Content uploaded by Christophe Minier
Author content
All content in this area was uploaded by Christophe Minier on Aug 26, 2014
Content may be subject to copyright.
112
Environmental Toxicology and Chemistry, Vol. 25, No. 1, pp. 112–119, 2006
q
2006 SETAC
Printed in the USA
0730-7268/06 $12.00
1
.00
A POLLUTION-MONITORING PILOT STUDY INVOLVING CONTAMINANT AND
BIOMARKER MEASUREMENTS IN THE SEINE ESTUARY, FRANCE, USING ZEBRA
MUSSELS (
DREISSENA POLYMORPHA
)
C
HRISTOPHE
M
INIER
,*† A
LAIN
A
BARNOU
,‡ A
GNE
`S
J
AOUEN
-M
ADOULET
,† A
NNE
-M
ARIE
L
E
G
UELLEC
,‡
R
ENAUD
T
UTUNDJIAN
,† G
ILLES
B
OCQUENE
´
,§ and F
RANC¸OIS
L
EBOULENGER
†
†Laboratory of Ecotoxicology (LEMA), EA 3222, Universite´ du Havre, 25 rue Philippe Lebon, BP 540, 76058 Le Havre, France
‡Institut franc¸ais de recherche pour l’exploitation de la mer, De´partement Bioge´ochimie et Ecotoxicologie, BP 70, 29280 Plouzane´, France
§Institut franc¸ais de recherche pour l’exploitation de la mer, De´partement Bioge´ochimie et Ecotoxicologie, rue de l’Ile d’Yeu, BP 21105,
44311 Nantes Cedex 03, France
(
Received
4
March
2005;
Accepted
24
June
2005)
Abstract—Zebra mussel (
Dreissena polymorpha
) is an invasive species that has proliferated in European and North American
rivers and lakes during the last century. In this study,
D. polymorpha
has been used to provide information on contaminationlevels
and biological effects in the Seine Estuary (France). The bivalves accumulated polychlorinated biphenyls (PCBs) and polycyclic
aromatic hydrocarbons (PAHs) to a high degree with values reaching 800 ng/g dry weight for PCBs (sum of 20 congeners), and
1,000 ng/g dry weight of PAHs (sum of 14 compounds) in the whole body. These values are among the highest reported of PCBs
and, to a lesser extent, of PAHs in other contaminated areas in the world. Toxic equivalent quantities of PCBs and PAHs detected
in zebra mussels varied from 20 to 40 pg dioxin equivalents/g dry weight for PCBs and up to 120 ng benzo[
a
]pyrene equivalents/
g dry weight for PAHs, indicating a high potential risk for animals feeding on them. Biological impacts, such as altered condition
index, decreased lysosomal stability, and high levels of multixenobiotic resistance (MXR) proteins also were detected in mussels
living downstream of Rouen, the main city of the Seine Estuary. Taken together, these results indicate that the Seine Estuary is a
heavily polluted area with the potential to cause deleterious health effects in some endogenous living organisms. This study also
shows that chemical and biological measurements bring different but complementary results that can help diagnose environmental
health.
Keywords—
Dreissena polymorpha
Polychlorinated biphenyls Polycyclic aromatic hydrocarbons Multixenobiotic
resistance Biomarkers
INTRODUCTION
Ecotoxicology is a multidisciplinary research area. It is now
evident that any reliable environmental risk assessment of ex-
posure to a mixture of contaminants should involve several
disciplines such as environmental chemistry, toxicology, and
ecology. Nevertheless, in many studies, integrating all the ex-
isting data and concepts still is an unachieved goal. Environ-
mental chemists and biologists both extensively have used
mussels as living organisms for their studies. Indeed, mussels
are very useful as biomonitors to overcome difficulties in mea-
suring chemical contamination directly from water, i.e., deal-
ing with very low concentrations and fluctuations over time.
Most importantly, they also allow discrimination between the
bioavailable fraction and the unavailable forms of potentially
toxic compounds and, thus, help to consider the dose that might
affect biological activities. The rationale for using mussels as
bioindicators is that they can (to a certain extent) accumulate
contaminants proportionally to the exposure over a large range
of concentrations. They provide valuable information on the
associated potential chemical risk and, because these organ-
isms are widespread and easy to collect, mussel watch pro-
grams have been conducted for more than three decades using
marine bivalves [1,2].
The zebra mussel,
Dreissena polymorpha
, more recently
has been used as an indicator organism in various countries
* To whom correspondence may be addressed
(minier@univ-lehavre.fr).
[3,4], and its physiological and demographic properties make
it an interesting alternative for freshwater environments. This
species has spread over European rivers (and in North Amer-
ica) during the 20th century and now can be found in many
countries [5]. Zebra mussels have high filtering capacities (1
to 2 L/individual/d) [6,7] and can tolerate high concentrations
of many types of pollutants. Although they are smaller than
the blue mussels, they can be numerous in some places and
thus provide enough samples for environmental analysis.
In this study, as part of the multidisciplinary research pro-
gram Seine-Aval [8], we used zebra mussels as sentinel or-
ganisms in order to provide information on both the levels of
contamination and their associated biological effects in the
Seine Estuary. Polychlorinated biphenyls (PCBs) and poly-
cyclic aromatic hydrocarbons (PAHs) are well-known and
widespread contaminants. Concentrations of these classes of
pollutants in
Mytilus edulis
tissues collected near the mouth
of the estuary are among the highest in the European coasts
(600 ng PCB153/g dry wt) [9]. Because these contaminants
might arise from inland waters,
D. polymorpha
could be ex-
posed to high doses of these compounds, which represent a
potential threat to living organisms. Both PCBs and PAHs
usually exist as complex mixtures of numerous congeners in
environmental samples. To assess the toxicological risks of
these mixtures, it has been proposed to use the toxic equiva-
lency approach [10,11]. The toxic potency of each congener
is compared with that of a reference compound, 2,3,7,8-tet-
rachlorodibenzo-
p
-dioxin or benzo[
a
]pyrene (B
a
P), given a
Contaminant and biomarker analysis in the Seine Estuary
Environ. Toxicol. Chem.
