in the Sea Star Asterias rubens L. - Université Libre de Bruxelles

UNIVERSITE LIBRE DE BRUXELLES – FACULTE DES SCIENCES
LABORATOIRE DE BIOLOGIE MARINE
Bioaccumulation and Effects of Polychlorinated Biphenyls (PCBs)
Thesis submitted in fulfilment of
the degree of :
Doctor in Sciences
in the Sea Star Asterias rubens L.
1
Bruno Danis
March 2004
Supervisors:
Dr Michel Warnau
Dr Philippe Dubois

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
2

Si les pétroliers transportaient de l'eau de mer, on s'en foutrait qu'ils fassent naufrage…
3
Philippe Geluck
A ceux qui nous manquent et qui veillent sur nous…

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
4

ACKNOWLEDGEMENTS-REMERCIEMENTS
5
ACKNOWLEDGEMENTS-REMERCIEMENTS
Quand vient le moment d’écrire les remerciements, des tonnes de souvenirs s’engouffrent
dans votre tête, faisant apparaître les personnes sans qui une thèse ne serait pas ce qu’elle est,
sans qui tous les éléments qui la composent ne se seraient pas ajustés comme ils le sont…
Beaucoup de mes phrases vont commencer par « je remercie le Dr… », mais bon c’est comme
ça… [on se croirait un peu à une cérémonie de remise des Oscars…]
Je commence évidemment par remercier le Professeur Michel Jangoux, qui m’a ouvert les
portes de son laboratoire il y a six ans déjà (!) et qui, j’en suis sûr, a gardé un œil bienveillant
sur moi pendant les moments difficiles ou joyeux qui ont échelonné cette période.
Je remercie le Dr Philippe Dubois, mon copromoteur et voisin de palier. Ton sens critique
affuté m’a plus d’une fois poussé à puiser au fond de mes ressources. Merci d’avoir été
toujours présent dans les moments de doutes, et merci pour ta franchise.
Je remercie le Dr Michel Warnau, mon autre copromoteur, mon quasi-grand frère. Généreux
et impartial, sans compter, tu m’as donné le goût de la recherche, celle que l’on mène en
poussant toujours plus loin ses limites, celle pour laquelle on a tant besoin de ses proches.
Merci à Geneviève et aux trois petits warnouilles: Nathan, Luane et Max, pour la gentillesse
sans fin que vous avez montré à mon égard.
Je remercie le Dr Scott Fowler, un autre protagoniste bienveillant de ma thèse, qui m’a
accueillit au sein de son laboratoire à plusieurs reprises, et m’a permis de concrétiser mes
fantasmes expérimentaux pratiquement sans limite.
Je remercie Jean-Louis Teyssié, technicien de l’extrême dopé à la salade qui, tel un compteur
Geiger, détecte la moindre trace de radioactivité. Tu m’as appris tous les rouages de la
manipulation « à chaud », et tu n’as jamais reculé devant les défis que je souhaitais que nous
relevions ensemble.
Je remercie Olivier Cotret, autre technicien de l’extrême, qui a été ma main droite pendant
toute la durée des expérimentations monégasques. Merci pour tes Hénaurmes coups de main !
Je remercie le Dr Jean-Paco Bustamante, mon binôme Rochelais, avec qui nous avons bataillé
à mort, à coup de cerises, dans le jardin de M. Verola, ex-champion de boules de son petit
état.

6
ACKNOWLEDGEMENTS-REMERCIEMENTS
Je remercie le Dr Jean-Pierre Villeneuve pour son aide précieuse dans l’analyse des PCB
« froids », et pour sa grande patience.
Je remercie le Dr Chantal Cattini pour sa gentillesse, son aide, ainsi que pour le temps qu’elle
a passé pour moi devant sa colonne, à voir couler de l’extrait d’étoile de mer.
Je remercie le Dr Véronique Flamand pour ses précieux conseils en matière d’ELISA et son
ouverture d’esprit.
Je remercie le Dr Virginie De Backer pour son aide et ses conseils concernant la famille des
dioxines.
Je remercie le Dr Patrick Flammang pour ses conseils pour la réalisation des Western Blots.
Je remercie le Dr Pascale Wantier pour sa gentillesse et pour m’avoir initié aux joies de
l’analyse des PCBs.
Je remercie le Prof Robert Flammang pour son intérêt pour notre projet, et l’implication de
son laboratoire pendant plusieurs années.
Je remercie le Dr 2 Stanislas Goriely, qui m’a initié aux joies de l’ELISA, mais qui est surtout
mon ami depuis quinze ans. Merci aussi à Nanou et au petit Kolya pour les inoubliables
brunchs du dimanche. Je suis sûr que nos bambins joueront encore longtemps ensemble.
Je remercie le Dr Drossos, dextre chirurgien de la main, à qui je dois le sauvetage de mon
annulaire droit après une rencontre indésirée, un soir de noël 2003, entre les tendons de mon
cher doigt et une rogneuse mal léchée de la marque Ideal, qu’au passage je ne remercie pas,
car elle ne place pas de garde sur de tels engins.
Je remercie évidemment l’ensemble du Biomar, dans l’ordre du début à la fin du couloir
« Hippie» Chantal, « Gomez da Costa » Sergio, «DivX le Breton » Vinz, « le Dubbe » Phil,
« Sand » Cristi, « le Dim-Dim », « le Loron », « Chipito» GillesD, « Snore » Geoff, « Mac
Guy-ver » Guy, « Never-short-of-a-joke » Herwig, « Oui-oui » Richard, « Vivi » Viviane,
« Psycho » Marcelle, « Happy hour » Edwin, « Drine-drine » Sandrine, « Radar» Phil,
« Buffalo » Beber, « Coup’-coup’ » Dev, « MasterMind » Didje, « Isaac » GillesR, « Aussie »
Raph, « Cotten » Guillemette, « Never-steack never again» den Daaav-id, « Tam-tam »
Tamar, « Civette » Yves, « Jedi » Eugene, « Chewbacca » Jean-Marc, « Batman » Walter, et
« Dites-Edith » Edith. Merci à vous tous de m’avoir supporté pendant toutes ces (courtes)
années (six ans, je n’en reviens pas !).
Je remercie chaleureusement l’équipage du Belgica, composé d’hommes dévoués, qui me
laisseront des souvenirs de moments hors du commun, parfois surréalistes (surtout au cours de
nos premières sorties…).

7
ACKNOWLEDGEMENTS-REMERCIEMENTS
Je remercie les GMs de mon Comité d’Accompagnement, les Drs Christiane Lancelot et Guy
Josens, pour leur constante attention tout au long de ma thèse.
Je remercie le FRIA, qui m’a accordé une bourse de thèse pendant les premières années de
celle-ci.
Je remercie la fondationVan Buuren qui m’a apporté un ballon d’oxygène non négligeable
pendant les mois de disette.
Je remercie l’ONEM pour son soutien financier (pendant les nombreux mois de disette encore
et quand le ballon d’oxygène s’est envolé…).
Je remercie la Communauté Française de Belgique pour la bourse de voyage qu’elle m’a
accordée, et qui m’a permis de -presque- joindre les deux bouts au cours de mes séjours
monégasques.
Je remercie le Dr Gwenaëlle Leclercq et le presque-Dr Grégory Sempo pour leur amitié si
simple qu’elle m’a fait oublier bien des prises de têtes. Spéciale dédicace au petit Loup, dont
les piles ne sont jamais à plat pour jouer avec ma petite Zoé.
Je remercie mon cousin David pour le soin avec lequel il a court-circuité mon ordinateur tout
neuf au champagne –s’il vous plaît- à quelques semaines du dépôt de ma thèse (je t’avais dit
que je te louperais pas !!).
Je remercie Chanda & Maxime, de chez Macline, et qui ont réussi à sauver in extremis toutes
mes données après l’incident cité ci-dessus.
Je remercie ma sœur, Muriel, pour sa présence rayonnante dans les moments très difficiles qui
ont émaillés la période de thèse. Gilles t’es quelques lignes au dessus…
Je remercie ma marraine Janine, ainsi que Jacques pour nous avoir accueillis mille fois autour
d’un incroyable festin qui a toujours eu l’art de nous remonter le moral à l’infini.
Je remercie ma belle-famille, Monique –notre ange gardien-, Aude et Yvan, toujours prêts à
nous changer les idées.
Je remercie évidemment tout le reste de la famille pour l’affection et l’attention qu’elle m’a
toujours prodigué.
Je remercie mes parents-adorés qui m’ont soutenu sans faillir depuis de nombreuses années (je
n’ose même pas les compter) et m’ont montré l’importance de l’ouverture d’esprit, de la
curiosité, de la ténacité et du bon vin.
Enfin, bien sûr, je remercie Céline, avec qui j’ai traversé en très peu de temps les plus dures
tempêtes, mais aussi les plus grandes joies de ma vie. Je crois qu’après ce que nous avons vécu,
aucun ouragan ne pourra nous éloigner.

8
ACKNOWLEDGEMENTS-REMERCIEMENTS
Zoé, mon bébé-crabe, et Liam, mon bébé-grogne, papa vous dédie ce travail.

TABLE OF CONTENTS
9
TABLE OF CONTENTS
ACKNOWLEDGEMENTS-REMERCIEMENTS........................................................................................................5
TABLE OF CONTENTS................................................................................................................................................9
SUMMARY ..................................................................................................................................................................11
I. GENERAL INTRODUCTION ..........................................................................................................................15
I.1. POLYCHLORINATED BIPHENYLS.........................................................................................................................15
I.1.1. General information ..................................................................................................................................15
I.1.2. Analysis ......................................................................................................................................................16
I.1.3. International recommendations ................................................................................................................18
I.2. PCBS IN THE MARINE ENVIRONMENT................................................................................................................19
I.2.1. Caracterization of PCB contamination ....................................................................................................20
I.2.2. Biological effects of PCBs.........................................................................................................................22
I.2.3. Biomarkers of PCB exposure....................................................................................................................27
I.3. CONTAMINATION OF THE NORTH SEA BY PCBS ...............................................................................................31
I.3.1. The North Sea ............................................................................................................................................31
I.3.2. Origin and fluxes of PCB contamination in the North Sea......................................................................32
I.3.3. PCBs in benthic ecosystems of the North Sea ..........................................................................................33
II. OBJECTIVES ................................................................................................................................................35
III. EXPERIMENTAL CONDITIONS...............................................................................................................37
III.1 NON-COPLANAR VS. COPLANAR CONGENER-SPECIFICITY OF PCB BIOACCUMULATION AND
IMMUNOTOXICITY IN SEA STARS ..............................................................................................................................39
III.2 DELINEATION OF PCB UPTAKE PATHWAYS IN A BENTHIC SEA STAR USING A RADIOLABELLED CONGENER
...................................................................................................................................................................................59
III.3 COPLANAR PCB UPTAKE KINETICS IN THE COMMON SEA STAR ASTERIAS RUBENS AND SUBSEQUENT
EFFECTS ON ROS PRODUCTION AND CYP1A INDUCTION.......................................................................................73
III.4 CONTRASTING EFFECTS OF COPLANAR VS NON-COPLANAR PCB CONGENERS ON IMMUNOMODULATION
AND CYP1A LEVELS (DETERMINED USING AN ADAPTED ELISA METHOD) IN THE COMMON SEA STAR ASTERIAS
RUBENS L. .................................................................................................................................................................95
IV. FIELD CONDITIONS.................................................................................................................................113
IV.1 CONTAMINANT LEVELS IN SEDIMENTS AND ASTEROIDS (ASTERIAS RUBENS L., ECHINODERMATA) FROM
THE BELGIAN COAST AND SCHELDT ESTUARY: POLYCHLORINATED BIPHENYLS AND HEAVY METALS...............115
IV.2 BIOACCUMULATION AND EFFECTS OF PCBS AND HEAVY METALS IN SEA STARS (ASTERIAS RUBENS, L.)
FROM THE NORTH SEA: A SMALL SCALE PERSPECTIVE..........................................................................................141
IV.3 ECHINODERMS AS BIOINDICATORS, BIOASSAYS AND IMPACT ASSESSMENT TOOLS OF SEDIMENT-
ASSOCIATED METALS AND PCBS IN THE NORTH SEA............................................................................................159
IV.4 LEVELS AND EFFECTS OF PCDD/FS AND C-PCBS IN SEDIMENTS, MUSSELS AND SEA STARS OF THE
INTERTIDAL ZONE IN THE SOUTHERN NORTH SEA AND THE CHANNEL.................................................................185
V. GENERAL DISCUSSION ..........................................................................................................................205
THE BIOACCUMULATION OF PCBS IN SEA STARS...................................................................................................205
THE EFFECTS OF PCBS IN SEA STARS .....................................................................................................................208
CONCLUSIONS-RECOMMENDATIONS......................................................................................................................211
VI. REFERENCES.............................................................................................................................................213
VII. ANNEX STUDIES ......................................................................................................................................249
VII.1 BIOACCUMULATION OF PCBS IN THE SEA URCHIN PARACENTROTUS LIVIDUS: SEAWATER AND FOOD
EXPOSURES TO A 14 C-RADIOLABELLED CONGENER (PCB 153).............................................................................251
VII.2 BIOACCUMULATION OF PCBS IN THE CUTTLEFISH SEPIA OFFICINALIS FROM SEAWATER, SEDIMENT AND
FOOD PATHWAYS. ...................................................................................................................................................263

