.5370.33 mg/g dry weight in 13 mM-exposed mussels
Before biochemical analysis, digestive glands were homogenized in phosphate
after 8 days of exposure. We also studied the ability of mussels
buffer (0.1 M, pH 7.5). The homogenate obtained was centrifuged at 9000g for to uptake Ni after the exposure period fitting chemical data
cytosolic fractions (S9) or at 20,000g for S20. S9 fraction was used to carry out GST, into a toxicokinetics model, thus deriving the uptake constant ‘‘k’’
CAT, MDA and AChE analyses. S20 fraction was used for MTs analysis. The for each condition (Table 2). Our data show that ‘‘k’’ clearly
quantities of proteins present in S9 and S20 fractions were determined according
decreased when mussels were exposed to the highest Ni
to the Bradford (1976) method using Coomassie Blue reagent.
MTs content was evaluated in digestive gland according to the spectro- concentration.
photometric method described in Viarengo et al. (1997) based on cystein residues The results relative to the MT accumulation are reported in
titration of a partially purified metallothionein extract. MTs protein levels were Fig. 2. Our data show that MT accumulation significantly
determined using a spectrophotometric assay for MT using Ellman’s reagent increased over time and in a dose-dependent fashion in digestive
(0.4 mM DTNB in 100 mM KH2PO4) at pH 8.5 in a solution containing 2 M NaCl
and 1 mM EDTA. Reduced GSH standard solutions were used for calibration
gland tissues of mussels exposed to the two Ni concentrations.
(2–100 mM) and data were expressed as mg MT per mg of protein taking into The maximum were observed after 8 days of exposure with up
consideration mussel’s MT molecular weight and number of cystein residues. to 81.8 77.1 and 105.478.5 mg/mg proteins, respectively, in
1714 H. Attig et al. / Ecotoxicology and Environmental Safety 73 (2010) 1712–1719

Table 1
Nominal and measured Ni concentrations (mg/L) and retrieval rate of nominal concentrations (%); mean 7SD of six replicate measurements.

Ni (135 mg/L) Ni (mg/L) measured Retrieval rate of nominal Ni (755 mg/L) Ni (mg/L) measured Retrieval rate of nominal
nominal concentration concentration (%) nominal concentration concentration (%)

Control 2.3 70.4 – Control 2.3 7 0.42 –

24 h 137.9 712.8 102.27 9.5 24h 768.8 7 55.8 101.8 77.4
48 h 138.2 710.7 102.47 7.9 48h 778.8 7 50.8 103.1 76.7
72 h 137.9 711.4 102.77 8.8 72h 770.6 7 62.3 102.1 78.3
92 h 139.1 713.8 103.17 10.2 92h 780.7 7 73.9 103.4 79.8
8 days 138.1 79.6 102.37 7.1 8 days 777.9 7 59.4 103.1 77.9

10 120 ab
Control Control
9 2.5 µM ab 2.5 µM ab

8 13 µM 100 13 µM a a
a
ab a

µg MT/mg proteins
7 a
µg/g dry weight

a 80 a
6
5 60
4 ab
a
40
3
ab
2 a
a 20
1 ab
a
0 0
24h 48h 72h 96h 8days 24h 48h 72h 96h 8days

Fig. 1. Ni contents (mg/g dry weight) in the digestive gland tissues of mussels Fig. 2. Digestive gland metallothionein accumulation in mussel exposed for 24,
M. galloprovincialis exposed for 24, 48, 72, 96 h and 8 days, to 2.5 and 13 mM Ni 48, 72, 96 h and 8 days, to 2.5 and 13 mM Ni (NiCl2). The control group was
(NiCl2). The control group was maintained in clean seawater in the same maintained in clean seawater in the same conditions as the exposed groups. Data
conditions as the exposed groups. Data are expressed as mean 7SD (n ¼10 represent means7 S.D (n¼ 10). aP o0.05 significantly different, Tukey’s test
individual digestive glands). aPo 0.05 significantly different, Tukey’s test ANOVA ANOVA multiple comparison test versus test versus each respective control
multiple comparison test versus each respective control animal group, bPo 0.05 animal group, bP o0.05 significantly different, Tukey’s test ANOVA multiple
significantly different, Tukey’s test ANOVA multiple comparison test versus 2.5 mM comparison test versus 2.5 mM Ni-reated animals.
Ni-treated animals.