25, 2006 113
Fig. 1. Location of sampling sites in France; pk are distances (in km)
from Paris.
value of 1. The amount of each congener in the mixture then
is multiplied by its relative potency factor, and,assuming pure-
ly additive effects, the products are added together.
Many contaminants, such as dichlorodiphenyltrichloroeth-
ane, lindane, mercury, and cadmium, as well as PAHs and
PCBs, have been detected in the Seine Estuary [12]. Because
environmental contaminants are too numerous and chemical
analysis is expensive, it is an unachievable goal to monitor all
the compounds that are present in continental or marine waters.
Biological analysis has the potential to complement this chem-
ical approach because measurements are performed directly
on organisms living in the environment. Organisms such as
zebra mussels have to cope with numerous structurally and
functionally diverse hazardous compounds. Although these or-
ganisms may be considered as adapted or resistant to contam-
inated environments, they can provide information on suble-
thal effects of the actual mixtures to which they are exposed
[13,14]. Two types of biological responses have been studied.
One is the expression of the multixenobitic resistance (MXR)
system, which corresponds to a first line of defense against
various structurally and functionally unrelated compounds
[15]. The resistant cells and organs express high molecular–
weight membrane proteins that can transport several chemicals
out of the cell. This response has the advantage of being the
result of an exposure to numerous compounds and may help
to diagnose the environmental contamination [16]. The second
type of response corresponds to biological activities or struc-
tures whose impairment may lead to dramatic effects on the
organism health. In this respect, acetylcholinesterase (AChE)
activity and lysosome membrane stability have gained rec-
ognition as important mechanisms and biomarkers [17,18].
Acetylcholinesterase activity is involved in the deactivation of
acetylcholine at nerve endings, preventing continuous nerve
firings, which are vital for normal functioning of sensory and
neuromuscular systems. Lysosomes accumulate a diverse
range of chemical contaminants that can lead to membrane
damage, resulting in leakage of their contents into the cytosol
and subsequent damage to cells [19].
This paper presents the results of a two-year monitoring
study involving chemical analysis and biomarker measure-
ments. It shows that the Seine Estuary is a highly contaminated
area based on the measured concentrations of two major classes
of persistent contaminants and on the levels of biomarkers of
pollution detected in the resident zebra mussel population.
MATERIALS AND METHODS
Study area
The Seine River, which flows into the English Channel,
drains a large industrialized and urbanized basin in France
(78,000 km
2
) hosting 40% of the French industries and 16
million people. Mean river flow is 410 m
3
/s with low waters
of less than 200 m
3
/s in summer and high waters in winter of
more than 1,500 m
3
/s. It is an estuary of the macrotidal type
with tidal amplitude varying from 3 to 8 m at its mouth and
a tidal penetration of 170 km up to the dam located at Poses
(pk 202; pk are distances in km from Paris). Salinity varies
according to tides and river flows from pk 325 onward and a
maximal turbidity zone is located between pk 330 and 350
(Fig. 1). The mean annual suspended matter flow at Poses is
6.5
3
10
5
tons and the main sewage treatment work located
in Rouen (pk 247) discharges 1 m
3
/s of secondary treated
effluents (capacity: 4.5
3
10
5
population equivalent).
Sample collection and biometric analysis
Zebra mussels, 20- to 27-mm long, were sampled by scuba
diving at 1- to 2-m depth at low tide. Six sites (Poses, Oissel,
Rouen, La Bouille, Le Landin, and Petitville) from the fresh-
water zone of the estuary and one control site located at Val
de Reuil in a nearby lake (Fig. 1) were studied in July and
September in 1997 and 1998, and a temporal study was con-
ducted at Rouen by sampling every month (in 1997) or every
two months (in 1998). Forty-two mussels were sized, dis-
sected, drained, and weighed. A condition index (CI), derived
from these data, was calculated as follows: CI (in mg/mm)
5
wet weight/length. Surface water temperature, salinity, pH,
suspended organic matter, chlorophyll, and pheopigmentswere
recorded and are part of a database of the Seine-Aval program
[20].
Chemical analysis
Fifteen mussels were depurated for 24 h in clean freshwater
(from a local source) then shucked and kept at
2
20
8
C until
further analysis. Analytical procedure for both PCBs and PAHs
was performed as described previously [21]. Briefly, it in-
cluded a solid-liquid (Soxhlet, hexane:acetone 80:20) extrac-
tion step on freeze-dried samples, followed by an adsorption
chromatography on open alumina-silica columns to remove
lipids and other unwanted organic material. Twofractionswere
obtained. The first fraction,
F
1, eluted with
n
-pentane, con-
tained PCBs that were separated further using a Cosmosil 5-
PYE high-performance liquid chromatography to isolate non-
ortho
-chloro–substituted and coplanar PCBs from
ortho
-chlo-
ro–substituted compounds. Both subfractions then were ana-
lyzed by gas chromatography (with capillary columns:
CPSil5-C18 and DB1701; J & W Scientific, Folsom, CA, USA)
coupled to an electron capture detector. The second fraction,
F
2, eluted with methylene chloride/
n
-pentane (10/90), con-
tained PAHs that were analyzed by capillary gas chromatog-
raphy (DB5) and mass spectrometry. Ionization of samples
was performed in the electron impact mode and detection was
achieved by selected ion monitoring, recording the molecular
ion for each PAH. The limit of quantification was 0.1 ng/g
dry weight for individual PCBs including the coplanar con-
geners, and between 0.02 and 0.6 ng/g dry weight for indi-
vidual PAH components. Both for PCBs and PAHs, the relative
uncertainty was less than 15% except for the lower PAHs
(fluorene, phenanthrene, anthracene), which partially were lost
during the evaporation-concentration steps. One procedural
blank and one certified reference material (either cod liver oil
Bureau Communautaire de Re´fe´rence–certified reference ma-
114
Environ. Toxicol. Chem.
25, 2006 C. Minier et al.