10
TABLE OF CONTENTS
VII.3 MEASUREMENT OF EROD ACTIVITY: CAUTION ON THE SPECTRAL PROPERTIES OF THE STANDARDS USED
.................................................................................................................................................................................279
VII.4 EFFECTS OF PCBS ON REACTIVE OXYGEN SPECIES (ROS) PRODUCTION BY THE IMMUNE CELLS OF
PARACENTROTUS LIVIDUS (ECHINODERMATA).....................................................................................................287
APPENDIX I: CAPTIONS TO FIGURES...............................................................................................................299
APPENDIX II: CAPTIONS TO TABLES ...............................................................................................................304
APPENDIX III: CAPTIONS TO EQUATIONS......................................................................................................308

SUMMARY
11
SUMMARY
PCBs are among the most problematic marine contaminants. Converging towards the oceans
via the rivers and the atmosphere, they concentrate in sediments where they become a
permanent threat to organisms living at their contact. PCBs are extremely resistant,
bioaccumulated and some congeners are considered as highly toxic. The North Sea is
considered as a highly contaminated area; however little information is available regarding
the impact of PCBs on key benthic organisms of this region.
Ubiquist, abundant and generally recognized as a good bioindicator species, the common NE
Atlantic sea star Asterias rubens (L.) is an ecosystem-structuring species in the North Sea and
was chosen as an experimental model. The present study focused on the characterization of
PCB bioaccumulation in A. rubens exposed through different routes (seawater, food,
sediments) and on subsequent biological responses, at immune and sucellular levels. The
considered responses were respectively (i) the production of reactive oxyggen species (ROS)
by sea stars amoebocytes, which constitutes the main line of defence of echinoderms against
pathogenic challenges and (ii) the induction of a cytochrome P450 immunopositive protein
(CYP1A IPP) which, in vertebrates, is involved in PCB detoxification.
Experimental exposures carried out have shown that A. rubens efficiently accumulates PCBs.
Exposure concentrations were always adjusted to match those encountered in the field. PCB
concentrations reached in sea stars during the experiments matched the values reported in
field studies; therefore our experimental protocol was found to accurately simulate actual
field situations. Uptake kinetics were related to the planar conformation of the considered
congeners: non-coplanar PCB uptake was described using saturation models, whereas
coplanar PCBs (c-PCBs) were bioaccumulated according to bell-shaped kinetics. Non-
coplanar congeners generally reached saturation concentrations whithin a few days or a few
weeks, which means that sea stars can be used to pinpoint PCB contamination shortly after
occurrence. On the other hand, c-PCB concentrations reached a peak followed by a sudden
drop, indicating the probable occurrence of c-PCB-targeted metabolization processes in sea
stars. Our experimental studies also demonstrated that seawater was by far the most efficient
route for PCB uptake in sea stars and that even if PCB levels in seawater are extremely low
compared to sediment-associated concentrations, seawater constitutes a non-negligible route

12
SUMMARY
for PCB uptake in marine invertebrates. Among the different body compartments, bodywall
displayed the highest bioaccumulative potency and can therefore be considered as particularly
interesting for field biomonitoring applications. Rectal caeca, which play a central role in
digestion and excretion processes in sea stars, have also rised particular interest as results
suggest these organs could be involved in the elimination of PCB 77 degradation products.
The field work carried out during the present study showed that PCB concentrations measured
in A. rubens tissues reflect environmental levels of certain congeners. As it was the case in
experimental conditions, A. rubens differentially accumulated PCB congeners according to
their planarity. Strong relationships were found between concentrations measured in
sediments and those determined in sea stars body wall for certain non-coplanar congeners
(e.g. 118 and 138), thus allowing to consider A. rubens as a suitable bioindicator species for
medium-chlorinated PCB congeners. On the other hand, sea stars appeared to be able to
regulate -to a certain extent- their content in coplanar PCBs. This implies that (i) A. rubens
cannot be strictly considered as an indicator organism for c-PCBs and (ii) c-PCBs probably
affect essential aspects of sea star biology, potentially leading to deleterious effects.
The present study addressed effects of PCB exposure on A. rubens biology, in both
experimental and field conditions. In experimental conditions, PCBs were found to
significantly alter ROS production by sea stars amoebocytes. This alteration also occurred in
a congener-specific way: c-PCBs were found to significantly affect, and probably impair sea
stars immune system, whereas non-coplanar congeners had no effect. In the field, the PCB
contribution to immunotoxicity could not be determined because none of our studies
considered ROS production along with c-PCB concentration measurements. However, the
levels of ROS production by sea stars amoebocytes measured in field and experimental
conditions were found to potentially lead to altered immunity, and therefore to impair sea
stars defence against pathogenic agents.
A specially designed ELISA was used to measure CYP1A IPP in experimental and field
conditions. Experimental work has shown that the induction of this protein was related to
PCB exposure in a congener-specific fashion: c-PCBs alone were found to strongly induce
the production of CYP1A IPP according to a dose-dependent relationship. These results have
highlighted many similarities between the dioxin-like responsiveness of CYP1A IPP
induction in sea stars and that occurring in vertebrates. This strongly suggests similarities in
the toxicity-triggering mechanism of dioxins and c-PCBs. In the field, CYP1A IPP induction
was found to be significantly related to PCB levels determined in bottom sediments. It can
thus be considered as a valuable biomarker. Further research is however needed to better

13
SUMMARY
characterize the influence of physico-chemical and physiological parameters on CYP1A
induction to refine the interpretation of the information gathered via this biomarker.
Results obtained in our study have lead to questionning international regulations applying to
PCB biomonitoring in the marine environment. For instance, we strongly suggest that the
selection of congeners to be systematically considered should be revised to include c-PCBs.
Indeed, in our experiments PCB toxicity was almost always attributable to the sole c-
congeners. Historically, determination of c-PCB concentrations was extremely difficult due to
analytical limitations; however, nowadays, these problems have been overcome and do no
more justify their exclusion from monitoring studies.
Although A. rubens appeared to be quite resistant to PCB contamination, levels measured in
sea stars from the southern North Sea can possibly affect their immune and endocrine systems
in a subtle way, but with relatively low risk for this species at the short-term. However, this
does not mean that other species in this region undergo similarly low risks, or that sea star-
structured ecosystems may not become affected in the long-term.

I. GENERAL INTRODUCTION
I.1. Polychlorinated biphenyls
I.1.1. General information
15
GENERAL INTRODUCTION
Polychlorinated biphenyls (PCBs) have been in use in the industry since the 1930s, in
electrical equipment and in the manufacture of paints, plastics, adhesives, coating compounds
and pressure-sensitive copying paper (Clark 1997). Their remarkable physico-chemical
properties (very high stability, excellent electric and thermic insulation) led to their
proliferation in the industry to which they were marketed under various trade names
(Metcalfe 1994): Aroclor (Monsanto, United States), Clophen (Bayer, Germany), Phenoclor
(Caffaro, Italy), Pyralene (Prodelec, France), Kanechlor (Kanegafushi, Japan), Sovol
(Russia).
These commercial PCBs vary in texture from clear oils to powders, according to the needs of
industrial applications. These mixtures also vary in composition, containing between 20 and
60 % per weight of chlorine, with a varying number of chlorine atome per molecule.
Theoretically, PCBs include 209 possible compounds (congeners) with varying degree and
pattern of chlorine substitution (Fig. 1). Ballschmiter & Zell (1980) set up a nomenclature
system for PCBs that assigns each congener a number comprised between 1 and 209. This
system was adopted by the International Union of Pure and Applied Chemistry (IUPAC).
Figure 1. Numbering system for sites of chlorine on a biphenyl molecule (Metcalfe 1994)
PCBs were first reported in environmental samples in the 1960s (Jensen, 1966). Although
there are 209 possible PCB congeners, only around 90 of them have been detected in the

16
GENERAL INTRODUCTION
environment. The relative abundance of the different congeners in environmental samples
depends on factors such as the congener abundance in the initial commercial products, the
relative sales and uses of the products, and the relative persistence and transport in the
environment of the compounds.
Due to high persistence and relative volatility, on the long term PCBs may be transported
over considerable distances, as a consequence of mass movements of air or water. These
movements can also occur by diffusion, which may be very localized, but can take place over
large distances, especially in air. PCBs have been detected in the most remote regions, such as
the Arctic, the North Atlantic and even the Antarctic and deep-sea (e.g. AMAP 1998), where
there is no anthropogenic emission sources.
Data on the global production and use of PCBs has been collected for decades, but more work
is needed for the interpretation of past, present and future contamination levels around the
world: it is likely that PCB compounds will remain in the environment for a very long time
(Cummins 1988, Tanabe 1988, Voldner & Li 1995, Wania & Mackay 1996, Vallack et al.
1998). In the late 1980s, estimations indicated that there were still 374,000 tons of PCBs in
the environment, of which 232,400 tons dissolved in seawater, 3,500 tons dissolved in
freshwater, and 1580 tons circulating in the atmosphere (Tanabe 1988). According to the
same author, 783,000 additional tons of PCBs were remaining in storages or in landfills, of
which an undetermined part could be released in the environment.
I.1.2. Analysis
Most analyses of PCB levels in the environment have been reported as Aroclor equivalents.
This was once due to necessity because traditional packed-column gas chromatography (GC)
were not able to resolve individual PCB congeners. This lack of resolution limited the
capacity of analyses to accurately describe environmental PCB levels and patterns. Moreover,
these analyses were mostly based on Aroclor peaks from the packed-column chromatogram,
assuming that ratios among PCB congeners in the environment were the same as those found
in commercial mixtures (Metcalfe 1994).
It is now well-known that, once released in the environment, the composition of a commercial
PCB mixture changes over time, since different congeners display very different physico-
chemical properties (e.g. water solubility, vapor pressure, tendency to sorb to organic matter).
Individual congener partition behaviour differs among water, air and solid phases (Dickhut et
al. 1986, Lara & Ernst 1989, Brunner et al. 1990). Moreover, some congeners undergo

