Arch Environ Contam Toxicol (2009) 56:209–220

DOI 10.1007/s00244-008-9176-8

Effects of the Erika Oil Spill on the Common Starfish Asterias

rubens, Evaluated by Field and Laboratory Studies
Guillemette Joly-Turquin Æ Philippe Dubois Æ Geoffroy Coteur Æ
Bruno Danis Æ Sandra Leyzour Æ Karine Le Menach Æ Hélène Budzinski Æ
Monique Guillou

Received: 3 October 2007 / Accepted: 7 April 2008 / Published online: 6 May 2008
Ó Springer Science+Business Media, LLC 2008

Abstract Impacts of the Erika oil spill on the common Introduction

starfish Asterias rubens were investigated in the field and
using laboratory experiments based on contamination via In December 1999, the oil tanker Erika sank off the coast of
food at different stages of the starfish reproductive cycle. southern Brittany (France), breaking in two and releasing
Two months after the shipwreck, levels of hydrocarbons approximately 19,000 metric tons of heavy fuel oil (No. 2 or
characteristic of Erika fuel were significantly higher in 6 according to the French or British classification, respec-
pyloric ceca and body wall of A. rubens from a contaminated tively) along 400 km of the French Atlantic coast (Laubier
site, compared with control animals from an unpolluted et al. 2004) (Fig. 1). The substance released was not a true
reference area. Concomitant immunological responses and crude oil, but a residue of petroleum transformation char-
detoxification enzyme activity (CYP1A) were enhanced in acterized by high molecular weight polycyclic aromatic
the impacted starfish, suggesting rapid biotransformation hydrocarbons (PAHs), a specific gravity similar to that of
processes. This was confirmed by laboratory experiments seawater, and a high viscosity (20,000 cSt at 10°C). It was
which showed a fast PAH uptake during the 10 first days of composed of about 50% PAHs, with a high predominance of
contamination and the start of biotransformation processes chrysene, phenanthrene, pyrene, and benzo(a)anthracene
from the third day. Our study confirms benzo(a)pyrene (INERIS; National Institute for Environment and Industrial
hydroxylase activity (BPH) in A. rubens and demonstrates Risk [Diderich 2000]). The main oil discharges from the
the influence of CYP1A in the conversion of insoluble PAHs wrecked tanker occurred between 24 and 27 December and
into soluble derivatives in this species for the first time. The fouled the upper parts of the shorelines of Loire Atlantique
rapidity of decontamination could explain why starfish and Vendée counties. In order to study the ecological and
growth, level of motile activity, reproductive investment, ecotoxicologic consequences of this oil spill, the French
energy storage, and larval development were not signifi- Ministry of Ecology and Sustainable Development launched
cantly affected by these contaminants. a monitoring program, managed by IFREMER (French
Research Institute for the Exploitation of the Sea) and
INERIS. Two to three months after the oil spill, PAHs were
found in every compartment of the environment, the main
G. Joly-Turquin
P. Dubois
G. Coteur
B. Danis
Laboratoire de Biologie Marine (CP 160/15), Université Libre de types detected being alkyl-substituted phenanthrenes, pyr-
Bruxelles, 50 Av. F.D. Roosevelt, B-1050 Bruxelles, Belgium enes, and chrysenes and heterocyclic sulfur compounds such
as dibenzothiphenes and their alkyl derivatives (Tronczynski
G. Joly-Turquin
S. Leyzour
M. Guillou (&)
et al. 2004). Moreover, evidence of contamination associ-
Laboratoire de l’Environnement Marin (UMR 6539), Université
de Bretagne Occidentale, Institut Européen de la Mer, place N. ated with Erika oil was still detectable in water and
Copernic, F-29780 Plouzané, France suspended particulate matter in May 2002. However, the
e-mail: mguillou@univ-brest.fr shallow subtidal sediments exhibited little sign of oil con-
tamination at any time.
K. Le Menach
H. Budzinski
LPTC UMR CNRS 5472, Université de Bordeaux 351, crs de la Mass mortalities of the starfish species Asterias rubens
Libération, 33405 Talence, France and Marthasterias glacialis were observed on the highly

123
210 Arch Environ Contam Toxicol (2009) 56:209–220

1996). However, as these field studies dealt with density

measurements, it is difficult to dissociate effects of the oil
from those of other factors like postspill shoreline treat-
ments, the use of dispersants, or even natural cycles.
Previous experimental studies have mostly been focused on
the impact of the water accommodated fraction (WAF) or
water-soluble fraction (WSF) of crude oil on the survival,
behavior, or feeding of predatory starfish including Aste-
rias rubens (Georgiades et al. 2003; O’Clair and Rice
1985; Ordzie and Garofalo 1981; Temara et al. 1999;
reviewed by O’Clair and Rice 1985). Locomotive function,
prey localization ability, feeding rates, and growth have all
been found to be reduced at low oil concentrations, even
though sensitivity was shown to vary with the type of oil
and/or starfish species. Ability to avoid oiled sediment was
also demonstrated in one bottom-dwelling starfish (Ryder
et al. 2004).
The present study, therefore, aimed not only to quantify
the effects of the Erika oil spill on biomarker expression in
Asterias rubens, i.e., cytochrome P450 (CYP1A) and
immune responses, but also to determine whether such
pollution would have an impact on the growth or repro-
ductive cycle. Effects on growth and reproduction were
Fig. 1 Map showing the location of field sites and of the Erika investigated through controlled laboratory experiments,
shipwreck. Piriac was contaminated by oil from Erika, whereas Aber using oil-polluted mussels as a food supply for starfish.
remained free of this contamination. Solid line, shoreline affected by These experiments also aimed to investigate PAH uptake
the oil spill; dotted line, the main oil slick drift from December 12 to
December 24, 1999 (from Laubier et al. 2004)
and loss over time in the digestive structures of A. rubens,
in order to determine possible relationships between the
presence of PAHs in starfish and the biological effects
polluted shore of the Croisic headland (Loire Atlantique) observed.
a few days after the spill (Bruet, personal communica-
tion). A. rubens is a common marine predator of the
Atlantic coast and has been frequently used as a bioin- Materials and Methods
dicator of contamination of the benthic environment by
heavy metals and polychlorinated biphenyls (Coteur et al. Field Survey
2003b, c; Danis et al. 2004b; den Besten et al. 2001;
Guns et al. 1999; Stronkhorst et al. 2003; Temara et al. The field survey of the effects of the Erika oil spill on
1998). The expression of A. rubens cytochrome P450- Asterias rubens populations was conducted by sampling
immunopositive protein (Danis et al. 2004a; den Besten starfish from two intertidal populations: Piriac (Croisic
et al. 1993) and immune response (Coteur et al. 2005) are headland), contaminated by the Erika oil spill on 25
among the biomarkers used for such toxicity studies. December 1999 (Laubier et al. 2004), and Aber (Bay of
Moreover, contaminated sediments from the North Sea Douarnenez; Fig. 1), which remained free of pollution.
were proven to have deleterious effects on embryo Both field sites were visited at low tide to collect starfish
development of this species (Coteur et al. 2003b). The on days 40 (February 2000), 58 (February 2000) and 221
finding of sublethal effects on A. rubens skeletogenesis (August 2000) and 2 years after the oil spill (January
and growth following contamination by heavy metals 2002). Individuals with a 50- to 70-mm-long ray were
(Temara et al. 1997, 1998) suggests that pollution could sampled (the ray corresponds to the size of the longest arm
also have an impact on the demographic parameters of the measured from mouth to tip with a Vernier caliper). On day
species. 40 no starfish were found at Aber, and the five starfish
The literature contains few reports about the effects of collected at Piriac the same day were used for immune
oil spillage on starfish populations, apart from some mon- response measurements. On days 58 and 221 and 2 years
itoring carried out after the Exxon Valdez oil spill (Dean after the oil spill, four starfish from each site were dissected
et al. 1996; Hooten and Highsmith 1996; Jewett et al. in the field for cytochrome P450-immunopositive protein