Table 1. Average concentration of polychlorinated biphenyls (PCBs) in zebra mussels from different
rivers and lakes worldwide
Sampling sites
PCB congeners (ng/g wet wt)
52 101 118 153 180 % Body
lipid References
Seine (La Bouille, France)
Rhine (Lobith, The Netherlands)
Meuse (Eijsden, The Netherlands)
Lake St Clair (ON, Canada)
6.5
2.9
4.8
1.1
15.2
5.5
7.2
1.7
9.7
2.1
3.1
1.3
23.5
8.4
9.1
1.7
8.9
2.7
3.8
0.7
1.2
2
2
1.8
This study
Hendriks et al. [39]
Hendriks et al. [39]
Morrison et al. [40]
Lake Erie (ON, Canada)
Mississippi (Lynxville, WI, USA)
Lake Iseo (Sarnico, Italy)
1.7
2.4
—
4.9
8.0
14.5
—
4.1
6.6
8.9
4.7
18.3
7.3
1.4
6.0
1.7
1
1
Morrison et al. [40]
Cope et al. [4]
Binelli et al. [13]
terial 349 for PCBs or marine sediment National Institute of
Standards and Technology standard reference material 1941a
for PAHs) were analyzed in each 10-sample batch. Measured
PCB congeners were PCB 31, 28, 52, 101, 110, 149, 118, 153,
132, 105, 138, 187, 128, 156, 180, 170, 194, 77, 126, 169,
and PAH compounds were fluorene (F), phenanthrene (P), an-
thracene (A), fluoranthene (Fluor), pyrene (Pyr), benz[
a
]
anthracene (B
a
A), chrysene (Chrys), benzo[
b
]fluoranthene
(B
b
F), benzo[
k
]fluoranthene (B
k
F), benzo[
e
]pyrene (B
e
P),
B
a
P, indeno[1,2,3-
cd
]pyrene (Ipyr), dibenz[
a,h
]anthracene
(D
ah
A), and benzo[
a
]perylene (Bper). The tissue lipid content
was performed gravimetrically by weighing the hexane-ace-
tone extract.
Biochemical analysis
MXR protein analysis.
Gills from 42 fresh-collected mussels
(per sampling site) were dissected, transferred in six tubes
(pools of 7 gills), and frozen immediately at
2
80
8
C. Quanti-
fication of MXR protein was performed as described in Minier
et al. [22]. Briefly, membrane-bound proteins were extracted
in 0.1% sodium dodecyl-sulfate buffer and 10
m
g of protein
samples were loaded onto nitrocellulose membranes. The C219
monoclonal antibody (Abcys, Paris, France) was used as pri-
mary antibody (1
m
g/ml) and revealed using alkaline-phos-
phatase–conjugated secondary antibodies. Semiquantitation of
the resulting staining was carried out by image analysis
(ImageMasterTotalLab, Amersham, Piscataway, NJ, USA) and
intercalibration of the different membranes was performed us-
ing positive controls corresponding to extracts (4 replicates)
of a pool of mussels sampled in Rouen and of a human cancer
cell line (MCF7 cells).
AChE activity.
The AChE activity was assessed as described
by [23]. Briefly, six pools of seven gills (42 mussels per site)
were homogenized and centrifuged for 20 min at 10,000
g.
The AChE activity of 100-
m
l supernatant in the presence of
0.01 M dithionitrobenzoic acid and 2.8 mM acetylthiocholine
then was measured at 412 nm using a microplate reader. Both
MXR proteins and AChE activity assessments were standard-
ized to the protein content of the extracts. Protein concentra-
tions were measured using the Bradford method [24] with
bovine serum albumin as a standard.
Lysosomal stability.
Hemocytes were withdrawn from the
posterior adductor muscle of six mussels using a hypodermic
syringe and transferred to a physiological saline solution (20
mM
N
-2-hydroxyethyl piperazine-
N
9
-2-ethane sulfonic acid,
150 mM NaCl, 10 mM KCl, 10 mM CaCl
2
, pH 7.5). In a
humidity chamber, 40
m
l of cell suspensions were allowed to
attach to poly-lysine–treated microscope slides for 15 min in
the dark at 15
8
C. Excess suspension was removed from the
slide and the cell monolayers were incubated in physiological
saline buffer containing 40
m
g/ml neutral red (Sigma, Lyon,
France) in the humidity chamber for 60 min. Slides then were
removed and cells were examined under light microscope for
abnormal morphology of haemocytic lysosomes and lysosomal
integrity. Thirty granular haemocytes were observed and those
with marked increased lysosomal volume or cytosolic staining
(indicative of lysosomal membrane breakdown) were recorded.
Statistical analysis
Statistical analysis was performed using the SYSTAT pro-
gram (Systat Software, Richmond, CA, USA). Dependence of
chemical and biochemical data on sampling site or sampling
time was tested for statistical significance by one-way analysis
of variance. Correlation analysis was performed using the
Pearson product moment correlation coefficient.
RESULTS
PCB and PAH body burdens
High PCB and PAH concentrations were measured in
D.
polymorpha
in the Seine Estuary. The sum of all the measured
PAHs (14 compounds) was above 1,000 ng/g dry weight of
mussel tissues. The concentrations of chrysene, the PAH com-
ponent found in highest concentrations in this study, were in
the range of 300 to 700 ng/g dry weight of tissues. Total PCB
concentrations (sum of 20 congeners) generally exceeded 800
ng/g dry weight. The two main PCBs congeners measured
within the whole organism were PCB 153 and PCB 138, and
their body burdens varied between 100 and 180 ng/g dry
weight for PCB 153 and between 70 and 140 ng/g dry weight
for PCB 138. These values always were significantly higher
than those found in tissues of zebra mussels collected in the
nearby lake taken as reference site (located at Val de Reuil)
for both PAHs and PCBs. Comparison with other sites world-
wide shows that zebra mussels from the Seine Estuary are
more contaminated by PCBs than those from other investigated
areas (Table 1). In contrast to PCBs, the PAH body burden of
zebra mussels from the Meuse River is four times that found
in the Seine Estuary mussels (Table 2).