17
GENERAL INTRODUCTION
dechlorination by anaerobic bacterial action when present in sediment at threshold
concentrations, while others do not (Brown et al. 1987, Quensen et al. 1990, Mohn & Tiedje
1992). Also, “light” congeners can be subject to aerobic degradation in particular conditions
(Furukawa 1982). Consequently, original PCB mixtures become depleted in the most
degradable congeners and enriched in metabolites of the latter ones and in “resistant”
congeners. In biota, PCB composition can also be altered, depending on the uptake,
metabolization and depuration rates of individual congeners. PCBs in the environment take on
a congener composition that becomes dissimilar to the original Aroclor mixture. Thus, since
the early 1990s, congener-specific analysis of PCBs has progressively replaced traditional
Aroclor-equivalent based methods (Duinker et al. 1991, Eganhouse & Gossett 1991).
For routine analyses of PCB congeners in marine biota samples, the commonly used methods
are high resolution gas chromatography with capillary columns and electron capture detection
(HRGC-ECD) or high resolution gas chromatography with low resolution electron impact
mass spectrometry in selected ion mode (HRGC-LRMS-SIM) (Metcalfe 1994). Most
individual PCB congeners can be resolved using these methods at low parts-per-billion
concentrations (Schultz et al. 1989). An advantage of the ECD over LRMS for PCB analysis
is that it is halogen sensitive (Cairns et al. 1989): many coextractive compounds (e.g.
polynuclear aromatic hydrocarbons, phthalates) are not detected. A disadvantage of the
method over HRGC-LRMS is that the response is highly dependent on the degree and pattern
of chlorination, reducing sensitivity and accuracy of the method for lesser chlorinated
congeners.
Another approach used to address the toxic potential of PCBs is the use of toxic equivalency
(TEQ). In this approach, the biological or toxic potencies of individual congeners are
expressed related to a benchmark contaminants, usually 2,3,7,8 tetrachloro-dibenzo-p-dioxin
(TCDD), an extremely potent toxicant (Fig. 2). Using a variety of endpoints or responses, a
relative biological potency or toxic equivalency factor (TEF) can be determined for each
congener. The TEQ approach is an attempt to provide integrated assessment of the toxic
potential of environmental mixtures. It relies on a number of assumption, including the
absence of non-additive interactions (i.e. possible synergism or antagonism is not taken into
account) among the components of the mixture (Safe 1990, Ahlborg et al. 1992, 1994).
TCDD equivalents are being used increasingly in risk assessments as a replacement for
exposure measures based only on TCDD or total PCBs (Barron et al. 1994, Van den Berg et
al. 1998).

Figure 2. Molecular configuration of 2,3,7,8 TCDD and PCB 169 (Metcalfe 1994)
I.1.3. International recommendations
18
GENERAL INTRODUCTION
Persistent Organic Pollutants (POPs) is the common name refering to a group of organic
contaminants that comprises PCBs. POPs are semi-volatile, bioaccumulative, persistent and
toxic (Vallack et al. 1998). Although the occurrence of POPs at elevated levels is of great
concern in “hot spots”, the POPs issue has received increasing attention at regional and global
scales in the last decades (Wania & Mackay 1996, UNECE 1998, UNEP 2001).
Due to their beyond-boundaries transport, political problems have also arisen. International
agreements have thus come into effect, such as the 1998 Aarhus Protocol on POPs (UNECE,
1998). The overall and long-term objective of the Aarhus Protocol on POPs is to eliminate
any discharge, emission and loss of POPs to the environment. The international community
has called for action to reduce and eliminate production, use and releases of these substances
through: (i) the Protocol to the regional UNECE Convention on Long-Transboundary Air
Pollution (CLRTAP) on POPs, opened for signatures in June 1998 and (ii) the global
Stockholm Convention on POPs, opened for signatures in 2001. These instruments establish
strict international regimes for initial lists of POPs (16 in the UNECE Protocol and 12 in the
Stockholm Convention). Both instruments also contain provisions for including additional
chemicals into their list. They lay down the following control measures: prohibition or severe
restriction of the intentional production of POPs and their use, restrictions on export and
import of the intentionally produced POPs (Stockholm Convention) , provisions on the safe
handling of stockpiles (Stockholm Convention), provisions on the environmentally sound

19
GENERAL INTRODUCTION
disposal of POPs wastes and provisions on the reduction of emissions of unintentionally
produced POPs (e.g. dioxins and furans).
Regarding PCBs, the International Council for the Exploration of the Sea (ICES) has
recommended that congeners 28, 52, 101, 153, 138 and 180 should be selected for routine
analysis (Duinker et al. 1988). Several European Union (EU) countries have adopted these
congeners, with the addition of congener 118, for defining maximal levels of PCBs in edible
marine resources. These environmental quality standards and other international
commitments also arise from the 1984 International Conference on the Protection of the
North Sea, the 1995 Barcelona Convention for the Protection of the Mediterranean Seas
against Pollution, the Baltic States HELCOM, etc.
One of the most frequent objectives of monitoring is to assess seafood quality using estuarine
and marine water and sediments as a check for sources of possible pollution. The recent
emphasis on the monitoring of non-ortho and mono-ortho PCB congeners has necessitated an
expansion of the list of congeners to be considered in routine analysis. Because of their high
toxic potential (Safe 1990), it is most probable that all non-ortho substituted congeners should
be included in analysis programmes (Metcalfe 1994).
I.2. PCBs in the marine environment
The ultimate sink for many contaminants is the marine environment, following either direct
discharges or hydrologic and atmospheric processes (Stegeman & Hahn 1994). Since the late
1960s, PCBs are known to be present in substantial quantities in marine sediments, as well as
in marine biota (Jensen et al. 1969). PCBs accumulate in the organic phase, such as biota and
the organic fraction of sediments, transfering between these compartments according to the
model presented in Fig. 3. PCBs persist in the marine environment for several decades: most
PCBs only exist in trace concentrations, but all have extensive half-lives (degradation half-
lives ranging up to 200 years) in the environment (Howard et al. 1991, Haynes et al. 2000, Oh
2000, Moore et al. 2002, Wania & Daly 2002).

20
GENERAL INTRODUCTION
Figure 3. Contaminants transfers between compartments in a coastal model (Moore et al. 2002)
I.2.1. Caracterization of PCB contamination
a. Seawater
PCBs are hydrophobic compounds, i.e. they have extremely low water solubilities.
Concentrations in ocean water are generally very low, making reliable quantification
technically difficult. PCB concentrations in filtered ocean water are usually reported to be in
the low pg l -1 range. In contrast, PCBs are highly lipophilic and adsorb readily onto particles.
Their distribution in sea is thus far from being uniform.
The sea surface microlayer (SSM) is a film varying from a few µm to 1 mm in thickness. It is
extremely difficult to study, but is known to contain high levels of particulate organic carbon
and lipids compared to bulk water, thus allowing PCBs to accumulate (Daumas et al. 1976,
Hardy et al. 1988, Xhoffer et al. 1992, Garabetian et al. 1993). Elevated levels of dissolved
organic contaminants in the SSM have been reported with enrichment factors reaching one to
three orders of magnitude for PCBs (Duce et al. 1972, Bidleman 1973, Napolitano 1995).
While the total quantity may not be great, the PCB enrichment of the SSM may be of
considerable importance to surface-living organisms. Where water masses with variable
physico-chemical characteristics meet, they form a front where floating material gets
accumulated including surface oil. Fronts have a high productivity and attract a wide range of
animals, which thus receive a PCB-enriched diet. Since the upper millimetre of the sea is also
enriched in microorganisms and zooneuston (including larvae), great concern has been

21
GENERAL INTRODUCTION
expressed on the toxic effects of the high contaminant levels in the SSM (Hardy et al. 1990,
Hardy & Cleary 1992, Stebbing et al. 1992). The PCB enrichment in SSM microorganisms
also poses analytical difficulties in distinguishing the portion that is incorporated and the one
that is adsorbed onto it. The former may affect them, but the latter is bioavailable to animals
feeding on the contaminated organisms. PCBs adsorbed onto inorganic particles may
ultimately be carried to the seabed, which acts as a sink for these compounds. Moreover,
suspended or re-suspended particles are commonly ingested by filter-feeding animals,
entering food chains by this route.
b. Sediments
Sediments are repositories for physical and biological debris and are considered as sinks for a
wide variety of chemicals (Clark 1997). The concern associated with PCBs sorption to
sediments is that many organisms spend a considerable portion of their life-cycle on or in
marine sediments. This provides a path for PCBs to reach higher trophic levels. Direct
transfer of contaminants from sediments or interstitial water to organisms is considered to be
a major route of exposure (Walker & Peterson 1994). PCBs are present in much higher
concentrations in sediments than in overlying water. Sorption to sediments is the predominant
removing mechanism for PCBs from the water column. The analysis of PCBs in sediments
has the advantage of integrating time variations. Once contaminated, sediments can act
themselves as a slowly releasing source of PCBs, which causes chronic exposure of biota long
after the primary source of contamination has discontinued (Moore et al. 2002).
c. Organisms
As a consequence of their hydrophobic and persistent characteristics PCBs are
bioaccumulated and high concentrations are found in biota (Stebbing et al. 1992, Clark 1997,
OSPAR 2000). PCBs are efficiently accumulated by marine organisms by absorption across
outer surfaces (e.g. gills, skin), or by ingestion of contaminated food, seawater or sediments.
Once they have entered the organism, PCBs are stored within the fatty tissues, or in other
lipophilic sites, such as cell membranes or lipoproteins. In the long term, release from storage
may occur (e.g. in times of low food availability) during which organisms mobilize and use
their fat reserves, so increasing the concentration of PCBs in their body up to possibly
harmful levels (Walker et al. 1996). Delayed toxicity may therefore be observed some time
after initial exposure to the contaminant. Organisms have the capacity to bioaccumulate and
to biomagnify PCBs, which results in body concentrations several orders of magnitude higher

22
GENERAL INTRODUCTION
than in seawater or in the food (OSPAR 2000). In marine animals, contaminants tend to
concentrate in specific organs (Walker et al. 1996).
The fact that PCBs accumulate preferentially in fatty tissues implies that caution must be
taken in comparing levels of contamination in different organisms. Different amounts of
PCBs can be accumulate in the various organs, having quite different implication for a fat
animal than for an emaciated one. Accumulation rates vary among species, but also within a
species according to factors such as age, sex, stage in the breeding cycle, as well as exposure
concentrations or feeding habits (Van der Oost et al. 2003). Bioaccumulation is a precursor to
all chemical toxicity: without some degree of accumulation, even if slight, toxic action in
organism target site(s) cannot take place.
I.2.2. Biological effects of PCBs
Experimental studies have shown that PCBs are capable of producing a wide variety of toxic
effects in exposed organisms, some of the most common include neurotoxicity, immune
dysfunction, reproductive and developmental effects, and cancer (Harding & Addison 1986,
Zabel et al. 1995, Chapman 1996, Krogenaes 1998, Coteur et al. 2001). PCBs are of concern
primarily because of their potential for causing chronic effects following long-term, low-level
exposure (Walker et al. 1996, OSPAR 2000). The effects of substances on biota are
dependent on a number of factors and processes including bioavailability, bioaccumulation,
toxic potency and the capacity of the organism to metabolize the substance (Fig. 4). Marine
contamination by PCBs poses a relatively well-documented risk to the health of marine
organisms, which can occur at levels ranging from subcellular effects to ecosystem effects
(Tanabe & Tatsukawa 1992, Elkus et al. 1992, Norstrom & Muir 1994, Bello et al. 2001).
Figure 4. Model describing the fate of lipophilic xenobiotics in organisms (Hodgson & Levi 1993)

a. Subcellular and cellular effects
23
GENERAL INTRODUCTION
To gain a full understanding of the toxic effects of a chemical, it is necessary to link initial
molecular interactions to consequent effects at higher levels of organization. The extent to
which such a molecular interaction occurs is, in general, related to the dose received, although
the relationship is rarely a simple one (Walker et al. 1996). Molecular interactions between
the xenobiotic and sites of action, which lead to toxic manifestations, may be highly specific
for certain types of xenobiotics and organisms or non-specific, because of the variety of sites
of action, which can occur in one species and not in other ones (Fig. 5).
Figure 5. Pathways for activation and detoxification of organic chemicals (Walker et al. 1996)
Subcellular effects of pollutants can be out of two types: those which serve to protect the
organism against the harmful effects of the chemical (viz. detoxification via e.g. induction of
monooxygenases or induction of metallothioneins), and those which do not (e.g. inhibition of
AchE, formation of DNA adducts) (Table 1). Protective mechanisms function by reducing the
contaminant concentration in the cell (e.g. some PCB congeners induce enzymes that
metabolize them) or by reducing the bioreactive fraction of the contaminant concentration.
One of these mechanisms is achieved through the monooxygenase system, whose function is
to increase the rate of production of water-soluble metabolites and conjugates of low toxicity,
which can be excreted. However, in some cases, metabolism leads to the production of highly
reactive metabolites, that can cause more damage than the parent compound.