123
Arch Environ Contam Toxicol (2009) 56:209–220 211

induction measurement (CYP1A): 2 ml of pyloric ceca was Table 1 List of PAH compounds studied with the internal standards
blotted, immediately frozen in liquid nitrogen, and stored at used for GC/MS analysis
-80°C until analysis. On days 58 and 221, another 10 Compound Abreviation Internal
starfish were brought to the laboratory, where they were standard
kept for 24 h in aquaria supplied with clean running sea-
Naphthalene Nd8
water prior to immune response measurements. On day 58
Acenaphthylene Nd8
and 2 years after the oil spill, 10 starfish from each site
Acenaphthene Nd8
were dissected for PAH analysis; two pools of pyloric ceca
Fluorene Pd10
and two pools of body wall were made (10 starfish per
Phenanthrene Phe Pd10
pool) and stored at -20°C.
Anthracene Pd10
Fluoranthene Fluod10
Cytochrome P450-Immunopositive Protein Induction
Benzo(a)anthracene Chrys d12
Measurement (CYP)
Pyrene Pyr Pyrd10
The P450-immunopositive protein (CYP1A-IPP) contained Triphene + chrysene Chrys Chrys d12
in the postmitochondrial fraction of the pyloric ceca was Benzo(b + j + k) fluoranthene BePd12
quantified using a competitive ELISA method as described Benzo(e)pyrene BeP BePd12
by Danis et al. (2004a). The final results were expressed as Benzo(a)pyrene BaP BaPd12
inhibition factors, i.e., the ratio of CYP1A-IPP levels Perylene BaPd12
between experimental and control groups. Indeno(1,2,3-cd)pyrene BPd12
Dibenz(a)anthracene + dac BPd12
Immune Response Measurement Benzo(ghi)perylenep BPd12
Sum of unsubstituted PAHs Uns-PAHs
The two essential parameters of the starfish immune system Sum methylnaphthalene Nd8
considered in this study were coelomic amoebocyte con- Sum methylphenanthrenes and MPhe Pd10
centration (CAC) and reactive oxygen species (ROS) dimethyl-phenanthrene
production by resting (unstimulated) or stimulated Methylchrysene MChrys Chrysd12
amoebocytes. Amoebocyte ROS production was measured Sum of methylated PAHs M-PAHs
by the peroxidase luminol-enhanced chemiluminescence Dibenzothiophene DBT DBTd8
(PLCL), method developed by Coteur et al. (2002, 2004, Methyldibenzothiophene MDBT DBTd8
2005). Measurements were normalized using the actual 2,1-Benzonaphtothiophene Chrysd12
amoebocyte concentration in each sample, and expressed Sum of all analyzed PAHs Sum PAHs
as the sum of all 10-min interval measurements (in Relative
Light Units [RLU], an internal scale of the instrument) for
106 cells ml-1 (total chemiluminescence) of resting or except for the oven temperature, which was programmed
stimulated amoebocytes. to increase by 2°C/min from 70°C to 300°C in the present
study. Different perdeuterated PAHs were used as internal
PAH Measurements standards for each class of aromaticity (Baumard and
Budzinski 1997). The response factor of a given PAH
For quantification of the PAHs listed in Table 1, tissue relative to the perdeuterated one was determined for each
samples were freeze-dried and ground prior to analysis. GC/MS sequence by injecting 1 lL of a standard solution
Among the alkylated PAHs, only certain compounds of containing a mixture of PAHs (SRM 2260, National
high molecular weight, known to be representative of the Institute of Science and Technology, Gaithersburg, MD,
crude oil from the Erika, and highly abundant, were USA) and the solution of perdeuterated PAHs used for
chosen for measurement. PAHs were extracted from 1 to sample spiking. The precision of this type of quantifica-
3 g of dry tissues using the microwave-assisted extraction tion has been evaluated to be between 88% and 98%,
method described by Budzinski et al. (1999). Filtration of depending on the compound, with reproducibility between
the extracts, reduction of their volume, and purification 2% and 5%. The entire analytical procedure was per-
were performed as described by Baumard and Budzinski formed several times with certified mussel tissue, SRM
(1997). Experimental blanks were also passed to assess 2974 (NIST, Gaithersburg, MD, USA). The recoveries on
the risk of potential contamination. Extracts were ana- this SRM were between 70% and 110% for five repli-
lyzed by HP GC/MSD according to the procedure used by cates, with reproducibility ranging from 7% to 17%
Geffard et al. (2004) in the Erika monitoring program, depending on the compound.