280
Table 2 Control
Toxicokinetic parameters: shown are: uptake constant ‘‘k’’ of Ni for each tested 260
condition empirically evaluated from model fitting (Wang and Rainbow, 2008). 2.5 µM
R2, variance explained by the model. Mussels M. galloprovincialis were exposed for
240
13 µM a a
220 a a
24, 48, 72, 96 h and 8 days to the tested Ni concentrations.
nmole/min/mg proteins

200
Condition 2.5 mM 13 mM 180
Uptake constant (L/g/d) 0.0477 0.002 0.028 7 0.001
R2 0.96 0.95 160
140
120
mussels treated with 2.5 and 13 mM Ni (37% and 76% increase 100
with respect to control). 80
The effects of Ni exposure on the mussel digestive gland GST
60
activity are shown in Fig.3. A weak but significant response
40
compared to control animals was recorded after 96 h and 8 days
of exposure for the two tested concentrations. The maxima were 20
reached after 8 days exposure with around 40% increase with 0
respect to control for both exposed groups. 24h 48h 72h 96h 8days
The results relative to the CAT activity in mussels exposed to Ni
Fig. 3. Digestive gland GST activity in mussel exposed for 24, 48, 72, 96 h and
are reported in Fig. 4. Our data show that CAT activity increased 8days, to 2.5 and 13 mM Ni (NiCl2). The control group was maintained in clean sea
over time reaching two slightly different plateau levels after 72 h water in the same conditions as the exposed groups. Data represent means 7 S.D
with an increase of about 37% and 50% compared to control values (n ¼10). aP o 0.05 significantly different, Tukey’s test ANOVA multiple comparison
test versus test versus each respective control animal group, bP o 0.05 significantly
in animals exposed to 2.5 and 13 mM Ni, respectively.
different, Tukey’s test ANOVA multiple comparison test versus 2.5 mM Ni-treated
The MDA accumulation evaluated as thiobarbituric acid reactive animals.
species (TBARS) in mussel digestive gland after exposure to Ni is
reported in Fig. 5. A time- and dose-dependent accumulation of
MDA was registered in animals exposed to Ni with a maximum of Fig. 6 shows the effect of Ni exposure on digestive gland’s AChE
26.572.2 and 44.472.5 nmole/mg proteins reached after 8 days activity. AChE activity exhibited two distinct trends in response to
of exposure, respectively, for 2.5 and 13 mM Ni. 2.5 and 13 mM Ni exposure. Indeed, in animals exposed to the
 H. Attig et al. / Ecotoxicology and Environmental Safety 73 (2010) 1712–1719 1715

260 35
Control Control
240 2.5 µM 2.5 µM
220 13 µM 30
13 µM
a a a

µmole/min/mg proteins
200 a
µmole/min/mg proteins

a a a
a 25
180 a a
160
20
140
120 15 ab ab
ab ab ab
100
80 10
60
40 5
20
0 0
24h 48h 72h 96h 8days 24h 48h 72h 96h 8days

Fig. 4. Digestive gland CAT activity in mussel exposed for 24, 48, 72, 96 h and Fig. 6. Digestive gland AChE activity in mussel exposed for 24, 48, 72, 96 h and
8days, to 2.5 and 13 mM Ni (NiCl2). The control group was maintained in clean sea 8days, to 2.5 and 13 mM Ni (NiCl2). The control group was maintained in clean sea
water in the same conditions as the exposed groups. Data represent means 7S.D water in the same conditions as the exposed groups. Data represent means7 S.D
(n¼ 10).aP o0.05 significantly different, Tukey’s test ANOVA multiple comparison (n¼10).aPo 0.05 significantly different, Tukey’s test ANOVA multiple comparison
test versus test versus each respective control animal group, bPo 0.05 significantly test versus test versus each respective control animal group, bP o 0.05 significantly
different, Tukey’s test ANOVA multiple comparison test versus 2.5 mM Ni-treated different, Tukey’s test ANOVA multiple comparison test versus 2.5 mM Ni-treated
animals. animals.

highest Ni concentration, a significant positive correlation was

50
Control ab evident also with GST activity and MT accumulation. Interest-
45 2.5 µM ingly, AChE activity showed no correlation with the other
13 µM ab biomarkers analyzed in the digestive glands of mussels exposed
40
to both the tested concentrations.
A PCA was applied to all biomarkers data along with Ni
nmole/mg proteins