Table 1. Protective and non-protective responses to chemicals (Walker et al. 1996).
Type of effects Example Consequences
24
GENERAL INTRODUCTION
Protective Induction of monooxygenases Increase in rate of metabolism of pollutant to more watersoluble
metabolite and thus increase in rate of excretion
Induction of metallothionein Increase the rate of binding sites with metals to decrease
bioavailability
Non-protective Inhibition of AChE Toxic effects seen above 50% inhibition
Formation of DNA adducts May cause harmful effects if leading to mutation
These chemical surveillance systems have evolved as mechanisms for recognizing a broad
range of chemical structures and initiating appropriate responses, such as the
biotransformation and elimination of toxic compounds (Brattsen 1979, Nebert & Gonzalez
1987, Gonzalez & Nebert 1990). The enzymatic components of this inducible
biotransformation system are now well-known and include monooxygenases in the
cytochrome P450 (CYP) superfamily as well as conjugating enzymes such as the
glutathionetransferases and glucuronosyltransferases. The sensory component of this system
consists of soluble receptors that regulate the expression of the biotransformation and
transporter genes in response to environmental chemicals. These receptors include several
members of the steroid/nuclear receptor superfamily (Kliewer et al. 1999a,b, Savas et al.
1999, Waxman 1999, Honkakoski & Negishi 2000) as well as the aryl hydrocarbon receptor
(AhR, Fig.6).
Figure 6. Hypothesized induction mechanism of CYP1A (Bucheli & Fent 1995)
The adaptive function of the AhR has been studied for more than 30 years, leading to the
prediction and then discovery of the AhR as an ‘induction receptor’ that controls the

25
GENERAL INTRODUCTION
induction of adaptive enzymes, especially CYP1A (Poland et al. 1976, Whitlock 1999). The
AhR is now known to recognize an impressive range of chemical structures, including non-
aromatic and non-halogenated compounds (Denison et al. 1998). In regulating
biotransformation enzymes, the AhR serves an important adaptive function, but the function
of this protein is much more complex: studies dealing with the toxicities associated with
exposure to TCDD and related compounds showed that these chemicals are interfering with
important physiological functions in addition to inducing biotransformation enzymes (Poland
& Knutson 1982, Pohjanvirta & Tuomisto 1994).
Physiological and morphological parameters are higher-level responses that follow chemical
and cellular interactions. They are generally indicative of irreversible damages (Hinton et al.
1992). When a pollutant enters a cell, it may trigger certain biochemical responses, or it may
be stored within a compartment, preventing interferences with essential biochemical
components of the cells.
Many alterations may persist even after the exposure to a toxicant has ceased so that host
responses to prior toxicity can also be used to determine effects. Responses are relatively
easily recognized, provided that proper reference and control data are available. Nowadays,
sufficient information is at hand to assemble cellular or histopathological biomarker
approaches and to apply them in integrated field studies (Hinton 1994).
b. Immunological effects
The immune system of an organism maintains a close and efficient surveillance of the body in
order to react against infection and infestation or eliminate dysregulated protein expression.
This system can be divided into two forms of immunity: acquired -or specific- immunity and
innate -or nonspecific- immunity (Roitt et al. 1993). Acquired immunity provides rapid,
specific, and selective reaction against a given infectious agent, but requires a previous
exposition to the same agent. Innate immunity is less specific, but protects the organism
whithout previous contact with the infectious agent. Innate immunity is present in all
metazoan animals, whereas acquired immunity would be present only in vertebrates.
Owing to the complexity of the immune system, several authors have suggested using a tiered
approach for examining immunotoxicity in mammals and lower vertebrates (Vos 1980, Miller
1985, Luster et al. 1988, Weeks et al. 1992). Although invertebrate immunity relies on the
innate system for host defence there are wide ranging strategies for eliminating microbes.
Assays of immunocompetence for invertebrates can therefore be approached at different
levels of organization (Pipe et al. 1995): (1) the apparatus for immunity (total/differential

26
GENERAL INTRODUCTION
blood cell counts, hemopoietic tissues), (2) the mechanisms of immunity (phagocytosis, blood
cell proliferation, release of antimicrobial molecules), and (3) the efficiency of the immune
response (susceptibility to infection by model agents). Most studies of environmental
modulation of the immune function in marine invertebrates have focused on heavy metals, but
some organic compounds have been adressed, both in laboratory and field studies (Fischer
1988, Anderson 1993, Pipe & Coles 1995, Coteur et al. 2001, 2003a).
c. Individual effects
Chemically-induced disorders have been attributed to PCBs at the level of the individual. The
existence of repairing and detoxification mechanisms (e.g. mixed-function oxidases) involves
that a biological response measured at a given level of biological organization might not be
detected at a higher organization level (Luoma & Carter 1991, George & Olsson 1994,
Goldstein 1995). Effects on individuals integrate these latter mechanisms and are therefore
highly relevant of the actual deleterious effects of a contaminant from a biological viewpoint,
but these responses are generally slow and not specific of a given contaminant. Individual
responses include for instance direct increase in mortality rates or interference with processes
of resource acquisition. These effects may result in slower population growth or population
decline. At the individual level, fertilization rate or embryonic development are commonly
used as markers, and represent a good compromise: these responses are quite fast (from a few
hours to a few days) and ecologically relevant because the reproductive success and the
maintaining of populations rely directly on these processes (Dinnel et al. 1988, Gray 1989,
Langston 1990, Weis & Weis 1991, Warnau et al. 1996a).
d. Population and ecological effects
The presence of PCBs into the marine environment is known to provoke toxic effects in biota,
which vary with the intensity and duration of exposure (Long et al. 1995). PCBs may exert
dramatic effects on relatively tolerant species (as determined by laboratory testing) by a
number of ecological mechanisms (Walker et al. 1996). Indeed, the direct influence of
contaminants on predators and grazers can lead to cascading indirect effects on more tolerant
species in other trophic levels. The direct effects of contaminants on sensitive species may
also alter competitive interactions within the resistant populations of producers and
consumers of a given community. Similarly, disturbance rates or resource availability may be
influenced by the presence of contaminants such as PCBs, leading to important modifications
in ecosystem processes, such as decomposition rates of the organic matter, oxygen dynamics

27
GENERAL INTRODUCTION
and nutrient cycling (Walker & Livingstone 1992). It has also been suggested that localized
toxicant-induced mortality may alter metapopulation dynamics, and have significant impacts
on non-exposed groups (Spromberg et al. 1998). Mechanisms associated to population and
community dynamics can vary in potentially complex fashion following PCB exposure. The
presence of PCBs in marine ecosystems can clearly cause a wide range of indirect ecological
effects that can be as or more significant than the direct (toxic) effects triggered by the
contaminant (Feldman et al. 2000).
It is now widely adopted that communities and ecosystems are much more than the sum of
their discrete parts and potentially intense indirect influences of realistic PCB exposures
should be incorporated into an integrated ecotoxicological approach (Walker et al. 1996) .
I.2.3. Biomarkers of PCB exposure
The need to detect and assess the impact of contaminations in the marine environment has led
to the development of markers of biological effect (biomarkers) at various organizational
levels (Huggett 1992, Livingstone 1991, Livingstone et al. 2000). Several definitions have
been given for the term ‘biomarker’, which is generally used in a broad sense to include
almost any measurement reflecting an interaction between a biological system and a potential
hazard, which may be chemical, physical or biological (WHO 1993). A biomarker is defined
as a change in a biological response (ranging from molecular through cellular and
physiological responses to behavioural changes) which can be related to exposure to
environmental chemicals or to their toxic effects (Peakall 1994). According to NRC (1987)
and WHO (1993), biomarkers can be subdivided into three classes:
•Biomarkers of exposure: allow detection and quantitation of an exogenous substance or its
metabolites or the product of an interaction between this xenobiotic and some target
molecules or cells (e.g. DNA or protein adducts, formation of specific metabolites,…),
•Biomarkers of effect: indicate measurable biochemical, physiological or other alterations
within tissues or body fluids of an organism that can be recognized as associated to an
established or possible health impairment or disease (e.g. reproductive, developmental,
endocrine or genetic toxicity),
•Biomarkers of susceptibility: indicate the inherent or acquired ability of an organism to
respond to the challenge of an exposure to a specific xenobiotic substance, including genetic
factors and changes in receptors which alter the susceptibility of an organism to that exposure
(e.g. activity of enzymes implied in activation or detoxification of a specific chemical or
DNA repair capacity for specific types of DNA damage).