123
212 Arch Environ Contam Toxicol (2009) 56:209–220

Laboratory Studies Parameters Measured

Our laboratory investigations complemented the field study The parameters studied were size structure, gonad and
by aiming to assess to what extent starfish somatic growth, pyloric cecum indexes, PAH concentrations, righting
reproduction, and development of gametes and larvae were activity coefficient (RAC), ovocyte diameter, fertilization
affected by oil exposure via food. Experiments were per- success, embryo development, and adambulacral spine
formed on starfish at different stages of development and regeneration. The parameters assessed depended on the
reproductive cycle. Uptake and loss of PAHs over time particular experiments.
were also analyzed. Before the experiments were started, starfish size dis-
tribution was carefully checked for likeness between the
groups used for the different treatments. Pyloric and gonad
Experimental Procedure indexes were calculated as follows: organ index = organ
dry weight/body wall dry weight. Gonads and pyloric ceca
Mussels were exposed for 3 days to ‘Erika oil’ samples, were dissected, and the body wall was carefully cleaned.
supplied by the CEDRE (Centre of Documentation, Body compartments were dried separately at 60°C for 48 h
Research and Experimentation on Accidental Water Pollu- and weighed.
tion, Brest, France). A preliminary study had first been RAC is indicative of the level of motion activity (Watts
carried out to establish a protocol to produce a PAH con- and Lawrence 1990). Here, the righting time corresponded to
centration in mussels that would be close to the maximum the time required by a starfish to turn back over once it had
PAH content observed in the mussels in the Bay of Vilaine been placed on its aboral face. RACs were calculated as
after the oil spill (sum PAH = 2,536 lg kg-1 dw in Feb- follows: RAC = 1,000/righting time(s). Among individuals,
ruary 2000 [Tronczynski et al. 2004]). A PAH concentration those showing no sign of righting response initiation within
of 3,420 lg kg-1 (268.5 lg g-1 lipids) ± 900 lg kg-1 10 min were assigned an RAC value of 1,000/(2 9 600).
(mean PAH concentration in three replicates of 30 g ww of Histological observations were made to detect any effect
mussel soft tissue) was obtained after 3 days of contamina- that oil pollution had on ovocyte diameter. Ovary samples
tion using the following method: 1,500g of mussels was were embedded in paraffin, sectioned into 5-lm-thick sli-
placed in an aquarium containing 2 liters seawater and 50 g ces, and colored using Trichrome staining. For each
Erika oil (25 g L-1 oil). The aquarium was covered and the female, digital pictures of 100 ovocytes exhibiting a
soluble phase homogenized by high-air bubbling. After nucleolus and a nucleus within the same section were taken
contamination, the mussels were carefully brushed and with a CCD camera (Iris Sony) and measured with Visilog
cleaned to eliminate any traces of oil. The PAH concentra- 4.0 software; diameters were then obtained by conversion.
tion in the control mussels was 14 lg g-1 lipids. Conta- Fertilization assays were made to assess the capacity of
minated and uncontaminated mussels of the same origin starfish ovocytes to be fertilized, as well as the quality of
were frozen following treatment and prior to use as food. One subsequent larval development. Starfish were sexed by
should note that after 24 h in running seawater, the frozen puncture of gonadic material through the integument.
mussels showed no significant decrease in PAH content Spawning and fertilization were performed as described by
(222 ± 134.3 lg g-1). Starfish were fed at the rate of one Coteur et al. (2003b). Individual female clutches were
mussel per individual per day. fertilized by a pool of sperm (three males). Fertilization
Two treatments were used in the laboratory experi- success was assessed after 1 h by counting the number of
ments—’contaminated treatment’ and ‘control treatment’— eggs with a fertilization membrane in three subsamples of
depending on whether or not the mussels supplied as food 100 eggs each and the embryo development quality was
to starfish were contaminated (details of the separate determined after 72 h. For each fertilization, the different
experiments are given below, under Descriptions of types of larvae (viable larvae, abnormal larvae, and dead
Experiments). For each experiment, the aquaria were filled eggs) were counted in each well of a six-well plate.
with clean running seawater from the nearby Bay of Brest; Adambulacral spine regeneration was induced and
salinity ranged from 33 to 35 PSU. Once a week, the studied according to the methods developed by Dubois and
photoperiod was adjusted to match the natural one. The Jangoux (1990).
mussels fed to each starfish were chosen according to
the size of the starfish under test, according to the rela- Descriptions of Experiments
tion established by Sommer et al. (1999) (mussels of 2-
to 50-mm shell length for starfish of 5- to 65-mm ray Impact of PAH Uptake on Adults at the Gametogenic Stage:
length). A 3-Month Experiment Two laboratory experiments

123
Arch Environ Contam Toxicol (2009) 56:209–220 213

(called 1A and 2A, respectively) were carried out over Statistical Analysis
3 months in 2 successive years. These were conducted dur-
ing the period of gonadal growth from December to the end The measured immune responses (ROS production and
of February. In each treatment (contaminated and control), CAC) were compared between treatments using nonpara-
60 starfish of 35- to 45-mm ray length were used (10 per metric tests since the data that did not fit a normal
aquarium). distribution and CYP1A expression were tested with two-
way ANOVA. In the laboratory contamination trials, gonad
Monitoring PAH Uptake and Its Potential Physiological and pyloric indexes, PAH levels, and physiological
Effects: A Short, 15-Day Experiment This experiment parameters were intercompared and statistically tested
(called PAH uptake) was conducted in June during the using t-tests or ANOVAs, depending on the number of
agametogenic stage on starfish of 50- to 70-mm ray length. samples. In experiments requiring the use of several
PAH accumulation in pyloric ceca (pool of 10 starfish per aquaria for each treatment, differences in the variables
aquarium) and RAC values were measured on days 0, 1, 2, (measured on all individuals) were tested using a nested
5, 7, 10, and 15 of exposure to oil. The regenerating ad- ANOVA with ‘‘aquarium’’ as a nested factor within
ambulacral spines were sampled on five starfish in each ‘‘treatment.’’ The nested factor allows one to distinguish
treatment, 15 days after fracture, and the regeneration size any discrepancy between pseudo-replicates within each
(RAS) measured with a scanning electron microscope as experimental treatment (Zar 1996).
previously described by Dubois and Jangoux (1990). PAH uptake and loss were modeled by quadratic curve
For the sum of all measured PAHs, the sum of parental fitting with Systat 9 software. Residuals were plotted, and
PAHs and PAHs of particular interest in this study, PAH checked to ensure that there was no serial correlation.
uptake vs. time was tentatively modeled with the following Whenever several statistical comparisons were per-
quadratic equation: formed either on the same set of data or on subsets, we
used the Bonferroni criterion and divided by the number of
CPAH ¼ aT 2 þ bT þ c
comparisons.
where CPAH is the concentration of the PAH concerned, T
is the time, expressed as days of exposure to contamina-
tion, and a, b, and c are constants. The inflection point of Results
the curve, I, corresponded to I ¼ b=2a. The transfer
factor (TF) of a given PAH in this study corresponded to Field Survey
the ratio of its concentration in the pyloric ceca (i.e., the
mean of data at days 10 and 15) to that in the food (lg g-1 Fifty-eight days after the oil spill, the PAH levels in the
lipids). pyloric ceca and body wall of starfish from Piriac were higher
than those in starfish from Aber (Fig. 2). The amounts were
Decrease in PAHs in the Pyloric Ceca: A 24-Day Exper- particularly high in pyloric ceca compared with the body
iment Juvenile starfish of about 20-mm ray length were wall. The most abundant PAHs in the fuel transported by the
contaminated over 4 months (from May to September) Erika were analyzed in starfish samples; these were alkyl-
under the conditions previously described. In September substituted homologues of phenanthrene (Phe), dibenzo-
they had grown to 40-65 mm in ray length. Contaminated thiophene (DBT), and chrysene (Chrys) (Tronczynski et al.
and control starfish were then both fed with pollution-free 2004), and their unsubstituted PAHs. Chrysene and its
mussels for 23 days. Temperature was constant and veri- alkylated homologues were far more abundant in starfish
fied daily during the experiment (17.5 ± 0.15°C). Pyloric from Piriac than in controls for both organs studied
index (fresh weight) and RAC value were determined on (Table 2). Moreover, the pyloric ceca of Piriac starfish
four starfish per treatment on days 0, 1, 2, 4, 7, 10, 14, 18, contained more Phe-alkylated homologues and fewer DBT
and 23 of the PAH loss experiment. For the sum of all and DBT-alkylated homologues than did the body walls.
measured PAHs, the sum of parental PAHs and PAHs of Figure 3 illustrates the changes occurring over time in
particular interest in this study, PAH loss by starfish was the responses of the biomarkers according to sampling site.
tentatively modeled using the following equation: At day 58, ROS production by both resting and stimulated
amoebocytes (Mann-Whitney p B 0.005; n = 19), CAC
CPAH ¼ d  ekT þ g
and CYP1A levels were significantly higher in starfish from
where CPAH is the concentration of the concerned PAHs, T Piriac than in those from Aber. However, at day 221, only
is the number of days under PAH loss conditions, and d, k, CAC (p B 0.005; n = 37) and CYP1A levels were more
and g are constants. The biological half-lives of the dif- elevated (two-way ANOVA, factor: site p = 1 9 10-8;
ferent compounds were estimated as Tb = ln2/k. n = 16).