Only small and nonsignificant variations of PCB body bur-
den, as exemplified by PCB 153 concentrations, were observed
between sites of collection within the Seine Estuary (Fig. 2).
Nevertheless, concentrations tended to be higher in Poses and
La Bouille. This could indicate that those compounds origi-
nated from the upper part of the river (allowing a greater
amount of PCBs in organisms living upstream, i.e., at Poses)
and that another source possibly was located in the industrial
area of Rouen (resulting in the small increased PCBs body
burden measured in La Bouille). The temporal study of PCB
body burden performed in zebra mussels collected at Rouen
Contaminant and biomarker analysis in the Seine Estuary
Environ. Toxicol. Chem.
25, 2006 115
Table 2. Average concentration of individual and total polycyclic aromatic hydrocarbons (PAHs) in zebra mussels from different rivers
River
PAH concn. (ng/g wet wt)
a
AB
a
AB
a
PB
e
P Chrys Fluor Pyr Total PAHs References
Seine (France)
Rhine (The Netherlands)
Meuse (The Netherlands)
2
1
21
28
20
250
21
6
15
30
13
55
54
17
65
29
1
250
36
12
120
290
137
1,189
This study
Hendriks et al. [39]
Hendriks et al. [39]
a
A
5
anthracene; B
a
A
5
benz[
a
]anthracene; B
a
P
5
benzo[
a
]pyrene; B
e
P
5
benzo[
e
]pyrene; Chrys
5
chrysene; Fluor
5
fluoranthene; Pyr
5
pyrene.
Fig. 2. Zebra mussel tissue concentrations of the polychlorinated bi-
phenyl congener 153 and chrysene in the Seine Estuary (Rouen,
France; mean of values obtained in June and September 1997 and
1998). Error bars represent confidence intervals (
a5
0.05,
n
5
4).
Fig. 3. Seasonal variations of the polychlorinated biphenyl congener
153 (solid bar: In ng/g dry wt; solid line: In ng/g lipid wt) and chrysene
body burden (in ng/g dry wt) in
Dreissena polymorpha
sampled in
Rouen, France (pools of 15 mussels).
during two years (1997–1998) did not reveal any significant
trend during the first year, whereas in 1998, a bell-shape curve
was observed with a rise in tissue concentrations during the
spring and a subsequent decrease during the autumn of the
second year (Fig. 3). These PCB levels were not correlated
with lipid content (
r
5
0.41,
p
.
0.1,
n
5
11) but with pheo-
phytin (
r
5
0.65,
p
5
0.03,
n
5
11) and suspended particulate
matter (
r
5
0.54,
p
5
0.08,
n
5
11) concentrations in water
samples, indicating that PCB body burden might be influenced
significantly by filtration rate and algae consumption.
Significant variations of PAH body burdens were observed
along the six sampling sites from the Seine Estuary (
p
,
0.001,
n
5
6). The PAH concentrations were 50% lower in Le Landin
and Petitville when compared to the other sites located up-
stream (Fig. 2). Even if the difference was not statistically
significant, La Bouille always was recorded as the site where
the highest tissue concentrations were determined in zebra
mussels, thus indicating the existence of a source of contam-
ination immediately downstream of Rouen. Monthly mea-
surements performed at Rouen showed that highest concen-
trations were found between March and June for each for the
two years of study (Fig. 3), but there also was evidence that
high variations (25–50% increase or decrease) in PAH body
burdens could arise in a relatively short period of time, i.e.,
from one month to the other. Noticeably, PAH and PCB con-
centrations varied in a similar way during the second year of
the study (
r
5
0.76,
p
5
0.01,
n
5
7).
Fingerprint analysis
The fingerprint of the 16 PCB congeners measured in the
mussel body and normalized against the main congener, PCB
153, is given in Figure 4. The data indicated that, whatever
the site or time of sampling, the contaminant pattern was sim-
ilar and characterized by the presence of high concentrations
of the PCB congeners 153, 138, 101, and 149. These finger-
prints were nearly identical to those observed in the blue mus-
sel,
M. edulis
, at the mouth of the Seine Estuary, but different
from the one obtained in zebra mussel from the reference site,
which showed a lack of lower chlorinated PCBs (Fig. 5) and
a different ratio of specific congeners such as PCB 101/PCB
118.
For PAHs, the fingerprint was normalized against totalPAH
body burdens (sum of 14 measured compounds). Patterns were
very similar in the four upstream sites with chrysene, pyrene,
benzo[
a
]anthracene, fluorine, and benzo[
e
]pyrene, respective-
ly, the compounds found at the highest level. The pattern dif-
fered in Le Landin and Petitville and was dominated by
benzo[
e
]pyrene, chrysene, benzo[
b
]fluoranthene, phenan-
threne, and pyrene. The distribution between aromatic classes
was site-dependent (Fig. 6). A decrease in the four-ring PAHs
at La Bouille became more pronounced at sites further down-
stream. The four-ring PAHs gradually were replaced by others,
mainly five-ring PAHs.
Calculation of specific compounds ratio [25,26] did not
provide a clear-cut partition as to whether or not the PAHs
were of pyrogenic or petrogenic origin. Nevertheless, pyrene/
anthracene, chrysene/benzo[
a
]anthracene, and benzo-
[
e
]pyrene/benzo[a]pyrene ratios were greater in Le Landin and
Petitville, suggesting a predominance of petrogenic com-
pounds in these sites (Table 3).
Quantification of PAH- and PCB-associated toxicity
In order to assess potential effects of both PCBs and PAHs,
toxic equivalents (TEQ) were estimated for these mixtures.
With toxic equivalent factors taken from [10,27], dioxin TEQ
(2,3,7,8-tetrachlorodibenzo-
p
-dioxin toxicity equivalent) were
evaluated for two di-
ortho
–substituted (PCB congeners 170
and 180), three mono-
ortho
–substituted (PCB congeners 105,
118, and 156), and three non-
ortho
–substituted congeners
(PCB congeners 77, 126, and 169). Results showed that TEQ
varied between 0.02 and 0.03 ng/g dry weight according to
116
Environ. Toxicol. Chem.