Table 2. Biomarkers at different organizational levels (Walker et al. 1996)
Organizational level Example of biomarker
Binding to a receptor TCDD binding to Ah receptor
Nonphenyls binding to oestrogenic receptor
Biochemical response Induction of monooxygenases
28
Vitellogenin formation
Physiological alterations Eggshell thinning
Feminization of embryos
Effects on individuals Behavioural changes
a. Molecular markers of PCB exposure
Scope for growth
GENERAL INTRODUCTION
Molecular markers have been used extensively in environmental monitoring as part of
integrated programmes (Bayne et al. 1988, Hylland et al. 1996, Schlenk et al. 1996). The
main advantages provided by this level of organization are:
• an integrated measure of the bioavailable fraction of contaminants
• the demonstration of causality through mechanistic understanding
• the identification of different routes of exposure and their relative importance
• the detection of exposure to readily metabolized contaminants.
The most widely and best studied biomarker of PCB exposure is the induction of cytochrome
P450 (CYP)-dependent monooxygenase. Payne & Penrose (1975) and Payne (1976) were
among the first to make use of this enzymatic complex as a biomarker, reporting elevated
cytochrome P450 activity in fish from petroleum-contaminated sites. The multiple forms of
CYP catalyze a wide variety of monooxygenation reactions that contribute to cellular
oxidative metabolism in both prokaryotes and eukaryotes (Gibson & Skett 1994, Nelson et al.
1996). The products of the CYP super gene family undertake the oxidation of endogenous
substrates, e.g. fatty acids and steroid hydroxylation but some CYP gene families (CYP1,
CYP2 and CYP3) can also catalyze the oxidation of xenobiotics (Nelson et al. 1996). CYP1
may be induced in organisms exposed to specific aromatic and chlorinated hydrocarbons,
such as dioxins, furans, polyaromatic hydrocarbons (PAHs), or PCBs (Stegeman & Hahn

29
GENERAL INTRODUCTION
1994). An elevation of CYP1A * levels may therefore indicate exposure to these inducers. The
use of CYP1A induction as a biomarker for the pollution of aquatic ecosystems by organic
contaminants has mainly been based on fish.
Whereas the enzyme system and its inductibility have been studied extensively in vertebrates,
less is known in invertebrates (Livingstone et al. 2000). Different invertebrates have been
screened for the occurrence of the CYP1A system. It has been reported in four phyla:
Annelida, Arthropoda, Echinodermata and Mollusca (Lee 1981). Moore et al. (1980) found
that some components of a xenobiotic detoxification system were present in the blue mussel
Mytilus edulis, but with limited metabolizing capacity for organic xenobiotics. The use of
CYP1A as a biomarker was assessed with several molluscs (Stegeman 1985, Livingstone et
al. 1989). Yawetz et al. (1992) observed an induction of CYP1A content by PCBs in
molluscs. Enhanced CYP1A activity was found in the marine polychaete Nereis virens after
exposure to benzo[a]pyrene or PCBs (Lee et al. 1981). Organisms from oil-contaminated sites
showed several times higher CYP1A activities and lacked or had undevelopped gametes
(Fries & Lee 1984). Studies on crustaceans provided controversial results, but in any case,
crustacean CYP1A is less sensitive to induction than fish (James 1989). The CYP1A system
is also present in several echinoderms species (den Besten et al. 1991). Evidence for the
presence of P450 enzymes belonging to the CYP1, CYP2, and CYP3 subfamilies have been
obtained in sea stars (den Besten et al. 1993). Recently, the first echinoderm CYP genes were
identified by Snyder (1998) in digestive tissues of an echinoid (Lytechinus anamesis).
b. Immunological markers
Pollution-induced suppressive effect on the immune system was found to lead to enhanced
disease in organisms (Pipe & Coles 1995). Therefore, immunocompetence assays have been
increasingly used as biomarkers of environmental contamination in the last years. Monitoring
the immune system as a target for toxicity is difficult, given the complexity and self-
regulatory nature of the immune network, so that conventional dose-response relationships
may not always be observed. As with other biomarker responses, immune responses provide
an integrated measure of exposure over time and may reflect the combined results of
simultaneous exposure to several chemicals. It is, however, not possible to determine which
* According to Stegeman et al. (1992), the hydrocarbon-inducible isoenzyme cytochrome P4501A is referred to
as CYP1A. Hitherto, only the respective isoenzyme of rainbow trout (Oncorhyncus mykiss) can conclusively be
termed CYP1A1 (Heilmann et al. 1988), whereas, for all other species, CYP1A is more appropriate.

30
GENERAL INTRODUCTION
chemical has caused the observed effect as none of the changes in immune function can be
attributed to a specific compound or class of chemicals (Wester et al. 1994).
The use of invertebrate immunotoxicology, although of increasing interest whithin
environmental monitoring studies, is still very much in its infancy (Livingstone et al. 2000).
Risk assessment of specific compounds in terms of immunomodulation leading to enhanced
disease susceptibility for individuals or populations whithin a particular ecosystem has not yet
been attempted. Indeed, much of the fundamental information on invertebrate immune
responses and disease susceptibility is not available.
Among the immune functions, oxidative stress is widely investigated. The interest of
oxidative stress in ecotoxicological applications is based on the oxygen paradox: this
molecule is fundamental for many biochemical pathways in aerobic organisms, but its
consumption generates the intracellular formation of potentially toxic reactive oxygen species
(ROS). Despite the fact that basal oxyradical production is normally counteracted by a
complex antioxidant system, several pollutants are known to enhance the intracellular
generation of ROS through different mechanisms including the redox cycle, the cytochrome
P450-dependent oxidative metabolism of aromatic hydrocarbons and the Fenton reaction in
the presence of some transitional metals (Livingstone 1998). From the standpoint of
biomarkers it is useful to understand how antioxidants react to xenobiotic-mediated
enhancement of oxyradical production but the complexity of interactions between pro-oxidant
factors and cellular targets often precludes this possibility. Variations of individuals
antioxidants are difficult to predict and they often vary according to the class of chemicals
tested, species sensitivity and several environmental and biological factors (Winston & Di
Giulio 1991). Induction of antioxidant defences is referred to as a counteracting response of
exposed organisms but the same antioxidants can be depleted when overwhelmed. Depending
on the duration and intensity of the pro-oxidant stressor, antioxidant defences might only be
induced during the first phase of the response, while in other conditions organisms can exhibit
no variations or transitory responses before adaptive mechanisms occur (Regoli & Principato
1995). All these possibilities (and their combinations) have been reported (Winston & Di
Giulio 1991) and the complexity of antioxidant responses to pollutant exposure often leads to
a controversy about the use of oxidative stress in ecotoxicological applications.

I.3. Contamination of the North Sea by PCBs
I.3.1. The North Sea
31
GENERAL INTRODUCTION
The depth of the North Sea is not uniform: shallowest region is located near Dover (30 m),
getting deeper towards the west (up to 100 m) and the north (up to 700 m). Water masses
result from the mixing of NE Atlantic waters, precipitations, and river inputs. Seasonal
variations of salinity are relatively low (salinity remains around 35‰ all year long), except in
coastal regions, where the influence of large estuaries can bring it down to 32‰. Water
masses circulation has been modelized using radionuclide data; grossly, water masses
circulate according to an anti-clockwise direction in the North Sea (Fig. 7).
Figure 7. Diagram of the general water circulation in the North Sea (NSTF 1993)
Approximately 164 million people live in the North Sea catchment area. Numerous large
rivers (e.g. Rhine, Scheldt, Elbe, Thames) flow through this heavily urbanized and
industrialized region, providing significant inputs of several different pollutants into the North
Sea. Hence, the North Sea constitutes the ultimate repository for a large range of domestic
and industrial contaminants.
The southern region of the North Sea (the southern bight) is considered as one of the most
contaminated area. Certain researchers consider that there is no compartment of the southern

32
GENERAL INTRODUCTION
bight (seawater, sediments, biota) which is not altered anthropogenically in a way or another
(Rygg 1985, Kersten et al. 1994).
I.3.2. Origin and fluxes of PCB contamination in the North Sea
Inputs of contaminants in the North Sea occur via three main routes: direct, riverine and
atmospheric inputs. The relative importance of each input route differs among the regions,
and according to contaminants considered (OSPAR 2000). Direct inputs of contaminants arise
mainly as a consequence of municipal and industrial discharges in coastal waters and from
offshore activities and dumping. Riverine inputs extend along the coasts, and constitute
another important contribution to contamination. Atmospheric inputs are an important source
to the marine environment for several substances including heavy metals (e.g. mercury and
lead), PCBs and some nitrogen compounds. The sources of atmospheric inputs may be
located within or outside the North Sea area as PCBs can be transported on a global scale
through the atmosphere.
Ocean currents are also important in the transport and distribution of PCBs in the North Sea.
Although PCB concentrations in seawater are extremely low, the largeness of the water
volumes transported implies that fluxes are large. PCBs, as many other contaminants, get
adsorbed onto particulate matter upon which the transport path and fate of substances largely
depend (Olsen et al. 1982, Balls 1988). The residence time of dissolved substances in the
North Sea is 1 to 3 years (Otto et al. 1990). However, over 70% of the substances associated
with the suspended matter remain in the North Sea or in associated sedimentation areas such
as Wadden Sea, Skagerrak, Norwegian Trench and estuaries (Eisma 1973, Eisma & Kalf
1987, Eisma & Irion 1988). Whereas the Atlantic Ocean is the major source of suspended
matter in the southern North Sea (McManus & Prandle 1997), in the Dutch coastal zone the
dumping of dredged material and the riverine input from the Scheldt and Rhine are also
relatively important sources of suspended matter and associated substances (Eisma 1973,
Eisma & Kalf 1987, Van Alphen 1990, Lourens 1996).
Sediments are subject to resuspension and bioturbation, which can lead to the remobilization
of PCBs (which become available again to organisms) or to their burial in deeper layers of
bottom sediments. Although it is not possible to derive reliable estimates of inputs because
most PCB concentrations are below the limit of detection, estimated fluxes derived for the
North Sea are in the range of 0.13 – 2.4 t yr -1 for the 1990 to 1995 period (OSPAR 2000).

I.3.3. PCBs in benthic ecosystems of the North Sea
33
GENERAL INTRODUCTION
PCB contamination levels in North Sea biota have been mainly characterized in dab (Limanda
limanda), blue mussels (Mytilus edulis), and common sea stars (Asterias rubens) (e.g.
Stebbing et al. 1992, den Besten et al. 2001, Coteur et al. 2003a, Stronkhorst et al. 2003).
Dab is the most abundant flatfish species in the North Sea, with an estimated biomass of
about 2 million tons (Daan et al. 1990). Because it is a demersal fish with a large geographic
distribution and abundance, and because it is sensitive to PCB exposure (Sleiderinck et al.
1995), it has been routinely used in pollution monitoring programmes in the North Sea (North
Sea Task Force; Joint Monitoring Programme) (Stebbing et al. 1992, NSTF 1993a,b).
However, regarding indicating purposes, dabs present a major flaw as they are known to
migrate during spawning periods over relatively long distances, which makes difficult to
determine the origin of the contamination (Rijnsdorp et al. 1992).
The blue mussel (M. edulis) is a sedentary, filter-feeding bivalve of commercial importance,
which has long been considered amongst the best suited sentinel organisms for monitoring
marine pollutions (Goldberg et al. 1978). It has been therefore widely used as bioindicator in
North Sea pollution studies. M. edulis efficiently takes up and concentrates PCBs to levels
well above those present in the surrounding seawater. It provides information on spatial and
temporal pollution trends and enables the identification of contamination «hot spots» in
coastal areas (e.g. Phillips 1990). Moreover, the blue mussel exhibits a series of biochemical
(sublethal) responses to pollutants (see Livingstone 1991 for a review) which may be used as
early warning signals of exposure (McCarthy & Shugart 1990, Huggett 1992).
The common NE Atlantic sea star Asterias rubens (Fig. 8) is also an interesting test organism
because of its key position as top predator in the food chain “seston-mussels-sea stars”. In the
North Sea, this echinoderm is known to influence the structure and functioning of benthic
communities (Menge 1982, Hayward & Ryland 1990, Hostens & Hammerlink 1994). It lives
on or in proximity of sediments (the main reservoir of contaminants in the marine
environment) and can be found in very diversified biotopes from the surface to depths
reaching 650 m. A. rubens is also able to colonize low salinity areas, such as estuaries, which
are under direct influence of contamination carried by large rivers.