35
ab
digestive glands levels (6 variables) for each exposure concentra-
30 a
a a tion: in both treatments, biomarkers showed a clear trend
25 a reflecting stress syndrome development over time (Fig. 8). Follow-
a a
ing Keiser–Guttman criterion (Cliff, 1988; Shaw, 2003) in the
20
examination of eigenvalues and eigenvectors for both treatments,
15 PC1 and PC2 were selected for data interpretation (Table S1).
The first axis (PC1), explaining 64.4% and 70.7% of the total
10
variance, respectively, in the lower and higher Ni exposure
5 condition, showed the most evident biomarker responses in the
mussels exposed to Ni for longer periods, for GST and CAT
0
activities, MDA and MT accumulation, associated with Ni levels.
24h 48h 72h 96h 8days
Furthermore, the variance explained by the second axis (PC2)
Fig. 5. Digestive gland MDA accumulation in mussel exposed for 24, 48, 72, 96 h (17.8% and 16% of the total variance, respectively, in 2.5 and
and 8days, to 2.5 and 13 mM Ni (NiCl2). The control group was maintained in clean 13 mM Ni exposures), was mainly related to AChE activity.
sea water in the same conditions as the exposed groups. Data represent
means7 S.D (n¼ 10).aP o0.05 significantly different, Tukey’s test ANOVA multiple
comparison test versus test versus each respective control animal group, bP o0.05
4. Discussion
significantly different, Tukey’s test ANOVA multiple comparison test versus 2.5 mM
Ni-treated animals.
The study of the biological responses of organisms to different
environmental conditions and the quantitative evaluation of their
lowest Ni concentration, AChE showed a transient increase at 48 h physiological status are being considered as a successful approach
exposure (35.60% increase with respect to control) and a return to for the assessment of environmental quality (Banni et al., 2009,
control values thereafter. However, 13.5 mM Ni exposure caused a 2010; Dondero et al., 2010). Invertebrates, particularly bivalve
pronounced AChE inhibition during all the exposure periods. molluscs such as mussels are suitable organisms for studying the
Therefore a significant difference was registered between the two biological effects of pollutants (Viarengo et al., 2007).
tested Ni concentrations. In the last decades, increasing attention on the ecological and
Fig. 7 displays the correlation obtained amongst the investi- ecotoxicological effects of Ni contamination has been considered
gated biomarkers (including Ni concentration) in digestive gland due to its large industrial application (Papachristou et al., 1993;
of mussels exposed to 2.5 and 13 mM of Ni. All biomarkers showed Kienle et al., 2009; Vandenbrouck et al., 2009; Banni et al., 2010).
significant positive correlation with exposure time and Ni uptake Several studies indicated that metal accumulation in molluscs
in the digestive glands except AChE and CAT activities in the first was; dose- and time-dependent (Das and Jana, 1999; Roesijadi
experimental condition and AChE activity in animals exposed to and Unger, 1993; Wang and Evans, 1993). In the present study, Ni
13 mM Ni. GST activity showed a significant positive correlation in accumulation in the digestive gland tissue of mussels—metabo-
both the treatments with MDA and MT accumulation; CAT lically the most active organ—was found to be regulated by the
activity showed a positive significant correlation in the two ambient Ni concentration and positively correlated with the
exposure conditions only with MDA accumulation, while at the exposure time (r ¼ 0.98 in animals exposed to 2.5 mM Ni and
1716 H. Attig et al. / Ecotoxicology and Environmental Safety 73 (2010) 1712–1719

[Ni]
MDA
AChE
CAT
GST
MT

TIME [Ni] MDA AChE CAT GST

[Ni]
MDA
AChE
CAT
GST
MT

TIME [Ni] MDA AChE CAT GST

Fig. 7. Correlation matrix of studied biomarkers and exposure time in Ni-exposed M. galloprovincialis (A: 2.5 mM; B: 13 mM). Shown are data plots and correlation
coefficient for each test. *Significant at p o0.01.

r ¼0.93 in animals exposed to 13 mM Ni, Fig. 7). However, our data suggest that uptake is reduced at high aqueous Ni
data showed also that the bioaccumulation was higher in animals concentrations, an evidence of regulation.
exposed to the lower Ni concentration, as indicated by the In the present paper, we reported a significant increase in MT
k values obtained fitting the toxicokinetic model (Table 2). Our accumulation starting after 48 h exposure to Ni and further
 H. Attig et al. / Ecotoxicology and Environmental Safety 73 (2010) 1712–1719 1717