25, 2006 C. Minier et al.
Fig. 4.
Dreissena polymorpha
polychlorinated biphenyl (PCB) fingerprints in sampling sites along the Seine Estuary (Rouen, France).
Fig. 5. Polychlorinated biphenyl (PCB) relative concentrations in ze-
bra mussels from the Seine Estuary (Rouen, France) and a nearby
lake (Val de Reuil, France) and from blue mussels collected at the
mouth of the Seine Estuary (Le Havre, France, from Abarnou et al.
[9]). Fig. 6. Polycyclic aromatic hydrocarbon (PAH) relative concentrations
in zebra mussels collected in the Seine Estuary (Rouen, France).
site or time of sampling (Table 4). Noticeably, PCB congeners
118 and 126 were by far the main contributors to the calculated
dioxin equivalent (15–30% and 40–60%, respectively).
For PAHs, TEQ were expressed as B
a
P equivalent [11].
The assessed toxicity was highest at La Bouille with values
exceeding 120 ng/g wet weight. Benzo[
a
]pyrene itself was
contributing to more than 60% of the calculated toxicity.
Finally, using the approach described by Willett et al. [28],
induction equivalents were calculated to compare PAH and
PCB’s potency to induce 7-ethoxyresorufin-
O
-deethylase ac-
tivity. Results showed that PAHs contributed to more than 95%
of the obtained values that varied between 0.3 and 0.8 ng
dioxin equivalent/g dry weight.
Biological responses
The MXR protein content in mussels from the Seine Estuary
was double that of the reference site, indicating a higher ex-
posure to pollutants in the river sites (Fig. 7a). In mussels
sampled at Rouen, these proteins reached their highest con-
centrations in summer and decreased to low concentrations in
winter (
F
5
8.42,
p
,
0.001,
n
5
6; Fig. 8). These seasonal
variations, similar to the ones previously reported for the blue
mussel [29], correlated with water temperature (
r
5
0.72,
p
5
0.01,
n
5
16) and chlorophyll
a
content (
r
5
0.46,
p
5
0.07,
n
5
16), suggesting that the MXR protein concentration
might be related to filtering rates (which are temperature-de-
pendent) and food supply. When comparing sampling sites
along the estuary, values tended to be higher in downstream
sites (from Rouen to Petitville), with La Bouille and Petitville
having significantly higher amounts of MXR proteins than any
other site (
p
,
0.05,
n
5
6).
Similar discrepancies were observed between mussels liv-
ing upstream or downstream Rouen considering the condition
index (
F
5
58.8,
p
,
0.001,
n
5
42; Fig. 7b). Mussels from
La Bouille, Le Landin, and Petitville had lower relative
weights than those from the other sites within the estuary,
although values were reduced significantly only in La Bouille
and Petitville. Furthermore, CI of animals leaving in the Seine
also were reduced dramatically (by 50%) when compared to
CI of mussels living in the reference site.
Measurements of AChE activity in gills were performed
the first year of the study. However, results showed that this
activity was very low regardless of the site or date of sampling.
The values were below 1 nmole/min/mg protein and close to
detection limits of the assay. The second year, assessment of
lysosome stability then was performed. Results showed that
membrane stability was affected significantly in haemocytes
of mussels from La Bouille and Petitville compared to mussels
from the reference site and river sites upstream Rouen (
F
5
6.7,
p
,
0.001,
n
5
6; Fig. 7c).
DISCUSSION
This study reports a first attempt to combine chemical and
biological measurements using zebra mussel,
D. polymorpha
,
as indicator organism in the Seine Estuary. The study showed
that, like other bivalves such as the marine mussel (
M. edulis
),
D. polymorpha
do concentrate lipophilic compounds such as
Contaminant and biomarker analysis in the Seine Estuary
Environ. Toxicol. Chem.
25, 2006 117
Table 3. Ratio between structural isomers (mean of ratios obtained in June, July, and September 1997
and 1998) of polycyclic aromatic hydrocarbons (PAHs) in zebra mussels from the SeineEstuary (France)
with confidence interval in brackets (
a5
0.05)
Station
a
P/A
b
Fluor/Pyr
c
Chrys/B
a
A
d
B
e
P/B
a
P
e
Poses
Oissel
Rouen
La Bouille
Le Landin
Vieux Port
7.33 (2.98)
9.875 (4.38)
10.025 (3.33)
14.65 (9.71)
14.25 (0.07)
19.8 (10.27)
0.67 (0.06)
0.65 (0.06)
0.675 (0.05)
0.675 (0.28)
0.45 (0.07)
0.625 (0.12)
2.57 (0.20)
2.775 (0.17)
2.475 (0.15)
2.35 (0.63)
6.2 (0.42)
4.425 (0.51)
3.47 (1.73)
3.1 (1.13)
2.65 (0.72)
2.375 (1.14)
3.5 (1.25)
3.6 (1.31)
a
Station location sites given in Table 1.
b
P/A
5
ratio of phenanthrene concentration to anthracene concentration.
c
Fluor/Pyr
5
ratio of fluoranthene concentration to pyrene concentration.
d
Chrys/B
a
A
5
ratio of chrysene concentration to benz[
a
]anthracene concentration.
e
B
e
P/B
a
P
5
ratio of benz[
e
]pyrene concentration to benz[
a
]pyrene concentration.
Table 4. Toxic equivalents quantities (TEQs) for, and induction equivalents (IEs) of polychlorinated
biphenyl (PCB) and polycyclic aromatic hydrocarbon (PAH) concentrations assessed in zebra mussel
tissues collected in various sites along the Seine Estuary (France) in September 1998. Toxic equivalent
factors were taken from [11,27,28]
Sampling sites
a
Poses Oissel Rouen La Bouille Le Landin Petitville
TEQ PCBs (pg dioxin
equivalent/g dry wt) 27.7 30.6 23.8 28.6 21.8 26.4
TEQ PAHs (ng ben-
zo[
a
]pyrene equiva-
lent/g dry wt) 81.8 75.0 80.5 126.8 40.2 48.4
IE (pg dioxin equiva-
lent/g dry wt) 597 568 560 802 306 350
a
Sampling site locations given in Table 1.