34
GENERAL INTRODUCTION
Figure 8. The common NE Atlantic sea star Asterias rubens L. (Hayward & Ryland 1996)
In semi-field studies, A. rubens has been demonstrated to efficiently accumulate PCBs,
leading to deleterious effects on reproductive processes (den Besten et al. 1989, 1990a). This
sea star has largely proved its value or potential value as a bioindicator species for a wide
range of anthropogenic contaminants (e.g. PCBs, metals, organometals) in laboratory and/or
in field studies (e.g. Bjerregaard 1988, Everaarts & Fischer 1989, Temara et al. 1997a,
1998a,b, Warnau et al. 1999, Coteur et al. 2003a, Stronkhorst et al. 2003). Although there is a
wealth of studies showing the quality of the sea star as a bioindicator, no study has
investigated bioaccumulation processes of PCBs in the sea star. However, such data is a
prerequisite to assess the value of A. rubens as a bioindicator of PCB contamination.
Available data on effects of PCB exposure in sea stars are also scarce, but have shown that
these organisms are affected (den Besten 1998, den Besten et al. 1990a, 1993). Existing
studies have focused on the subcellular level using the microsomal activity of the CYP
enzyme system (den Besten et al. 1991, 1993) and steroid metabolism (den Besten et al.
1991). In vitro or in vivo exposure of sea stars to PCBs has elsewhere been reported to
decrease DNA integrity (Sarkar & Everaarts 1995).

II. OBJECTIVES
35
OBJECTIVES
Oceans are the ultimate receptacle for many anthropogenic contaminants that converge
towards them via the rivers and the atmosphere. Often displaying low solubility in seawater,
these contaminants concentrate in sediments, where they become a persistant threat for
benthic communities, particularly in coastal areas. PCBs are among the marine contaminants
of highest concern. Their unique physico-chemical characteristics, at first exploited by the
industry, have rapidly become a cause of concern: PCBs are resistant to degradation, readily
accumulated by marine organisms and highly toxic. Among the 209 possible congeners,
coplanar PCBs display a dioxin-like conformation and have the highest toxic potential.
Information available about the impact of PCBs on marine benthic species in the North Sea is
scarce. In addition most of the existing studies addressed species that are poorly
representative of North Sea benthic ecosystems, which leads to uncertain characterization of
the principal ecological impact of these contaminants.
The common NE Atlantic sea star Asterias rubens (L.) can be considered as an ecosystem-
structuring species in the North Sea. In addition it is ubiquist, abundant and generally
recognized as a good bioindicator species. Therefore, A. rubens was chosen as an
experimental model to study PCB bioaccumulation and effects. Sea stars have already been
used in the field to characterize the contamination status of a region, but no data regarding
bioaccumulation efficiencies, body distribution or relative importance of the different uptake
routes are available in the literature.
The main objectives of the present study were to examine -both in experimental and natural
conditions- the bioaccumulation and body distribution of PCBs in A. rubens, and the
biological consequences attributable to PCB exposure.
For this purpose, the accumulation biokinetics of structurally-contrasting PCB congeners
were studied by exposing sea stars via different exposure routes (seawater, sediments or
food), using either a mixture of PCB congeners (Chap. III.1) or single-congeners (Chap. III.2
& III.3). The effects of PCB exposure on the immune response (ROS production) and on the
induction of detoxification mechanisms (CYP1A induction) have been studied in parallel
(Chap. III.4).
The existing relationships between PCB levels measured in the environment and those
measured in sea stars were also examined (Section IV) in order to «calibrate» the

36
OBJECTIVES
bioindicating value of A. rubens. The health status of sea stars was studied alongside (same
Section) by measuring different biological responses (ROS production and CYP1A induction)
in field-collected individuals. Finally, the last objective of this fourth section was to assess the
contamination status of the southern North Sea by measuring concentration levels of PCBs
and of other contaminants of concern, viz. heavy metals and dioxins, in sediments and biota.

III. EXPERIMENTAL CONDITIONS
37

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
III.1 Non-coplanar vs. coplanar congener-specificity of PCB bioaccumulation
and immunotoxicity in sea stars
Danis B 1 , Cattini Ch 2 , Cotret O 2 , Teyssié JL 2 , Coteur G 1 , Villeneuve JP 2 ,
Fowler SW 2 & Warnau M 2
1 Laboratoire de Biologie marine (CP 160/15), Université Libre de Bruxelles, Av. F.D.
Roosevelt 50, B-1050 Brussels, Belgium
2 International Atomic Energy Agency - Marine Environment Laboratory, 4 Quai Antoine Ier,
MC-98000 Monaco
39

ABSTRACT
Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
The group of polychlorinated biphenyls compounds (PCBs) are among the contaminants of
greatest concern in the marine environment. Resistant to degradation, bioaccumulated and
highly toxic, they represent a continual threat for marine organisms and ecosystems. The
toxicity of PCBs is now thought to be mainly attributable to a few congeners only, viz. the
coplanar congeners. In the present study, a representative species of North Sea benthic
ecosystems, the sea star Asterias rubens (L.), was exposed experimentally to a mixture of 10
selected PCB congeners (3 coplanar and 7 non-coplanar) in contaminated sediments. Both the
degree of bioaccumulation and subsequent immunotoxic effects of these PCBs were
determined. A strong congener-specificity for both bioaccumulation and immunotoxicity was
found as well as a probable induction of a congener-specific detoxification mechanism
resulting in the specific metabolization of the three coplanar congeners tested. Moreover, a
strong correlation was observed between the bioaccumulation of coplanar PCBs and
immunotoxic effects. These findings suggest that coplanar congeners need to be included in
the list of congeners recommended to be systematically analyzed for monitoring the water
quality of the marine environment. If similar relationships were found to occur in other taxa,
there would be strong grounds for a possible revision of the current international
recommendations on water quality criteria for these compounds.
KEYWORDS
Polychlorinated biphenyls; echinoderms; bioaccumulation; immunotoxicity; detoxification;
congener-specificity.
40

INTRODUCTION
Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
Owing to their persistence in the environment and the fact that they are readily
bioaccumulated and highly toxic for aquatic organisms, polychlorinated biphenyls (PCBs) are
among the pollutants of greatest concern in marine ecosystems (Stebbing et al. 1992, OSPAR
2000). Although PCB levels tend to decrease regionally, a global decline in PCB
concentrations is thought to be unlikely because of continual inputs into the environment
mainly due to leakages from landfills and emissions from incinerators (Tanabe 1988).
Another worrisome fact is that the quantities of PCBs still in use in the late 1980s exceeded
the amount that had been released into the environment until that time (ibid.). In the marine
environment, these contaminants mainly become associated with bottom sediments,
representing a chronic threat for benthic ecosystems (e.g., Fowler et al. 1978, Boese et al.
1996, Koponen et al. 1998).
The PCB family is composed of 209 different congeners and about half of them have been
detected in significant concentrations in environmental samples (Metcalfe 1994). Until the
late 1970s, PCBs were mainly analyzed as commercial mixture equivalents (e.g. Aroclor,
Kaneclor) (Duinker et al. 1991, Eganhouse & Gossett 1991). By the late 1980s, the need to
conduct congener-specific approaches became widely adopted, since it appeared that
processes such as bioaccumulation, metabolization, or toxicity could differ considerably from
one PCB congener to another (Duinker et al. 1989, Skerfving et al. 1994).
The International Council for the Exploration of the Sea (ICES) has recommended the
systematic consideration of 6 PCB congeners for monitoring purposes (i.e. IUPAC#28, 52,
101, 138, 153 and 180). This selection (with congener #118 added) was subsequently adopted
and recommended by the World Health Organization (WHO 1999). However, it is now
widely accepted that the toxicity of such mixtures is mainly due to a few congeners, viz. the
non-ortho-substituted and mono-ortho-substituted congeners (Duinker et al. 1989, Safe 1990,
Metcalfe 1994). These congeners appear to be more problematic than other PCBs; indeed, the
former can display planar configuration (c-PCBs) which is very close to the configuration of
the highly toxic 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD). In vertebrates, the
toxicity of coplanar PCBs is known to take place through a receptor-mediated response
involving the binding of the contaminant to the cytosolic aryl hydrocarbon receptor (AhR),
followed by changes in gene expression (Kohn 1983, Shugart et al. 1992, Safe 1995, Hahn
1998, Nebert et al. 2000). Furthermore, coplanar PCBs are known to induce CYP1A which is
the major enzyme responsible for the metabolic activation of promutagens and
procarcinogens. However, not all the adverse biological effects of PCBs are attributable to
41

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
this mechanism; for example, many PCBs can exert toxicity as endocrine disrupters or
immunosuppressants through diverse pathways such as alteration of enzymatic activities
(kinases and phospholipases), disturbance of Ca 2+ homeostasis, or modulation of gene
expression (e.g. Satar 2000, Arukwe 2001).
Echinoderms are a major phylum of marine benthic invertebrates which includes a number of
species playing key roles in various ecosystems (Harrold & Pearse 1987, Birkeland 1989,
Menge et al. 1994). Since they are commonly found in coastal and estuarine waters,
echinoderms are directly exposed to anthropogenic contaminants which may affect several
aspects of their physiology such as reproduction, early development, somatic growth, and
neurophysiology (den Besten et al. 1989, Mallefet et al. 1994, Kobayashi 1995, Warnau et al.
1996, Temara et al. 1997). Several authors have proposed the sea star Asterias rubens as a
valuable sentinel organism (sensu Philips, 1976, 1990) for surveying and monitoring marine
contamination (e.g. Bjerregaard 1988, Temara et al. 1998b, 2002, OSPAR 2000, Coteur et al.
2003a). This top-predator feeds mainly on filter-feeding bivalves and is widely distributed
both geographically and bathymetrically in the NE Atlantic and North Sea (Hayward &
Ryland 1990). Furthermore, it is easily identified, collected, and maintained under laboratory
conditions and several experimental and/or field studies have shown that A. rubens efficiently
bioconcentrates marine contaminants such as metals and PCBs (Binyon 1978, Bjerregaard
1988, den Besten et al. 1990a, 2001, Sørensen & Bjerregaard 1991, Rouleau et al. 1993,
Hansen & Bjerregaard 1995, Temara et al. 1996, 1997a, 1998b, Everaarts et al. 1998, Warnau
et al. 1999, Coteur et al. 2003a,b, Danis et al. Chap. III.2). Therefore, the common sea star has
been considered as an excellent bioindicator species for monitoring a range of contaminants
in the North Sea (Everaarts et al. 1998, den Besten et al. 2001, Coteur et al. 2003a,
Stronkhorst et al. 2003); however, few studies have investigated in detail PCB
bioaccumulation processes in sea stars.
The effects of contaminants on A. rubens have focused mainly on its reproductive success
(den Besten et al. 1989), early development (Kobayashi 1995, Coteur et al. 2003a), immune
system (Coteur et al. 2003a,c, Danis et al. Chap. III.3, III.4), DNA integrity (Sarkar &
Everaarts 1995) and induction of the Cytochrome P450 (Everaarts et al. 1998, den Besten et
al. 2001, Stronkhorst et al. 2003, Danis et al. Chap. III.3, III.4). ROS production is one of the
main immune responses of echinoderms (Chia & Xing 1996). This response is triggered by
amoebocytes, which are the most active free-circulating cells found in echinoderm coelomic
cavities. Echinoderms lack a specific adaptive immune system and rely on innate responses
involving both humoral and cellular components (Chia & Xing 1996). These processes seem
42