1
[Ni]
PC2 (17.75%)

GST
0
-4 -3 -2 MT -1 0 1 2 3 4
MDA
-1 24h
48h
CAT 72h
-2 96h
8 days
AChE
-3
PC1 (64.37%)

[Ni] 2

MDA
1
PC2 (15.95%)

GST
0
-4 -3 -2MT -1 0 1 2 3 4
-1
CAT
24h
-2 48h
72h
-3 96h
AChE 8 days
-4
PC1 (70.65%)

Fig. 8. Biplot from additional principal component analysis of all variables studied in animals exposed to 2.5 mM Ni (A) and 13 mM Ni (B). In the first condition (2.5 mM Ni),
the first component explained 64.37% of the total variance and the second 17.75%. In the second condition (13 mM Ni), the first component explained 70.65% of the total
variance and the second 15.95%. (See also Table S1 for Eigenvalues and percentage variance, Table S2 for details about scores of the different analytical replicates in PC1 and
PC2 and Table S3 for details about Eigenvectors loadings of the different variables analyzed.)

maintained over time. Correlation between Ni exposure and MT other metals (Cu, Zn, Hg, Cd, Ag) tested in the same organism
induction is controversial. Indeed, Amiard et al. (2008) found no (Barka et al., 2001).
correlation between MT and Ni concentrations in mussels Ni was reported to produce low, but measurable levels of free
exposed to Ni in field and laboratory conditions. However, radicals in CHO cells (Huang et al., 1994). In addition to the direct
recently our research group demonstrated that mussels exposed production of free radicals, Ni was suspected to cause depletion of
to Ni displayed a significant increase in mt10 mRNA abundance the antioxidant enzymes system (Denkhaus and Salnikov, 2002)
and no effects on the cognate mt20 (Dondero et al., 2008), and therefore it should be considered a pro-oxidant agent. In
indicating that Ni ions behave in the same way as copper to which mussels, the antioxidant defense system includes enzymes such
mt20 is almost insensitive (Dondero et al., 2005). Such a different as CAT, GST, superoxide dismutase (SOD), glutathione peroxidase
outcome with Amiard’s findings can be explained by seasonal (GR) and other low molecular weight scavengers such as reduced
effects, which are known to affect both MT gene and protein glutathione (GSH) (Livingstone, 2001). In this work, we found
expression in the tissues of mussels as previously argued some evidence that Ni can produce oxidative stress in the
(Dondero et al., 2006). Moreover, Ni ions are known to have digestive gland tissue of mussels due to the fair increase in GST
a high affinity for sulfydrylic groups of proteins (Costa et al., 1994) and CAT activities. Moreover, a time- and dose-dependent
and MT induction by Ni was reported not only in vertebrates such increase in MDA was observed along with a positive and
as the cod E. navaga (Eriksen et al., 1990), but also in the copepod significant correlation between MDA accumulation and Ni uptake
T. brevicornis (Barka et al., 2001), although with less extent than at the two exposure levels (r ¼0.83 and 0.97, respectively,
1718 H. Attig et al. / Ecotoxicology and Environmental Safety 73 (2010) 1712–1719