123
214 Arch Environ Contam Toxicol (2009) 56:209–220

Fig. 2 Total PAH levels in field-collected starfish 58 days and 2

years after the Erika oil spill. Piriac, oil-exposed site; Aber, reference
site. All data are means; n = 2 pools (10 starfish per pool). The
numbers above the histogram bars indicate the lipid contents (mg g-1)
of the organs

Table 2 Characterization of PAH levels in organs of starfish

Body wall Pyloric ceca
Aber Piriac Aber Piriac

Uns-PAHs 38% 32% 41% 30%

M-PAHs 62% 68% 59% 70%
Phe 712 9 10-4 506 9 10-4 346 9 10-4 262 9 10-4 Fig. 3 Temporal changes in biomarker responses after the oil spill in
Alkyl-Phe 319 9 10 -3
349 9 10 -3
180 9 10 -3
428 9 10-3 field-collected starfish from Piriac (oil-exposed site) and Aber
(reference site). All data are means ± standard deviations. (1) ROS
DBT 31 9 10-4 78 9 10-4 39 9 10-4 24 9 10-4
-4 -4 -4
production by resting amoebocytes (n = 5 or 10); (2) ROS produc-
MDBT \10 122 9 10 156 9 10 176 9 10-4 tion by stimulated amoebocytes (n = 5 or 10); (3) coelomic
-4 -4 -4
Chrys 56 9 10 478 9 10 146 9 10 734 9 10-4 amoebocyte concentration (n = 5 or 10); (4) CYP1A induction
MChrys \10-4 1,787 9 10 -4
39 9 10-4
1,984 9 10-4 (n = 4). Histogram bars with the same letters are not significantly
different. Lowercase letters are for comparison of dates at Piriac,
Note. Piriac, oil-exposed population; Aber, control. All data are whereas capital letters are for comparison of dates at Aber. A star
means; n = 2 pools (10 starfish by pool). Uns (unsubstituted)- and M indicates a statistical difference between the sites at a given date.
(methylated)-PAH are expressed as percentage of the sum of all ROS, reactive oxygen species; CYP1A, cytochrome P450-immuno-
analyzed PAHs. Phe, phenanthrene; DBT, dibenzothiophene; Chrys, positive protein content
chrysene; Alkyl-Phe, alkyl-phenanthrene. DBT and Chrys are
expressed as micrograms per gram lipids. Lipid contents are those
measured at day 58 (Fig. 2) 40 than at day 58 (Kruskall-Wallis p = 0.007; n = 23),
CAC response, however, showed no significant difference
The immune response parameters showed contrasting between the two dates. ROS production by both resting and
results (Fig. 3): at Piriac, ROS production by both resting stimulated amoebocytes and CAC levels did not change
and stimulated amoebocytes was significantly higher at day significantly between day 58 and day 221 at Piriac. At

123
Arch Environ Contam Toxicol (2009) 56:209–220 215

Aber, in contrast, amoebocyte ROS production (of both

types) was higher on day 221 than on day 58 (Mann-
Whitney p B 0.01; n = 19). CYP1A levels were higher on
day 58 than on day 221 in both populations (two-way
ANOVA; factor, time; p = 23 9 10-8; n = 16).
In January 2002 (after 2 years), starfish from both sites
had a low PAH content with respect to lipids (Fig. 2);
moreover, no significant difference was found in mean
CYP1A expression at Aber (2.12 ± 0.87) vs. Piriac
(1.90 ± 0.71; t–test, p = 0.70).

Laboratory Studies

Impact of PAH Uptake on Adults at the Gametogenic

Stage: a 3-Month Experiment

The 3-month experimental contamination (experiment 1A)