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
to be very efficient and certainly constitute the main defence against foreign agents. Reactive
oxygen species (ROS) production has recently received much attention in the field of
invertebrate immunology and appears to respond to various xenobiotics (Anderson et al.
1997, Coteur et al. 2001, 2002c, 2003a, Danis et al. Chap. III.3, III.4).
The present study focuses on the bioaccumulation and immunotoxicity of key congeners (viz.
those considered as the most abundant or toxic ones) in the sea star A. rubens in order to
further assess its value as a bioindicator species. Particular care was taken to design
experimental conditions to simulate, as closely as possible, natural conditions that may be
encountered in the marine environment (e.g. levels of contaminants and complex exposure).
MATERIALS AND METHODS
Test organisms and sediments
Sea stars Asterias rubens L. and sediments (5 cm upper layer) were collected in Audresselles
(Nord-Pas-de-Calais, France) in March 1999 and transported to the IAEA-MEL premises in
Monaco. Prior to experimentation, animals were acclimated to laboratory conditions for 1
month (constantly aerated open-circuit 3000 l aquarium; flow rate 300 l hr -1 ; salinity 34 p.s.u.;
temperature 17 ± 0.5°C; light/dark cycle 12 hrs/12 hrs) during which time they were fed
mussels (Mytilus edulis L.) ad libitum. The sediments were conditioned in a constantly
aerated closed-circuit 400 l aquarium (salinity 34 p.s.u.; temperature 4 ± 0.5°C; light/dark
cycle 12 hrs/12 hrs), and before utilization they were characterized for total and organic
carbon content and grain-size distribution (Table 3).
Table 3. Sediments characteristics: total and organic carbon content (mg C g -1 ), and grain-size distribution (%)
PCB congeners and sediment preparation
Total carbon 9.29 ± 2.75 mg g -1
Organic carbon 0.27 ± 0.09 mg g -1
Size fractions % (mean ± SD)
500-1000 µm 0.04 ± 0.03
250-500 µm 1.97 ± 0.22
125-250 µm 96.2 ± 0.33
63-125 µm 1.70 ± 0.19
< 63 µm 0.08 ± 0.01
Ten PCB congeners were purchased from Promochem GmbH (Germany) as single congeners
of high certified purity (from 99.1 to 99.9% purity according to congener). For the
43

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
experiments, a stock solution of each congener was prepared in acetone (Ultrapure grade,
Sigma).
Three kg of dry sediments were spiked according to the method of Murdoch et al. (1997).
Sediments were placed in a 5 l glass bottle containing 500 ml of decanted natural seawater.
An aliquot (~1 µl) of each PCB stock solution was then added using a 5 µl Hamilton glass
syringe. After 5 min (to allow acetone to evaporate), sediments were agitated using a rotating
plate (150 rpm) for 35 hrs. At the end of that period, the supernatant seawater was removed
and sediments were placed in a 70 l glass aquarium, thoroughly mixed, and maintained for 24
hrs under flowing seawater (flow rate 15 l hr -1 , salinity 34 p.s.u.; temperature 17 ± 0.5°C;
light/dark cycle 12 hrs/12 hrs) to allow leaching of weakly-bound PCBs. The quantity of
spiked sediments allowed to form a continuous layer of 2 cm depth in the experimental
aquarium.
Exposure of sea stars to PCBs
Fifty sea stars of similar size (radius: 59 ± 8 mm) were held for 28 d in the aquarium
containing the spiked sediments under the same open-circuit conditions as previously
described. Every fourth day the sea stars were allowed to feed overnight on fresh mussels M.
edulis; the mussels (one per sea star) were supplied in the evening, and the next morning any
uningested food and empty shells were removed. Duration and frequency of feeding were
designed in this way in order to minimize any ingestion of PCB-labelled food. At the
beginning and end of the experiment, sediments were sampled to check the PCB
concentrations. Periodically throughout the experiment, 3 to 4 sea stars were collected, rinsed
in fresh seawater (particular care was taken to eliminate sediment particles), rapidly drained
of excess water, sampled for their coelomic fluid (see below), and then dissected. Two body
compartments (bodywall and pyloric caeca) were isolated, weighed (wet weight) and kept
frozen (-80°C) for subsequent analysis of PCB congener concentrations.
PCB analyses
Extraction of freeze-dried samples: A 50 to 100 g fresh weight sub-sample was taken from
the frozen sample. This sub-sample was freeze-dried and carefully pulverised in a cleaned
pestle and mortar. About 5 to 10 g of this material was placed into a pre-cleaned extraction
thimble in a Soxhlet apparatus. The size of the sub-sample was adjusted in order to obtain
about 100 mg of extractable organic matter (“lipid”). A known amount of internal standard
(2,4,5 trichlorobiphenyl, PCB 29) was added to the sub-sample in the thimble before Soxhlet
extraction. About 250 ml of hexane were added to the extraction flask and the sample was
extracted for 8 hours cycling the solvent through at a rate of 4 to 5 cycles per hour.
44

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
Concentration of the extract: The extracts were concentrated on a rotary evaporator (30 °C) to
obtain a final volume of about 15 ml, were dried with anhydrous sodium sulphate and further
concentrated to 1 ml by evaporating excess solvent under a gentle stream of clean dry
nitrogen. The sample extract was analysed gravimetrically for extractable organic matter
(EOM) content.
Removal of lipids by concentrated sulphuric acid: The concentrated extract was transferred
into a separatory funnel with enough hexane (40 to 50 ml) in order to dilute the sample. Five
ml concentrated sulphuric acid were added to the extract and the separatory funnel was
vigorously shaken. The separatory funnel was placed in a rack and the phases were allowed to
separate. The hexane phase was recovered into a glass beaker. Dried with sodium sulphate
and the hexane was transferred into a graduated tube. The volume of the extract was reduced
by evaporating the solvent with a gentle stream of pure nitrogen to about 1 ml.
Fractioning: A Florisil column was used for this separation. The Florisil was pre-extracted in
the Soxhlet apparatus with methanol for 8 hours, then with hexane for another 8 hours.
Activation was achieved by heating the dried Florisil at 130 °C for 8 hours. It was then
partially deactivated with 0.5 % water by weight and stored in a tightly sealed glass jar with
ground glass stopper. The water was well mixed into the Florisil and the mixture was allowed
to equilibrate for one day before use. A 1 cm diameter burette with Teflon stopcock was
plugged with pre-cleaned glass wool. 18 grams of Florisil were weighed out in a beaker and
covered with hexane. A slurry was made by agitation and poured into the glass column. The
solvent was drained to just above the Florisil bed. Individual columns were prepared
immediately before use, and a new column of Florisil used for each sample.
The extract, reduced to 1 ml, was applied to the Florisil column. It was carefully eluted with
65 ml of hexane and the fraction, containing the PCBs, collected and concentrated to 1 ml by
evaporating the hexane with a gentle stream of pure nitrogen.
Gas chromatography conditions: Samples were analyzed using an HP 6890 gas
chromatograph equipped with an ECD- 63 Ni detector and an SE-54 capillary column (30 m
length, 0.25 mm id, 0.25 µm film thickness). Initial temperature of the column was 70°C.
Final temperature was 260°C, reached at a rate of 3 °C min -1 . The detector temperature was
300°C. The carrier gas was helium, at a flow rate of 1.0 ml min -1 . Injection mode was
“splitless”.
Quality assurance/quality control: QA/QC was performed according to the guidelines on the
QA/QC requirements for analysis of sediments and marine organisms as described in
45

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
Reference Method N°57, “Contaminant monitoring programs using marine organisms:
Quality assurance and good laboratory practice” (UNEP 1990).
Biokinetic analyses
PCB uptake in the two sea star body compartments (bodywall and pyloric caeca) was
expressed as change in concentration (ng PCB g -1 lipid) over time of either single congeners
or the sum of the 10 considered congeners. When uptake kinetics tended to reach a steady
state during the experiment, they were best fitted by a single-component first-order kinetic
model (Equation 1):
Equation 1: C t = C 0 + C ss (1 - e -kt ),
where C 0 and C t are concentrations (ng PCB g -1 lipid) at time 0 (background concentration)
and t (days), respectively, and C ss is the incorporated concentration (ng PCB g -1 lipid) at
steady state, and k is the depuration rate constant (day -1 ) (Brown et al. 1995). Constants of the
uptake equation and their statistics were estimated by iterative adjustment of the model and
Hessian matrix computation, respectively, using the nonlinear curve-fitting routines in the
Systat 5.2.1 software (Wilkinson 1988). Possible correlations between concentration values
for the 10 congeners in the different body compartments were calculated using bivariate
correlation in the SPSS 11 software.
ROS production measurements
At days 0, 2, 7, 16, and 28 during the contamination period, coelomic fluid was collected
from 4 sea stars for ROS production analysis according to the method described by Coteur et
al. (2002). An aliquot of 3 ml of coelomic fluid was obtained by cutting the tip of the longest
arm of the sea stars and poured into 3 ml anticoagulant buffer (1.2 10 -2 M EDTA in Ca-, Mg-
free artificial seawater -CMFASW-; Noble 1970) at 4°C. The coelomocyte concentration of
this suspension was determined using a Thoma haemocytometer. The suspension was then
centrifuged (400 g for 10 min, 4°C) and the supernatant replaced by CMFASW to obtain a
final concentration of 1 ± 0.25 x 10 6 cells ml -1 . This concentration was double-checked as
described above.
ROS were measured using peroxidase/luminol-enhanced chemiluminescence (PLCL) as
described in Coteur et al. (2002), using a EG&G LB-9507 luminometer. Measurements were
normalised with the actual coelomocyte concentration in each sample and expressed either as
the sum of all 10 min-interval measurements for 10 6 cells ("total chemiluminescence").
Possible correlations between ROS production and PCB concentrations in sea star tissues
were tested using bivariate correlation procedures in the SPSS 11 software.
46

RESULTS
PCB congener bioaccumulation
Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
The uptake kinetics of the 10 PCB congeners tested were determined in 2 different body
compartments (bodywall and pyloric caeca) and are shown in Fig.9 (sum of 10 PCBs) and
Figs 10 and 11 (individual congeners).
Figure 9. Uptake kinetics of the sum of 10 PCB congeners (mean concentration in
ng g –1 total lipids) in bodywall and pyloric caeca of A. rubens exposed to spiked
sediments
A.
47

A.
B.
B.
Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
Figure 10. Uptake kinetics of 7 non coplanar PCB congeners (mean concentration
in ng g –1 total lipids) in (A) bodywall and (B) pyloric caeca of A. rubens exposed
to spiked sediments
PCB concentration (ng g -1 lipids)
800
700
600
500
400
300
200
100
0
PCB concentration (ng g -1 lipids)
PCB 77 PCB 126 PCB 169
0 5 10 15 20 25 30
300
250
200
150
100
50
0
Time (d)
PCB 77 PCB 126 PCB 169
0 5 10 15 20 25 30
Time (d)
Figure 11. Individual uptake kinetics of 3 c-PCB congeners (mean concentration in ng g –1 total
lipids ± SD, n = 3) in (A) bodywall and (B) pyloric caeca of A. rubens exposed to spiked
sediments.
48

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
The monitoring of PCB congener concentrations in sediments after spiking and after 28 days
of exposure (Table 4) showed that S 10 PCB concentration decreased from 178 ng g -1 dry wt at
the beginning of the experiment to 129 ng g -1 dry wt at day 28. Concentrations decreased with
similar rates regardless the congener, indicating that no particular process other than
desorption from sediments was responsible for this loss. Assuming the concentration decrease
rate was constant this would result in a time-integrated mean sediment concentration of 154
ng g -1 dry wt.
Table 4. Concentration of different PCB congeners in the sediments (mean concentration; ng g -1 dry wt; n=3).
Before PCB addition After PCB addition and 24 hrs under
flowing seawater
49
After 28 days of experiment
PCB 28 0.044 5.9 4.2
PCB 52 0.004 7.1 5.1
PCB 77 0.003 5.2 3.5
PCB 101 0.008 26 18
PCB 118 0.005 27 19
PCB 126