PCBs and PAHs to a high extent with values reaching 800 ng/
g dry weight for PCBs (sum of 20 congeners) and 1,000 ng/
g dry weight of PAHs (sum of 14 compounds) in the whole
body. These values are among the highest reported in other
contaminated areas in the world for PCBs and, to a lesser
extent, for PAHs (Tables 1 and 2). Although comparison be-
tween two species should be taken with caution, it can be
observed that similar PCB concentrations have been reported
in
M. edulis
tissues living near the mouth of the Seine Estuary
[9]. Furthermore, relative PCB concentrations were nearly
identical in the two bivalve species collected either in the Seine
Estuary (zebra mussels) or at the mouth of the Seine Estuary
(blue mussels) [21], thus indicating common sources of PCB
contaminants. On the contrary, PAH body burdens were 25
times higher in zebra mussels than in
M. edulis
[2]. This dif-
ference might be explained by the animals’ physiology.
Dreis-
sena polymorpha
generally show higher bioaccumulation ca-
pacities than
M. edulis
[30]. Another explanation would be
that the two species were exposed to different sources and
concentrations of PAHs. Our results suggested that large
amounts of PAHs could be released from industries just down-
stream the city of Rouen, thus contaminating the endogenous
zebra mussels.
Seasonal variations of total PCBs and PAHs were recorded.
Values generally tended to be higher during spring and sum-
mer. Interestingly, similar variations of MXR protein concen-
trations in gill tissues were measured. It generally is accepted
that bioaccumulation in zebra mussels is related to the octanol/
water partition coefficient and also to the contamination of
sediments rather than water [31,32]. Thus, sediment, partic-
ulate organic matter, and algae might be potential sources of
contamination for zebra mussels. This indeed was observed
in this study, as analysis of seasonal variations of PCBcontent
in zebra mussels were correlated to water pheopigments and
suspended matter concentration. High filtering rates increase
the probability that populations of zebra mussels will be ex-
posed to a wide range of pollutants, including hydrophobic
contaminants such as PCBs and PAHs. As this filteringactivity
is affected by temperature and food availability [30,31], zebra
mussels might accumulate more contaminants during the pe-
riod of elevated temperature and algal growth. This increase
in contaminant exposure, in turn, might have led to induction
of the MRX system. The expression of MXR protein is induced
by many lipophilic compounds [15] and, in this study, their
seasonal variations were correlated to water chlorophyll
a
con-
centrations and temperature. Taken together, these results in-
dicate that suspended matter and food might be major con-
tamination sources for mussels in the Seine Estuary.
Calculation of toxic equivalent quantities associated with
PCBs and PAHs detected in zebra mussels might indicate po-
tential effects mediated by the aryl hydrocarbon receptor. Our
results showed that TEQ associated with the PCBs varied from
20 to 30 pg dioxin equivalents/g dry weight, which is more
than the concentration limit recommended in any human food
in Europe. The PCB congeners 118 and 126 were the main
contributors to the calculated dioxin equivalent despite their
low concentrations in mussel tissue. The TEQ associated with
PAHs were up to 120 ng B
a
P equivalents/g dry weight for
mussels living immediately downstream Rouen at La Bouille,
corroborating previous results from Munschy et al. [33], who
reported that the sediment B
a
P concentrations per g of dried
material were high at La Bouille and exceeded 2 mg/g. These
118
Environ. Toxicol. Chem.
25, 2006 C. Minier et al.
Fig. 7. Multixenobiotic (MXR) protein expression levels (A), con-
dition index (B), and lysosomal stability (C) in zebra mussels collected
in the Seine Estuary (Rouen, France). Error bars represent confidence
intervals (
a5
0.05,
n
5
6 for MXR proteins;
n
5
42 for condition
index;
n
5
6 for lysosomal stability).
Fig. 8. Seasonal variations of multixenobiotic (MXR) protein ex-
pression in zebra mussels sampled in Rouen, France (March 1997–
January 1999). Error bars represent confidence intervals (
a5
0.05,
n
5
6).
results also are in accordance with previous works that showed,
in suspended matter, concentrations of PCBs within the whole
estuary and of PAHs at sites downstream Rouen were higher
than the acceptable limit effect defined by the Ospar conven-
tion [34]. The potential mutagenicity of PAHs is evidenced
further by the work of Boillot et al. [35] who reported that
flounder (
Platichthys flesus
) sampled in La Bouille had high
amounts of DNA adducts in the liver.
The biological responses of
D. polymorpha
living in the
estuary are consistent with the potential (calculated) effects of
their body burdens of contaminants. Our results showed that
the health and development of mussel collected in the river
sites, as assessed by their condition index, were poor compared
to those of mussels living in a nearby lake taken as reference
site. Other differences in markers related to mussel health also
were evident between sites within the estuary. They showed
that, not only the site of La Bouille, but also the other sites
downstream of the city of Rouen, had reduced growth leading
to decreased relative weight, higher MXR protein content, and
altered lysosomal membrane stability. Because mussels have
few biotransformation capacities, these results indicate that
adverse effects could arise from other mechanisms of toxicity
and probably from other compounds than PAHs and PCBs. In
this study, no specific mode of action was searched for (except
for the AChE activity, which did not provide any significant
result). Measurements of lysosomal stability and MXR protein
expression allow monitoring of biological effects that can be
due to a variety of compounds. They provide results comple-
mentary to the chemical analyses as they integrate potential
multiple mechanisms of action of compounds that are present
as complex mixtures. Zebra mussels also have been used to
give more specific information on effects of contaminants,
such as endocrine-disrupting effects in other areas, including
European and American rivers and lakes [13,14]. Similar anal-
yses should be performed in the Seine, because intersex fish
(flounder, roach, and gudgeon) have been observed in this area
[36,37].