showed an accumulation of PAHs in the pyloric ceca of con-
taminated starfish (49.3 ± 18.8 lipids vs. 1.7 ± 0.7 lg g1
lipids in the control) and in their gonads (32.2 ± 6.7 vs.
0.4 ± 0.2 lg g-1 lipids in the control). Differences were
highly significant between treatments (two-way ANOVA,
p = 1 9 10-4) but not significantly different between body
compartments (p = 0.170) or treatment 9 body compart-
ment interaction (p = 0.199). PAH uptake was the same in
pyloric ceca and gonads. Apart from RAC values, which
were higher in experiment 1A in contaminated starfish than
in controls (t-test, p = 0.02; n = 36), no other significant
differences were shown between control and contaminated Fig. 4 Kinetics of PAH uptake in the pyloric ceca of starfish (n = 2
treatments in the parameter measured: (i) mean size (p = pools of 10 starfish) fed with contaminated mussels over the
agametogenic stage. Sum PAHs, sum of PAHs analyzed in this
0.81 and 0.13, with n = 63 and 50, for experiments 1A and study; uns-PAHs, sum of unsubstituted PAHs; Pyr, pyrene; Phe,
2A, respectively), (ii) gonad index (p = 0.43, with n = 63, phenanthrene; DBT, dibenzothiophene; Chrys, chrysene
in experiment 1A), (iii) pyloric index (p = 0.26 and 0.26,
with n = 63 and 50, for experiments 1A and 2A, respec- Table 3 Kinetics of PAH uptake and results of the quadratic curve
tively), and (iv) ovocyte diameters (p = 0.81, with n = 10 in Curve fitting TF
experiment 1A). 2
p R I (days)
Spawning and fertilization were induced at the end of
these experiments. At least 90% of eggs presented a fer- Sum PAHs \1 9 10-3 0.823 9.48 0.110
tilization membrane, demonstrating their fertilization Uns-PAHs \1 9 10-3 0.881 9.11 0.125
quality. No significant difference was found in the per- Pyr \1 9 10-3 0.909 9.19 0.121
centage of dead eggs and viable larvae between larval Phe \1 9 10-3 0.874 8.82 0.159
batches produced from contaminated or uncontaminated DBT \1 9 10-3 0.756 9.37 0.118
parents (ANOVA, p = 0.76; t-test, p = 0.95; n = 12). Chrys \1 9 10 -3
0.849 10.12 0.086
Note. n = 2 pools of 10 starfish fed contaminated mussels during the
Monitoring PAH Uptake and Its Potential Physiological agametogenic stage (one mussel per individual starfish per day; PAH,
Effects: A Short, 15-Day Experiment 268.5 ± 78.5 lg/g lipids). I, inflection point of the fitted curve; TF,
transfer factor; p, curve-fitting probability; R2, correlation coefficient.
Starfish in the contamination treatment consumed the same Sum PAHs, sum of all analyzed PAHs; uns-PAHs, sum of analyzed
unsubstituted PAHs; Pyr, pyrene; Phe, phenanthrene; DBT, diben-
number of mussels as control starfish. During this experi- zothiophene; Chrys, chrysene
ment the changes in daily seawater temperature were low
(15.4 ± 0.4°C). The patterns of PAH accumulation in the observed in the first days in the two pools analyzed from
pyloric ceca appeared to be the same for unsubstituted and contaminated starfish was followed by a decrease between
alkylated PAHs (Fig. 4, Table 3): the marked uptake day 9 and day 10. Transfer factors were always less than

123
216 Arch Environ Contam Toxicol (2009) 56:209–220

0.2. The evolution over time in the ratios of benzo(e)pyrene No differences were observed in RAC, pyloric indexes,
(BeP) to benzo(a)pyrene (BaP), and the ratios of some or size of regenerating adambulacral spines between the
parent PAHs: chrysene, phenanthrene, and dibenzothio- two groups of contaminated starfish (t-tests, p [ 0.05), and
phene to their more hydrophobic methyl-substituted mean RAC showed no temporal evolution whatever the
compounds, showed a preferential accumulation of the treatment (Pearson correlation coefficients r = -0.492,
more soluble PAHs over the first 3 days of the experiment, -0.573, and -0.502, respectively, for control starfish,
except for the chrysene family (Fig. 5). Such an evolution contaminated starfish in aquarium 1, and contaminated
is indicative of a biotransformation process. A slight rise in starfish in aquarium 2; p C 0.2, n = 7, for each). As the
chrysene was noted throughout the experiment. An accu- starfish consumed food during the experiment and accu-
mulation of DBT and Phe during the first days was mulated reserves throughout the agametogenic stage, their
followed by a large increase in methyl-substituted com- pyloric indexes under both contaminated and control con-
pounds from the fifth day. From the second day BeP ditions were higher compared to those measured at the
increased faster than BaP, indicating preferential metabo- beginning of the experiment (day 0) (ANOVA, p = 4.10-
4
lization of BaP compared with BeP. After the seventh day, ; n = 40), but no difference was detected between treat-
none of these ratios changed very much. Overall contam- ments (t-test, p [ 0.1). Similarly, neither RAC values nor
ination of starfish under controlled conditions showed no the sizes of regenerating adambulacral spines were found
evolution over time, though, with the sum of analyzed to differ between contaminated and control starfish (t-test,
PAHs remaining within the range 0.3 to 0.6 lg g-1 lipids. p = 0.68 and 0.344, n = 30 and 14, respectively).

Decrease in PAHs in the Pyloric Ceca: a 24-Day

Experiment

Mean daily seawater temperature in the tanks

(17.5 ± 0.1°C) did not change during this experiment.
PAH levels decreased rapidly in the pyloric ceca (Fig. 6),
and calculated half-lives were within 2 and 4 days
(Table 4). After 23 days of PAH loss, most of the indi-
vidual PAHs of particular interest were close to zero apart
from a few PAHs including pyrene, chrysene, and phen-
anthrene that had concentrations much higher than in the
control. No significant differences in RAC were observed
between treatments or days (two-way ANOVA, p = 0.23,
0.39, and 0.75 for the factors ‘‘treatment,’’ ‘‘days,’’ and
their interaction, respectively; n = 74). No significant
difference was detected in pyloric indexes between
decontaminated starfish (PI = 0.496 ± 0.099) and the
control (PI = 0.440 ± 0.064; t-test, p = 0.12; n = 74).

Discussion

Biomarker Expression and PAH Dynamics in Asterias

rubens Pyloric Ceca

This study has provided evidence for contamination of the

pyloric ceca and body wall of Asterias rubens in the field
by the oil from the Erika, spilt along the shoreline of South
Brittany over the first weeks following the shipwreck. Less
than 2 months after the event, high PAH levels character-
Fig. 5 Temporal changes in biotransformation ratios in the pyloric istic of fuel from the Erika were detected in pyloric ceca of
ceca of starfish (n = 2 pools of 10 starfish) fed with contaminated
starfish at Piriac. ROS production by resting or stimulated
mussels. BeP, Benzo(e)pyrene; BaP, benzo(a)pyrene; Phe, phenan-
threne; MPhe, methyl-phenanthrene; DBT, dibenzothiophene; MDBT, amoebocytes and coelomic amoebocyte concentration
methyl-dibenzothiophene; Chrys, chrysene; MChrys, methyl-chrysene (CAC) were simultaneously enhanced in the impacted