in animals exposed to 2.5 and 13 mM Ni). However, it is known that References

MDA measured as TBARS can be overestimated as already reported
(Mendes et al., 2009). In support of this hypothesis, it should be Alexandrova, M.L., Bochev, P.G., 2005. Oxidative stress during the chronic phase of
stroke. Free Radical Biol. Med. 39, 297–316.
noted that in the same exposure conditions after 96 h, no change in Amiard, J.C., Pineau, A., Boiteau, H.-L., Metayer, C., Amiard-Triquet, C., 1987.
the lysosomal lipofuscin levels—the end-product of biological Application de la spectrométrie d’absorption atomique Zeeman aux dosages de
membrane LPO—could be observed (Dondero et al., 2008). huit éléments traces (Ag, Cd, Cr, Cu, Mn, Ni, Pb et Se) dans les matrices
biologiques solides. Water Res. 21, 693–697.
AChE activity of different marine species has been found to be Amiard, J.C., Journel, R., Bacheley, H., 2008. Influence of field and experimental
modulated by environmental contaminants other than organo- exposure of mussels (Mytilus sp.) to nickel and vanadium on metallothionein
phosphate and carbamate insecticides, such as metals and concentration. Comp. Biochem. Physiol. C. 147, 378–385.
Banni, M., Bouraoui, Z., Clerendeau, C., Narbonne, J.F., Boussetta, H., 2009. Mixture
hydrocarbon compounds (Jebali et al., 2006; Viarengo et al., toxicity assessment of cadmium and benzo[a]pyrene in the sea worm Hediste
2007; Banni et al., 2010). Data reported in the present work diversicolor. Chemosphere 77, 902–906.
showed two distinct response patterns of the AChE activities in Banni, M., Dondero, F., Jebali, J., Guerbej, H., Boussetta, H., Viarengo, A., 2007.
Assessment of heavy metal contamination using real time PCR analysis of
the digestive gland of mussels exposed to 2.5 and 13 mM Ni.
mussel metallothionein mt10 and mt20 expression: a validation along the
Indeed, while AChE showed a transient up-regulation at 48 h, it Tunisian coasts. Biomarkers 12 (4), 369–383.
was drastically inhibited at the higher exposure level. Interest- Banni, M., Jebali, J., Daubeze, M., Clerendeau, C., Guerbej, H., Narbonne, J.F.,
ingly, a significant difference in the AChE activity was observed Boussetta, H., 2005. Monitoring Pollution in Tunisian coasts: application
of a classification scale based on biochemical markers. Biomarkers 10,
during all the exposure periods between the two experimental 105–116.
conditions. Recently, Jebali et al. (2006) demonstrated that Cd Banni, M., Negri, A., Dagnino, A., Jebali, J., Ameur, S., Boussetta, H., 2010. Acute
treatment in S. dumerilli results in significant induction of AChE effects of benzo[a]pyrene on digestive gland enzymatic biomarkers and DNA
damage on mussel Mytilus galloprovincialis. Ecotoxicol. Environ. Saf. 10.1016/
activity at low doses, and a strong and dose-dependent AChE j.ecoenv.2009.12.032.
inhibition with higher doses of the same metal. Barka, S., Pavillon, J.F., Amiard, J.C., 2001. Influence of different essential and non-
Biomarker data and Ni uptake results were summarized by essential metals on MTLP levels in the Copepod Tigriopus brevicornis. Comp.
Biochem. Physiol. C 128, 479–493.
means of PCA (Fig. 8). PC1 discriminated data mainly following Bradford, M., 1976. A rapid and sensitive method for the quantification of
exposure time: results from Ni-exposed organisms showed microgram quantities of protein utilizing the principle of protein–dye binding.
positive scores in PC1 after 24 and 48 h; differently, data showed Anal. Biochem. 72, 248–254.
Buege, J.A., Aust, S.D., 1978. Microsomal lipid peroxidation. In: Packer, L. (Ed.),
negative scores at 96 and 8 days (Table S2). PC1 was able to
Methods in Enzymology, vol. 52; 1978, pp. 302–310.
summarize biomarker responses in mussels: in particular eigen- Claiborne, A., 1985. Catalase activity. In: Greenwald, R.A. (Ed.), CRC Handbook of
vectors loadings for PC1 indicated negative values to biomarkers Methods in Oxygen Radical Research. CRC Press, Florida, pp. 283–294.
Cliff, N., 1988. The eigenvalues-greater-than-one rule and the reliability of
such as GST, MT, MDA and CAT, as well as [Ni], and a positive
components. Psychol. Bull. 103, 276–279.
loading to AChE (Table S3). PC2 described only a minimal amount Costa, M., Zhuang, Z., Huang, X., Cosentino, S., Klein, C.B., Salnikov, K., 1994.
of data variance, near the threshold of triviality (Cliff, 1988; Shaw, Molecular mechanisms of nickel carcinogenesis. Sci. Total Environ. 148,