CONCLUSION
By combining chemical and biological measurements, this
study provides concordant information on the presence of con-
taminants and biological consequences. It shows that the Seine
Estuary is highly contaminated by lipophilic compounds such
as PAHs and PCBs that bioaccumulate in zebra mussel tissues
and might have measurable and significant biological effects
on living organisms feeding on mussels. The calculation of
TEQ indicated that effects might arise, especially for verte-
brates that have high metabolizing capacities. The decrease in
fish-population density, which has led to the disappearance of
any professional fishing activity in the Seine Estuary since the
1970s [38], at least partly might be related to such toxic effects.
Even zebra mussels that might be considered as resistant or-
ganisms showed altered condition index, decreased lysosomal
stability, and high levels of resistance proteins in the Seine
Estuary.
Acknowledgement
—This study was conducted under the Seine Aval
research program, with financial support from the Agence de l’Eau
Seine-Normandie and the Re´gion Haute Normandie. The authors are
grateful to A. Ficht for help in physical and chemical measurements.
A. Jaouen-Madoulet and R. Tutundjian were recipients of a doctoral
grant from the Re´gion Haute Normandie Council. Special thanks are
due to E.M. Hill for reviewing the manuscript.
REFERENCES
1. Goldberg ED, Bowen VT, Farrington JW, Harvey GR, Martin JH,
Parker PL, Risebrough RW, Roberton WE, Schneider E, Gamble
E. 1978. The mussel watch.
Environ Conserv
5:101–125.
Contaminant and biomarker analysis in the Seine Estuary
Environ. Toxicol. Chem.
25, 2006 119
2. Claisse D. 1989. Chemical contamination of French coasts: The
results of ten years mussel watch.
Mar Pollut Bull
20:523–528.
3. Mersch J, Jeanjean A, Spor H, Pihan J-C. 1992. The freshwater
mussel
Dreissena polymorpha
as a bioindicator for trace metals,
organochlorines, and radionuclides. In Nalepa TF, Schloesser DW,
eds,
The Zebra Mussel, Dreissena polymorpha
. Gustav Fischer
Verlag, New York, NY, USA, pp 245–254.
4. Cope WG, Bartsch MR, Rada RG, Balogh SJ, Rupprecht JE,
Young RD, Johnson DK. 1999. Bioassessment of mercury, cad-
mium, polychlorinated biphenyls, and pesticides in the upper Mis-
sissippi river with zebra mussels (
Dreissena polymorpha
).
En-
viron Sci Technol
33:4385–4390.
5. Johnson LE, Carlton JT. 1996. Postestablishment spread in large-
scale invasions: Dispersal mechanism of the zebra mussel
Dreis-
sena polymorpha
.
Ecology
77:1690–1697.
6. Akopian M, Garnier J, Testard P, Ficht A. 2001. Estimating the
benthic population of
Dreissena polymorpha
and its impact in
the lower Seine River, France.
Estuaries
24:1003–1014.
7. Berny P, Lachaux O, Buronfosse T, Mazallon M, Gillet C. 2002.
Zebra mussels (
Dreissena polymorpha
) as indicators of fresh-
water contamination with lindane.
Environ Res
90:142–151.
8. Lafite R, Romanˇa L-A. 2001. A man-altered microtidal estuary:
The Seine Estuary (France), introduction to special issue.
Estu-
aries
24:939.
9. Abarnou A, Loizeau V, Le Guellec AM, Jaouen-Madoulet A.
2002. Contaminants in marine food webs.
Rev Med Vet
153:425–
432.
10. Van den Berg M, Birnbaum L,Bosveld BTC, Brunstro¨ mB, Cook
P, Feeley M, Giesy JP, Hanberg A, Hasegawa R, Kennedy SW,
Kubiak T, Larsen JC, van Leeuwen FXR, Liem AKD, Nolt C,
Peterson RE, Poellinger L, Safe S, Schrenk D, Tillitt D, Tysklind
M, Younes M, Waern F, Zacharewski T. 1998. Toxic equivalency
factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wild-
life.
Environ Health Perspect
106:775–792.
11. Nisbet C, LaGoy P. 1992. Toxic equivalency factors (TEFs) for
polycyclic aromatic hydrocarbons (PAHs).
Regul Toxicol Phar-
macol
16:290–300.
12. Chevreuil M, Blanchard M, Teil MJ, Carru AM, Testard P, Ches-
terikoff A. 1996. Evaluation of the pollution by organochlorinated
compounds (polychlorobiphenyls and pesticides) and metals (Cd,
Cr, Cu, and Pb) in the water and in the zebra mussel (
Dreissena
polymorpha
Pallas) of the river Seine.
Water Air Soil Pollut
88:
371–381.
13. Binelli A, Bacchetta R, Vailati G, Galassi S, Provini A. 2001.
DDT contamination in Lake Maggiore (N. Italy) and effects on
zebra mussel spawning.
Chemosphere
45:409–415.
14. Quinn B, Gagne´ F, Costello M, McKenzieC, Wilson J, Mothersill
C. 2004. The endocrine-disrupting effect of municipal effluent on
the zebra mussel (
Dreissena polymorpha
).
Aquat Toxicol
66:
279–292.
15. Bard SM. 2000. Multixenobiotic resistance as a cellular defense
mechanism in aquatic organisms.
Aquat Toxicol
48:357–389.
16. Minier C, Eufemia F, Epel DE. 1999. The multixenobiotic resis-
tance phenotype as a tool to biomonitor the environment.
Bio-
markers
4:442–454.
17. Ringwood AH, Hameedi MJ, Lee RF, Brouwer M, Peters EC,
Scott GI, Luoma SN, Digiulo RT. 1999. Bivalve biomarker work-
shop: Overview and discussion group summaries.
Biomarkers
4:
391–399.
18. Brown M, Davies IA, Moffat CF, Redshaw J, Craft JA. 2004.
Characterization of choline esterases and their tissue and sub-
cellular distribution in mussel (
Mytilus edulis
).