123
Arch Environ Contam Toxicol (2009) 56:209–220 217

expression by exposure to PCBs had already been reported

in the pyloric ceca of A. rubens (Danis et al. 2004a, b), but
the present study is the first time that a high reactive signal
was clearly observed in the presence of PAHs. These
biomarker expression patterns suggest a rapid initiation of
the biotransformation process.
Laboratory experiments confirm this hypothesis and
allow a better understanding of the dynamics of some
specific aromatic hydrocarbons. While PAH uptake within
the pyloric ceca of adult starfish fed with oil-polluted
mussels was fast (the peak concentration was reached in 8
to 10 days after the start of contamination), elimination and
biotransformation were also rapid: even under conditions
with a constant supply of contaminated food. The most
soluble compounds, apart from the chrysene family, were
preferentially accumulated over the first few days. Indeed,
among the PAH parent compounds under study, chrysene
was the only one that showed a continuous increase in
relation to its alkylated homologues during the 15 days of
the PAH uptake experiment. This indicated that it was not
biotransformed, in contrast to benzo(a)pyrene, whose bio-
transformation was induced in the first days of
contamination. Benzo(a)pyrene is more reactive and oxi-
dizable than benzo(e)pyrene (Baumard et al. 1999) and the
observation of benzo(a)pyrene biotransformation is in
agreement with the study by den Besten et al. (1993),
Fig. 6 Loss kinetics of PAH in pyloric ceca of starfish (n = 1 pool of 4 where benzo(a)pyrene hydroxylase activity (BPH) was
starfish) fed with uncontaminated mussels over 23 days. sumPAHs, sum enhanced in A. rubens pyloric cecum microsomes in the
of all analyzed PAHs; uns-PAHs, sum of nonsubstituted PAHs; Pyr, days immediately following injection of benzo(a)pyrene
pyrene; Phe, phenanthrene; DBT, dibenzothiophene; Chrys, chrysene into the animals. Moreover, the low transfer factors of
PAHs (\1) in the pyloric ceca indicate that there was no
Table 4 PAH loss in the pyloric ceca of starfish and results of curve biomagnification through the mussel-starfish trophic link.
fitting for certain PAHs Following the end of the experimental pollutant expo-
p R2 D g h sure, half of the total PAHs were eliminated in the next 2 to
4 days. However, 20 days later, a certain amount of PAHs
Sum PAHs \1 9 10-3 0.958 92.84 3.50 3.03 remained stored in the pyloric ceca. Experimental obser-
-3
Uns-PAHs \1 9 10 0.983 21.323 1.45 2.77 vations therefore suggest that despite clear evidence for
Pyr \1 9 10-3 0.980 2.97 –0.0318 3.50 rapid biotransformation processes, the longer the exposure
Phe \1 9 10-3 0.976 7.05 0.284 2.02 to pollution continues, the higher the level of PAHs
DBT \5 9 10-3 0.856 2.27 –0.0065 2.12 accumulated in the pyloric ceca. Table 5 reports that a
Chrys \1 9 10-3 0.926 4.744 0.0117 3.42 constant supply of oil-polluted food lowered the biotrans-
Note. n = 1 pool of 4 starfish fed with two uncontaminated mussels formation capability of starfish. Measured PAH
per individual per week over 23 days. p, curve-fitting probability; R2, concentrations were much lower in the field than under
correlation coefficient; D and g, parameters of the fitted equation experimental conditions (24 times lower in the field
(lg g-1 of lipids); h, half-life of the compounds (days). Sum PAHs, 2 months after the oil spill than in the laboratory after
sum of all analyzed PAHs; uns-PAHs, sum of analyzed unsubstituted
PAHs; Pyr, pyrene; Phe, phenanthrene; DBT, dibenzothiophene; 15 days of controlled contamination). It is also worth
Chrys, chrysene noting that the high oil concentrations found in mussels
following the oil spill did not remain as constant as the
contamination under controlled laboratory conditions.
starfish. A high induction of CYP1A expression was also Several biomarkers were used in this study to monitor
observed. CYP1A converts particularly insoluble organic the physiological response of starfish to PAHs in the field.
compounds such as PAHs and PCBs (Bucheli and Fent All showed a significant enhancement in the impacted
1995) into soluble derivatives. The induction of CYP1A starfish compared with controls a few weeks after the oil

123
218 Arch Environ Contam Toxicol (2009) 56:209–220

Table 5 Comparison of PAH load in starfish pyloric ceca in the field and under experimental conditions
Field: day In-lab experiments
58, Piriac
Short-term Long-term contamination End of the PAH
contaminationa b c loss experimentd
Adults Juveniles

Sum PAHs 0.97 23.16 (24) 49.27 (51) 86.15 (89) 3.50 (4)
Uns-PAHs 0.29 6.53 (23) 11.37 (40) 21.43 (75) 1.45 (5)
Pyr 0.062 0.79 (13) 1.19 (19) 2.73 (44) -0.032 (-1)
Phe 0.026 2.29 (87) 3.12 (119) 7.12 (272) 0.28 (11)
DBT 0.0024 0.46 (193) 0.54 (228) 1.92 (803) -0.0065 (-3)
Chrys 0.073 1.00 (14) 1.72 (23) 3.97 (54) 0.012 (0.2)
a b c
Note. Laboratory experiments: day 15 of PAH uptake experiment; end of the first experiment on adults (1A); day 0 of the PAH loss
experiment; d day 23 of the PAH loss experiment (fitted values in g). Sum PAH (sum of all analyzed PAHs), uns-PAHs (sum of analyzed
unsubstituted PAHs), Pyr (pyrene), Phe (phenanthrene), DBT (dibenzothiophene), Chrys (chrysene) are expressed as micrograms per gram of
lipids. Numbers in parentheses are PAH concentrations expressed as multiples of the field-measured concentrations