2003): in particular data from 24 and 8 days of exposure showed 191–199.
Dagnino, A., Allen, J.I., Moore, M.N., Broeg, K., Canesi, L., Viarengo, A., 2007.
positive scores, suggesting this axis was able to describe Development of an expert system for the integration of biomarker
intermediate responses to Ni uptake (Table S2). Moreover, PC2 responses in mussels into an animal health index. Biomarkers 12 (2),
showed a negative loading to AChE (table S3), characterized by 155–172.
Das, S., Jana, B.B., 1999. Dose-dependent uptake and Eichhornia-induced elimina-
a particular response profile, different from other parameters. tion of cadmium in various organs of the freshwater mussel, Lamellidens
marginalis (Linn.). Ecol. Eng. 12, 207–229.
Denkhaus, E., Salnikov, K., 2002. Nickel essentiality, toxicity, and carcinogenicity.
Crit. Rev. Oncolog./Hematol. 42, 35–56.
5. Conclusion
Dondero, F., Piacentini, L., Banni, M., Rebelo, M., Burlando, B., Viarengo, A., 2005.
Quantitative PCR analysis of two molluscan metallothionein genes unveils
The present study demonstrated that sub-chronic exposure to differential expression and regulation. Gene 345, 259–270.
Dondero, F., Dagnino, A., Jonsson, H., Capri, F., Gastaldi, L., Viarengo, A., 2006.
two sublethal Ni concentrations resulted in different toxicoki- Assessing the occurrence of a stress syndrome in mussels (Mytilus edulis)
netics and toxicodynamics on the digestive gland of the using a combined biomarker/gene expression approach. Aquat. Toxicol. 78,
Mediterranean mussel M. galloprovincialis. Indeed, mussel’s ability S13–24.
Dondero, F., Boatti, L., Banni, M., Negri, A., Marsano, F., Jones, O., Griffin, L.,
to accumulate Ni was reduced when Ni concentration in water
Viarengo, A., 2008. Molecular analysis of mixture toxicity in the marine mussel
was higher suggesting a compensatory strategy to diminish the Mytilus galloprovincialis: a systems biology approach. In: Proceedings of the
intoxication level in Ni-exposed mussels. On the other hand, even, SETAC 18th Annual Meeting, Warsaw, Poland.
the temporal trend for most investigated biomarkers was similar Dondero, F., Negri, A., Boatti, L., Marsano, F., Mignone, F., Viarengo, A., 2010.
Transcriptomic and proteomic effects of a neonicotinoid insecticide mixture in
and depicting a positive response of animals to metallic stress, the the marine mussel (Mytilus galloprovincialis, Lam.). Sci. Total Environ. 408 (18),
AChE profile was completely different with a persistent inhibition 3775–3786.
at the higher dose. The lack of a compensatory response for AChE Durou, C., Poirier, L., Amiard, J.C., Budzinski, H., Gnassia-Barelli, M., Lemenach, K.,
Peluhet, L., Mouneyrac, C., Romeo, M., Amiard-Triquet, C., 2007. Biomonitoring
is the key discriminating factor between the two exposure levels. in a clean and a multi-contaminated estuary based on biomarkers and
chemical analyses in the endobenthic worm Nereis diversicolor. Environ. Pollut.
148, 445–458.
Eriksen, K.D.H., Andersen, T., Stenersen, J., 1990. Iso-metallothionein in cod
Acknowledgments
induced by cadmium, zinc and nickel. In: Proceedings of the 12th Annual
Conference on Physiological and Biochemical Approaches to the Toxicological
This work was supported by funds from ‘‘Ministe re de Assessment of Environmental Pollution.
Ellman, G.L., Courtneyk, D., Andres, V., Featherstone, R.M., 1961. A new and rapid
l’Enseignement Supérieur et de la Recherche Scientifique, Tunisia; colorimetric determination of acetylcholinesterase activity. Biochem. Pharma-
UR04A6R05. Biochimie et Toxicologie Environnementale’’, 6th col. 7, 88–95.
UE-FW-Project NoMiRACLE (Contract number, 003956). Habig, W.H., Pabst, M.J., Jakoby, W.B., 1974. Glutathione S-transferases, the first
step in mercapturic acid formation. J. Biol. Chem. 249, 7130–7139.
Huang, X., Klein, C.B., Costa, M., 1994. Crystalline Ni3S2 specifically enhances the
formation of oxidants in the nuclei of CHO cells as detected by dichloro-
Appendix A. Supplementary material fluorescein. Carcinogenesis 15 (3), 545–558.
Jebali, J., Banni, M., Guerbej, H., Almeida, E.A., Bannaoui, A., Boussetta, H., 2006.
Effects of malathion and cadmium on acetylcholinesterase activity and
Supplementary data associated with this article can be found metallothionein levels in the fish Seriola dumerilli. Fish Physiol. Biochem. 32,
in the online version at doi:10.1016/j.ecoenv.2010.08.007. 93–98.
 H. Attig et al. / Ecotoxicology and Environmental Safety 73 (2010) 1712–1719 1719