Mar Environ Res
57:155–169.
19. Lowe DM, Fossato VU. 2000. The influence of environmental
contaminants on lysosomal activity in the digestive cells of mus-
sels (
Mytilus galloprovincialis
) from the Venice Lagoon.
Aquat
Toxicol
48:75–85.
20. Hocde´ R. 2002. Cd-Rom de restitution Seine-Aval. Rapports
d’avancement 2001. Technical report. Seine Aval. Rouen, France.
21. Jaouen-Madoulet A, Abarnou A, Le Guellec A-M, Loizeau V,
Leboulenger F. 2000. Validation of an analytical procedure for
polychlorinated biphenyls, coplanar polychlorinated biphenyls,
and polycyclic aromatic hydrocarbons in environmental samples.
J Chromatogr
886:153–173.
22. Minier C, Akcha F, Galgani F. 1993. P-glycoprotein expression
in
Crassostrea gigas
and
Mytilus edulis
in polluted seawater.
Comp Biochem Physiol
106:1029–1036.
23. Bocquene´ G, Galgani F. 1998. Biologicaleffects of contaminants:
Cholinesterase inhibition by organophosphate and carbamate
compounds.
International Council for the Exploration of the Sea
(ICES) Techniques in Marine Environmental Sciences
22:1–12.
24. Bradford MM. 1976. A rapid and sensitive method for the quan-
titation of microgram quantities of protein utilizing the principle
of protein-dye binding.
Anal Biochem
72:248–254.
25. Sicre MA, Marty JC, Saliot A, Aparicio X, Grimalt J, Albaiges
J. 1987. Aliphatic and aromatic hydrocarbons in different-sized
aerosol over the Mediterranean Sea: Occurrence and origin.
Atmos
Environ
21:2247–2259.
26. Parlanti E. 1990. Utilisation des hydrocarbures comme traceurs
d’origine de la matie`re organique se´dimentaire en milieu marin.
PhD thesis. University of Bordeaux, Bordeaux, France.
27. Ahlborg UG, Becking GC, Birnbaum LS, Brouwer A, Derks
HJGM, Feeley M, Golor G, Hanberg A, Larsen JC, Liem AKD,
Safe SH, Schlatter C, Waern F, Younes M, Yrja¨nheikki E. 1994.
Toxic equivalency factors for dioxin-like PCBs: Report on WHO-
ECEH and IPCS consultation, December 1993.
Chemosphere
28:
1049–1067.
28. Willett KL, Gardinali PR, Sericano JL, Wade TL, Safe SH. 1997.
Characterization of the H4IIE rat hepatoma cell bioassay for the
evaluation of environmental samples containing polynuclear ar-
omatic hydrocarbons (PAHs).
Arch Environ Contam Toxicol
32:
442–448.
29. Minier C, Borghi V, Moore MN, Porte C. 2000. Seasonal variation
of MXR and stress proteins in the common mussel,
Mytilus gal-
loprovincialis
.
Aquat Toxicol
50:167–176.
30. Fisher SW, Gossiaux DC, Bruner KA, Landrum PF. 1993. Inves-
tigation of the toxicokinetics of hydrophobic contaminants in the
zebra mussel (
Dreissena polymorpha
). In Nalepa TF, Schloesser
DW, eds,
Zebra Mussels: Biology, Impacts and Control.
Lewis,
Boca Raton, FL, USA, pp 465–490.
31. Gossiaux DC, Landrum PF, Fisher SW. 1996. Effect of temper-
ature on the accumulation kinetics of PAHs and PCBs in the zebra
mussel,
Dreissena polymorpha
.
J Gt Lakes Res
22:379–388.
32. Metcalfe CD, Metcalfe TL, Riddle G, Douglas-Haffner G. 1997.
Aromatic hydrocarbons in biota from the Detroit River and west-
ern Lake Erie.
J Gt Lakes Res
23:160–168.
33. Munschy C, Moisan K, Truquet I, Tronczynski J. 1997. Carac-
te´risation chimique de la contamination organique dans l’estuaire
de la Seine. In
Rapport d’Activite´ Seine-Aval, The`me Dynamique
des Contaminants.
Institut franc¸ais de recherche pour
l’exploitation de la mer, Issy-les-Moulineaux, France, pp 41–60.
34. Long ER, Mac Donald DD, Smith SL, Calder FD. 1995. Incidence
of adverse effects within range of chemical concentrations in
marine and estuarine sediments.
Environ Manag
19:81–97.
35. Boillot K, Burgeot T, Arnould JP, Pfhol-Leszkowicz A. 1997.
DNA adduct detection in the liver of the dragonnet
Callionymus
lyra
and the flounder
Platichthys flesus
.
Mutat Res
379:S116.
36. Minier C, Levy F, Rabel D, Bocquene´ G, Godefroy D, Burgeot
T, Leboulenger F. 2000. Flounder health status in the Seine Bay.
A multibiomarker study.
Mar Environ Res
50:373–377.
37. Minier C, Caltot G, Leboulenger F, Hill EM. 2000. An investi-
gation of the incidence of intersex fish in Seine-Maritime and
Sussex regions.
Analusis
28:801–806.
38. Costil K, Dauvin J-C, Duhamel S, Hocde´ R, Mouny P, de Roton
G. 2002.
Patrimoine biologique et chaıˆnes alimentaires.
Institut
franc¸ais de recherche pour l’exploitation de la mer, Plouzane´,
France, pp 47.
39. Hendriks AJ, Pieters H, de Boer J. 1998. Accumulation of metals,
polycyclic (halogenated) aromatic hydrocarbons, and biocides in
zebra mussel and eel from the Rhine and Meuse rivers.
Eviron
Toxicol Chem
17:1885–1898.
40. Morrison H, Yankovitch T, Lazar R, Haffner GD. 1995. Elimi-
nation rate constants of 36 PCBs in zebra mussels (
Dreissena
polymorpha
) and exposure dynamics in the Lake St. Clair—Lake
Erie corridor.
Can J Fish Aquat Sci
52:2574–2582.