spill. But although the CYP1A signal is clear, ROS pro- therein; Ordzie and Garofalo 1981), feeding rate and adult
duction by starfish from the contaminated field site were growth (Crapp 1971; Georgiades et al. 2003; O’Clair and
within the range of the values found by Coteur et al. (2004) Rice 1985; Temara et al. 1999), and embryo development
in natural uncontaminated populations, and CAC values (Davis et al. 1981). However, in these studies, the starfish
were below those previously measured in heavily polluted were exposed either to the water-accommodated fraction
locations (Coteur et al. 2003a, b). The kinetics of PAH loss (WAF) of different types of oil or to a complex mixture of
previously described suggest that rapid decontamination hydrocarbons and dispersants. Volatile and soluble oil
could possibly be occurring and thus that there may be a fractions may have greater direct effects on echinoderms
reduction in ROS production during the 24 h that elapsed than compounds obtained via food. In the sea urchin,
between starfish collection and response measurement. Paracentrotus lividus, the righting response was enhanced
Such an explanation cannot account for the CAC result, by the most volatile compounds of crude oil, but not by
though, since immune cell turnover in marine invertebrates weathered crude oil (Axiak and Saliba 1981). The effects
exceeds 24 h (McIntosh and Robinson 1999). CAC is of WAF are more harmful on embryo development in sea
known to be quite a robust biomarker for stress (Coteur urchins (references in Kobayashi 1995): adult exposure to
et al. 2004), and the differences observed between the two polluted effluents can lead to egg abnormalities at the first
populations would indeed have resulted from the PAH cleavage stage (Vashchenko 1980) or affect egg fertiliza-
level in the field. However, as it is well known that envi- tion and sperm motility (Krause 1994). It is not easy to
ronmental factors such as temperature can influence determine whether the reported differences resulted from
immune response in starfish (Coteur et al. 2004), the exposure conditions, pollutant sensitivity of the organism
immune biomarker levels could have been modulated by under study, or levels of contamination, because none of
changes in water temperature, notably in the impacted field these authors qualified or quantified the level of parental
site between day 40 and day 58. contamination. Once again, without additional data about
the relative toxicity of PAHs, it does not seem relevant to
Effect of PAHs on Asterias rubens Growth, speculate on whether volatile and soluble oil fractions are
Reproduction, and Embryo Development more or less toxic or bioaccumulated than PAHs taken in
with food. Only Barron et al. (1999) have suggested that
As observed in the present study, irrespective of which PAHs are not necessarily the major determinant of WAF
experiment is considered, neither growth nor reproductive toxicity.
capability of starfish was affected by PAH exposure, In conclusion, this study has demonstrated that uptake of
despite an increase in PAH content in the pyloric ceca. PAHs via food is an important route for contamination of
PAH accumulation in the gonads had no effect on ovocyte the common starfish A. rubens. Short-term impacts of oil
morphometry, and fertilization rates were only very contamination were easily quantified using detoxification
slightly affected. To our knowledge, no previous study has biomarkers and immunological responses and by moni-
dealt with the impact of oil contamination through food. toring PAH uptake and loss in the storage organs.
Oil has previously been shown to induce damage to starfish Moreover, A. rubens and suitable biomarkers (e.g., CAC
locomotive function (O’Clair and Rice 1985 and references and CYP1A) can be judged to be suitable for use as early

123
Arch Environ Contam Toxicol (2009) 56:209–220 219

warning signals for exposure to PAHs. However, behavior, enhanced chemiluminescence (PLCL) as an immunotoxicolog-
reproduction, and growth did not appear to be affected by ical tool. In: Mantraga V (ed) Progress in molecular and
subcellular biology subseries marine molecular biotechnology.
PAH exposure through food. Finally, the present data Springer-Verlag, Berlin, Heidelberg, pp 71–83
showed no significant long-term effects of the Erika Crapp G (1971) Chronic oil pollution. In: Cowell EB (ed) The
oil spill on A. rubens dynamics. The predation power of ecological effect of oil pollution on littoral communities.
A. rubens thus appears to be unaffected in the mussel beds Institute of Petroleum, London, pp 187–203
Danis B, Goriely S, Dubois P, Fowler SW, Flamand V, Warnau M
of the intertidal rocky temperate communities structured by (2004a) Contrasting effects of coplanar versus non-coplanar
this starfish. PCB congeners on immunomodulation and CYP1A levels
(determined using an adapted ELISA method) in the common
Acknowledgments This study was supported by the French Min- sea star Asterias rubens L. Aquat Toxicol 69:371–383
istry of Ecology and Sustainable Development within the framework Danis B, Wantier P, Flammang R, Dutrieux S, Dubois P, Warnau M
of the ‘‘Erika Monitoring’’ program, a CGRI-FNRS-CNRS coopera- (2004b) Contaminant levels in sediments and asteroids (Asterias
tion agreement, and by a French Research Ministry grant for PhD rubens L. Echinodermata) from the Belgian coast and Scheldt
cosupervision. We are grateful to Ph. Pernet (ULB) and M. Le Meur estuary: polychlorinated biphenyls and heavy metals. Sci Total
(LPTC) for their help in the analytic and experimental work, to Environ 333:149–165
M. Briand for her help with the preparation of figures, and to Davis PH, Schultz TW, Spies RB (1981) Toxicity of Santa Barbara
Dr. H. McCombie for revision of the English. seep oil to starfish embryos: part 2. The growth bioassay. Mar
Environ Res 5:287
Dean TA, Jewett SC, Laur DR, Smith RO (1996) Injury to epibenthic
invertebrates resulting from the Exxon Valdez oil spill. Am Fish
References Soc Symp 18:424–439
den Besten PJ (1998) Cytochrome P450 monooxygenase system in
Axiak V, Saliba LJ (1981) Effect of surface and sunken crude oil on echinoderms. Comp Biochem Physiol 121C:139–146
the behaviour of a sea urchin. Mar Pollut Bull 12:14–19 den Besten PJ, Lemaire P, Livingstone DR, Woodin B, Stegeman JJ,
Barron MG, Podrasbsky T, Ogle S, Ricker RW (1999) Are aromatic Herwig HJ, Seinen W (1993) Time course and dose response of
hydrocarbons the primary determinant of petroleum toxicity to the apparent induction of the cytochrome P450 monooxygenase
aquatic organisms? Aquat Toxicol 46:253–268 system of pyloric caeca microsomes of the female sea star
Baumard P, Budzinski H, (1997) Internal standard quantification Asterias rubens L. by Benzo[a]pyrene and polychlorinated
method and gas chromatograph-mass spectrometer (GC-MS): a biphenyls. Aquat Toxicol 26:23–40
reliable tool for polycyclic aromatic hydrocarbon (PAH) quan- den Besten PJ, Valk S, Weerle van E, Nolting RF, Postma JF,
tification in natural matrices. Analysis 25:246–252 Everaarts JM (2001) Bioaccumulation and biomarkers in the
Baumard P, Budzinski H, Garrigues P, Dizier H, Hansen PD (1999) seastar Asterias rubens (Echinodermata: Asteroidea) a North Sea
Polycyclic aromatic hydrocarbons in recent sediments and field study. Mar Environ Res 51:365–387
mussels (Mytilus edulis) from the Western Baltic Sea: occur- Diderich R (2000). Etude des risques des substances contenues dans
rence, bioavailability and seasonal variations. Mar Environ Res le fuel transporté par l’ERIKA sur les écosystèmes aquatiques.
47:17–47 Rapport 7 du dossier Erika. Institut National de l’Environnement
Bucheli TD, Fent K (1995) Induction of cytochrome P450 as a Industriel et des Risques, Ministère de l’Aménagement du
biomarker for environmental contamination in aquatic ecosys- territoire et de l’Environnement, Paris
tems. Crit Rev Env Sci Tech 25:201–268 Dubois P, Jangoux M (1990) Stereom morphogenesis and differen-
Budzinski H, Letellier M, Garrigues P, Le Menach K (1999) tiation during regeneration of adambulacral spines of Asterias
Optimisation of the microwave assisted extraction in open cell rubens (Echinodermata, Asteroı̈dea). Zoomorphology 109:263–
of PAHs from soils and sediments—study of moisture effect. J 272
Chromatogr A 837:187–200 Geffard O, Budzinski H, Le Menach K (2004) Chemical and
Coteur G, Warnau M, Jangoux M, Dubois Ph (2002) Reactive oxygen ecotoxicological characterization of the ‘Erika’ petroleum: bio-
species (ROS) production by amoebocytes of Asterias rubens tests applied to petroleum water-accommodated fractions and
(Echinodermata). Fish Shellfish Immunol 12:187–200 natural contaminated samples. Aquat Living Resour 17:289–296
Coteur G, Gillan D, Joly G, Pernet P, Dubois P (2003a) Field Georgiades ET, Holdway DA, Brennan SE, Butty JS, Temara A
contamination of the starfish Asterias rubens by metals. Part 2: (2003)The impact of oil-derivated products on the behaviour and
effects on cellular immunity. Environ Toxicol Chem 22:2145– biochemistry of the eleven-armed asteroid Coscinasterias muri-
2151 cata (Echinodermata). Mar Environ Res 55:257–276
Coteur G, Gosselin P, Wantier P, Chambost-Mancier Y, Danis B, Guns M, Hoeyweghen van P, Vyncke W, Hillewaert H (1999) Trace
Pernet P, Warnau M, Dubois P (2003b) Echinoderms as metals in selected benthic invertebrates from Belgian coastal
bioindicators, bioassays and impact assessment tools of sedi- waters (1981–1996). Mar Pollut Bull 38:1184–1193
ment-associated metals and PCBs in the North Sea. Arch Hooten A J, Highsmith RC (1996) Impact on selected intertidal
Environ Con Tox 45:190–202 invertebrates in Herring Bay, Prince William Sound, after the
Coteur G, Pernet P, Gillan D, Joly G, Maage A, Dubois P (2003c) Exxon Valdez oil spill. Am Fish Soc Symp 18:249–270
Field contamination of the starfish Asterias rubens by metals. Jewett SC, Dean TA, Laur DR (1996) Effect of the Exxon Valdez Oil
Part 1: short- and long-term accumulation along a pollution spill on benthic invertebrates in an oxygen-deficient embayment
gradient. Environ Toxicol Chem 22:2136–2144 in Prince William Sound, Alaska. Am Fish Soc Symp 18:440–
Coteur G, Corriere N, Dubois P (2004) Environmental factors 447
influencing the immune responses of the common European Kobayashi N (1995) Bioassay data for marine pollution using
starfish (Asterias rubens). Fish Shellfish Immun 16:51–63 echinoderms life. In: Cheremisinoff PN (ed) Encyclopedia of
Coteur G, Danis B, Dubois P (2005) Echinoderm reactive oxygen environmental control technology, vol. 9. Gulf, Houston, TX, pp
species (ROS) production measured by peroxydase, luminol- 539–609