Jolliffe, I.T., 1986. Principal Component Analysis. Springer Verlag, New York, NY, Tsangaris, C., Papathanasiou, E., Cotou, E., 2007. Assessment of the impact of heavy
USA. metal pollution from a ferro-nickel smelting plant using biomarkers.
Kienle, C., Kohler, H.R., Gerhardt, A., 2009. Behavioural and developmental toxicity Ecotoxicol. Environ. Saf. 66, 232–243.
of chlorpyrifos and nickel chloride to zebra fish (Danio rerio) embryos and Vandenbrouck, T., Soetaert, A., van der Ven, K., Blust, R., De Coen, W., 2009. Nickel
larvae. Ecotoxicol. Environ. Saf. 72, 1740–1747. and binary metal mixture responses in Daphnia magna: molecular fingerprints
Livingstone, D.R., 2001. Contaminant-stimulated reactive oxygen species produc- and (sub) organismal effects. Aquat. Toxicol. 92 (1), 18–29.
tion and oxidative damage in aquatic organisms. Mar. Pollut. Bull. 42, Viarengo, A., Ponzano, E., Dondero, F., Fabbri, R., 1997. A simple spectrophoto-
656–666. metric method for metallothionein evaluation in marine organisms: an
Michel, X., Narbonne, J.F., Mora, P., Daube ze, M., Ribera, D., Lafaurie, M., Budzinski, application to Mediterranean and Antarctic molluscs. Mar. Environ. Res. 44,
H., Garrigues, P., 1998. Indicateurs biogéochimiques de pollution des 69–84.
écosyste mes côtiers: expérience du groupe interface chimie-biologie des Viarengo, A., Burlando, B., Cavaletto, M., Marchi, B., Ponzano, E., Blasco, J., 1999.
écosyste mes marins (GICBEM). In: Lagadic, L., Caquet, T., Amiard, J.C., Ramade, Role of metallothionein against oxidative stress in the mussel Mytilus
F. (Eds.), Utilisation de biomarqueurs pour la surveillance de la qualité de galloprovincialis. Am. J. Physiol. 277, 1612–1619.
l’environnement. Lavoisier Tec et Doc, Paris, New York, Londres, pp. 9–30. Viarengo, A., Lowe, D., Bolognesi, C., Fabbri, E., Koehler, A., 2007. The use of
Mendes, R., Cardoso, C., Pestana, C., 2009. Measurement of malondialdehyde in biomarkers in biomonitoring: a 2-tier approach assessing the level of
fish: a comparison study between HPLC methods and the traditional spectro- pollutant-induced stress syndrome in sentinel organisms. Comp. Biochem.
photometric test. Food Chem. 112, 1038–1045. Physiol. C 3, 281–300.
Pan, J.F., Wang, W.X., 2004. Influences of dissolved and colloidal organic carbon on Vijayavel, K., Gopalakrishnan, S., Thiagarajan, B., Thilagam, H., 2009. Immunotoxic
the uptake of Ag, Cd, and Cr by the marine mussel Perna viridis. Environ. effects of nickel in the mud crab Scylla serrata. Fish Shellfish Immunol. 26, 133–139.
Pollut. 129, 467–477. Wang, W.X., Rainbow, P., 2008. Comparative approaches to understand metal
Papachristou, P., Haralambous, K.J., Loizidou, N., Spyrellis, N., 1993. Studies bioaccumulation in aquatic animals. Comp. Biochem. Physiol. C 148 (4), 315–323.
on the nickel removal from aqueous solutions. J. Environ. Sci. Health A 28 (1), Wang, Y., Evans, D.D., 1993. lnfluence of cadmium concentration on cadmium
135–142. uptake by the freshwater mussel, Elliptio complanata. Can. J. Fish. Aquat. Sci.
Roesijadi, G., Unger, M.E., 1993. Cadmium uptake in gills of the molluscs 50, 2591–2596.
Crassostrea virginica and inhibition by calcium blockers. Aquat. Toxicol. 24, Widdows, J., Donkin, P., Staff, F.J., Matthiessen, P., Law, R.J., Allen, Y.T., Thain, J.E.,
195–206. Allchin, C.R., Jones, B.R., 2002. Measurement of stress effects (scope for growth)
Shaw, P.J.A., 2003. Multivariate Statistics for the Environmental Sciences. Hodder and contaminant levels in mussels (Mytilus edulis) collected from the Irish Sea.
Arnold editor, London, UK. Mar. Environ. Res. 53 (4), 327–356.