123
220 Arch Environ Contam Toxicol (2009) 56:209–220

Krause PR (1994) Effects of an oil production effluent on gamete Stronkhorst J, Ariese F, van Hattum B, Postma JF, de Kluijver M, den
genesis and gamete performance in the purple sea urchin Besten PJ, Bergmann MJN, Daan R, Murk AJ, Vethaak AD
(Strongylocentrotus purpuratus, Stimpson). Environ Toxicol (2003) Environmental impact and recovery at two dumping sites
Chem 13:1153–1161 for dredge material in the North Sea. Environ Pollut 124:17–31
Laubier L, Le Moigne M, Flammarion P, Thybaud E, Cossa D (2004) Temara A, Nguyen QA, Hogarth AN, Warnau M, Jangoux M, Dubois
The monitoring programme of the ecological and ecotoxicolog- P (1997) High sensitivity of skeletogenesis to Pb in the Asteroid
ical consequences of the ‘‘Erika’’ oil spill. Aquat Living Resour Asterias rubens (Echinodermata). Aquat Toxicol 40:1–10
17:239–241 Temara A, Skei JM, Gillan D, Warnau M, Jangoux M, Dubois P
McIntosh LM, Robinson WE (1999) Cadmium turnover in the (1998) Validation of the Asteroid Asterias rubens (Echinoder-
hemocytes of Mercenaria mercenaria (L.) in relation to hemo- mata) as a bioindicator of spatial and temporal trends of Pb, Cd
cyte turnover. Comp Biochem Phys C 123:61–66 and Zn contamination in the field. Mar Environ Res 45:341–356
O’Clair CE, Rice SD (1985) Depression of feeding and growth rates Temara A, Gulec I, Holdway DA (1999) Oil-induced disruption of
of the seastar Evasterias troschelii during long-term exposure to foraging behaviour of the asteroid keystone predator, Coscinas-
the water-soluble fraction of crude oil. Mar Biol 84:331–340 terias muricata (Echinodermata). Mar Biol 133:501–507
Ordzie CJ, Garofalo GC (1981) Lethal and sublethal effects of short Tronczynski J, Miunschy C, Héas-Moisan K, Guiot N, Truquet I,
term acute doses of Kuwait crude oil and a dispersant Corexit Olivier N, Men S, Furaut A (2004) Contamination of the Bay of
9527 on bay scallops, Argopecten irradians (Lamarck) and two Biscay by polycyclic aromatic hydrocarbons (PAHs) following
predators at different temperatures. Mar Environ Res 5:195–210 the T/V ‘‘Erika’’ oil spill. Aquat Living Resour 17:243–259
Ryder K, Temara A, Holdway DA (2004) Avoidance of crude-oil Vashchenko MA (1980) Effects of oil pollution on the development
contaminated sediment by the australian seastar, Patiriella of sex cells in sea urchins. Helgoländ wiss Meer 33:297–300
exigua (Echinodermata: Asteroidea). Mar Pollut Bull 49:900– Watts SA, Lawrence JM (1990) The effect of temperature and salinity
909 interactions on rightning, feeding and growth in the sea star
Sommer U, Meusel B, Stielau C (1999) An experimental analysis of Luidia clathrata (Say). Mar Behav Physiol 17:159–165
the importance of body-size in the seastar-mussel predator-prey Zar JH (1996) Biostatistical analysis. Prentice-Hall, Englewood
relationship. Acta Oecol 20:81–86 Cliffs, NJ

123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.