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
Table 5. Parameters and statistics of the equations describing uptake kinetics of the non-coplanar PCB congeners
accumulated from sediments in two body compartments of Asterias rubens. Model used: C(t)=C 0+C ss.(1-e -k.t )
where C(t) and C 0 are PCB concentrations (ng g -1 lipids) respectively at time t (d) and at time 0 and C ss is the
PCB concentration incorporated at steady-state; k is the depuration rate constant (d -1 ); ASE is the asymptotic
standard error; C BKD is the background PCB concentrations, measured in sea stars before starting the
experiment; R 2 corr is the corrected determination coefficient.
Congener C BKD C 0(ASE) C ss(ASE) k (ASE) R 2 corr
Body wall PCB 28 8.33 - 367 (21.2) 0.33 (0.08) 0.69
PCB 52 31.1 - 593 (29.5) 0.22 (0.04) 0.83
PCB 101 200 191 (241) 3000 (426) 0.08 (0.03) 0.81
PCB 118 278 195 (287) 3530 (695) 0.06 (0.03) 0.76
PCB 138 989 629 (209) 4340 (1940) 0.03 (0.02) 0.82
PCB 153 1330 685 (261) 5580 (3750) 0.03 (0.03) 0.78
PCB 180 32.2 75.1 (134) 1520 (536) 0.05 (0.04) 0.72
∑ 10PCBs 2980 1510 (1300) 16600 (2910) 0.07 (0.03) 0.78
Pyloric caeca PCB 28 14.4 5.69 (68.3) 584 (89.1) 0.10 (0.05) 0.71
PCB 52 48.1 55.8 (84.2) 670 (116) 0.10 (0.05) 0.69
PCB 101 203 236 (183) 1470 (624) 0.06 (0.05) 0.58
PCB 118 214 204 (168) 1210 (498) 0.06 (0.06) 0.56
PCB 138 540 475 (163) 758 (243) 0.09 (0.09) 0.44
PCB 153 743 505 (211) 852 (221) 0.12 (0.11) 0.42
PCB 180 27.8 - 200 (31.5) 0.41 (0.32) 0.18
∑ 10PCBs 1890 1830 (941) 5230 (1390) 0.09 (0.08) 0.52
Table 6. PCB concentrations over time (d) (mean ± SD; ng g -1 lipids) in the bodywall (A) and pyloric caeca (B)
of the sea star.
A.Bodywall
Time PCB 28 PCB 52 PCB 77 PCB 101 PCB 118 PCB 126 PCB 138 PCB 153 PCB 169 PCB 180 ∑ 10PCBs
0 7.63 ± 1.21 25.5 ± 3.54 39.5 ± 3.54 112 ± 2.12 171 ± 1.41 30.7 ± 6.66 444 ± 38.4 559 ± 17.1 0.90 ± 0.12 28.0 ± 1.41 1600 ± 290
2 265 ± 66.8 292 ± 74.6 49.7 ± 18.2 987 ± 384 1020 ± 352 48.3 ± 10.7 1140 ± 246 1150 ± 322 0.85 ± 0.24 411 ± 170 5360 ± 1560
4 302 ± 84.5 411 ± 64.5 59.0 ± 33.0 1205 ± 377 1140 ± 359 41.3 ± 4.93 1230 ± 379 935 ± 240 0.76 ± 0.14 419 ± 222 5750 ± 1690
7 374 ± 72.9 501 ± 95.8 330 ± 83.3 1650 ± 453 1540 ± 397 473 ± 157 1740 ± 301 2060 ± 251 42.7 ± 24.8 578 ± 206 9280 ± 2000
11 398 ± 139 579 ± 167 298 ± 227 1950 ± 649 1820 ± 484 605 ± 154 1820 ± 436 2020 ± 517 41.2 ± 35.2 653 ± 301 9990 ± 3180
16 371 ± 61.6 531 ± 96.2 94.5 ± 51.3 2230 ± 714 2390 ± 744 51.5 ± 8.85 2570 ± 771 2820 ± 830 0.92 ± 0.11 765 ± 328 11800 ± 3470
21 423 ± 84.2 637 ± 148 151 ± 70.7 3320 ± 985 3240 ± 829 53.7 ± 8.50 3250 ± 1000 3790 ± 1240 0.74 ± 0.06 1630 ± 356 16700 ± 4800
28 318 ± 37.7 584 ± 50.3 57.5 ± 35.7 2905 ± 366 2960 ± 359 63.8 ± 7.59 3290 ± 484 3720 ± 613 0.91 ± 0.20 1180 ± 207 15100 ± 1690
B.Pyloric caeca
Time PCB 28 PCB 52 PCB 77 PCB 101 PCB 118 PCB 126 PCB 138 PCB 153 PCB 169 PCB 180 ∑ 10PCBs
0 12.7 ± 2.08 61.5 ± 14.9 42.7 ± 7.02 215 ± 99.1 209 ± 65.8 25.3 ± 10.2 382 ± 77.2 522 ± 167 0.07 ± 0.02 16.3 ± 4.16 162 ± 442
2 221 ± 87.5 294 ± 113 116 ± 32.35 624 ± 209 560 ± 179 59.0 ± 48.1 759 ± 184 913 ± 254 0.07 ± 0.02 177 ± 85.6 3700 ± 1020
4 214 ± 35.4 334 ± 118 160 ± 33.94 569 ± 97.6 532 ± 88.4 119 ± 39.6 756 ± 88.4 910 ± 63.6 5.71 ± 9.78 139 ± 89.1 3730 ± 350
7 340 ± 177 411 ± 208 172 ± 99.86 855 ± 504 776 ± 424 154 ± 148 983 ± 394 1170 ± 366 0.08 ± 0.01 218 ± 168 5080 ± 2380
50

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
11 362 ± 201 477 ± 63.7 147 ± 80.35 856 ± 314 707 ± 301 69.0 ± 78.4 788 ± 296 890 ± 311 0.44 ± 0.66 184 ± 149 4480 ± 1660
16 429 ± 26.7 499 ± 35.0 65.0 ± 14.00 1016 ± 126 887 ± 71.9 36.7 ± 3.21 1080 ± 55.0 1250 ± 80.8 0.06 ± 0.01 139 ± 31.4 5400 ± 351
21 595 ± 174 689 ± 235 72.3 ± 37.29 1360 ± 527 1110 ± 415 33.3 ± 12.7 1060 ± 259 1300 ± 398 0.06 ± 0.01 218 ± 175 6440 ± 2130
28 539 ± 141 688 ± 241 104 ± 62.34 1410 ± 630 1230 ± 570 42.5 ± 16.8 1220 ± 493 1370 ± 519 0.05 ± 0.01 227 ± 183 6830 ± 2810
When individual PCB congeners were considered, uptake in bodywall and pyloric caeca
displayed saturation kinetics for all congeners except the coplanar ones (Fig.10, Tables 5 and
6). In the case of coplanar congeners, kinetics showed a fairly unpredictable bioaccumulative
behaviour (Fig.11). Indeed, their concentration increased at the beginning of the experiment
(until day 7 to 11), and then dropped sharply. In bodywall, the estimated steady-state
concentrations (C 0 + C ss) of non-coplanar congeners ranged from 367 ng g -1 lipids (PCB 28) to
6.3 µg g -1 lipids (PCB 153) (ratio=1:17) whereas in the pyloric caeca, values were lower,
ranging from 590 ng g -1 lipids (PCB 28) to 1.7 µg g -1 lipids (PCB 101) (ratio=1:3).
In order to verify a possible metabolic transformation of the c-PCBs, ratios of the non
coplanar PCB 153 (which is not metabolized) to PCB 77, PCB 126 and PCB 169 have been
calculated for sediments and sea star tissues at different times during the experiment (Fig. 12).
The ratio did not vary much for sediments, indicating that no significant metabolization of
these congeners occurred. In contrast, the ratios of PCB 153 to c-PCBs in sea star tissues
increased up to a factor 30,000 (Fig.12).
Figure 12. Variation of the ratio between PCB 153 and c-PCBs 77, 126 and 169 in sediments (yellow) and sea
stars bodywall (orange) and pyloric caeca (green) at different times during the exposure period. Time 0:
background ratios (before spiking). To fit the figure, PCB 153:169 ratio is divided by a factor 100 in sea stars
bodywall, and by 1000 in pyloric caeca.
51

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
Correlations between incorporated concentrations of different congeners were also calculated.
Bioaccumulation of several congeners was found to be highly correlated in both tissues. The
strongest correlations were found for PCBs 28 vs. 52 (r ≥ 0.91), 101 vs. 118 (r ≥ 0.99), and
138 vs. 153 (r ≥ 0.98). Significant correlations were also found between c-PCB concentrations
in bodywall and pyloric caeca, both for intra- and inter-tissue comparisons. In particular, PCB
77 and PCB 126 concentrations were always closely correlated, sometimes with very high
correlation coefficient values (e.g., r = 0.95, p ≤ 0.0001 in bodywall).
Toxic Equivalents
Dioxin toxic equivalents (TEQs) were calculated for c-PCBs using toxic equivalency factors
(TEFs) as described by Smith & Gangoli (2002) (Fig.13). Toxicity was always dominated by
c-PCB 126 the contribution of which accounted for more than 98% of the total TEQ in sea
star tissues. The highest TEQ values calculated for c-PCB 126 were 60.5 ± 15.4 µg TEQ g -1
lipids in the bodywall (at day 11) and 15.4 ± 14.8 µg TEQ g -1 lipids in the pyloric caeca (at
day 7).
A.
B.
PCB concentration (pg TEQ g -1 lipids)
PCB concentration (pg TEQ g -1 lipids)
45
40
35
30
25
20
15
10
5
0
70000
60000
50000
40000
30000
20000
10000
0
Bodywall Pyloric caeca Sum LCL S
0 2 7 16 28
Time (d)
Bodywall Pyloric caeca Sum LCL S
0 2 7 16 28
Time (d)
52
90000
80000
70000
60000
50000
40000
30000
20000
10000
0
90000
80000
70000
60000
50000
40000
30000
20000
10000
0
Total Chemiluminescence (RLU)
Total Chemiluminescence (RLU)

C.
PCB concentration (100 pg TEQ g -1 lipids)
800
700
600
500
400
300
200
100
0
Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
Bodywall Pyloric caeca Sum LCL S
0 2 7 16 28
Time (d)
Figure 13. Comparison between c-PCB 77 (A), 126 (B) and 169 (C) concentrations
(pg WHO TEQ g -1 lipids; except PCB 169 in pyloric caeca: 100 pg WHO TEQ g -1
total lipids) in bodywall (white bars) and pyloric caeca (grey bars), and ROS
production by bacteria-stimulated amoebocytes (white dots; total
chemiluminescence, RLU) of A. rubens exposed to spiked sediments.
Effects of PCB congeners on ROS production
In order to gain some insight into the effects of PCB congeners in this species, reactive
oxygen species (ROS) production was analysed for nonstimulated and bacteria-stimulated
coelomocytes.
During the experiment, the variation of ROS production displayed a distinct behaviour, quite
comparable to that observed for c-PCBs uptake kinetics (Fig. 13). Indeed, ROS production in
both nonstimulated and bacteria-stimulated coelomocytes increased sharply during the first
week of exposure, then declined and reached a constant level, close to background levels,
during the two last weeks (Fig. 14).
A. Bodywall
PCB concentration (ng g -1 lipids)
500
450
400
350
300
250
200
150
100
50
0
PCB 77 PCB 126 PCB 169 Sum LCL NS Sum LCL S
0 5 10 15 20 25 30
Time (d)
53
90000
80000
70000
60000
50000
40000
30000
20000
10000
0
Total Chemiluminescence (RLU)
100000
90000
80000
70000
60000
50000
40000
30000
20000
10000
0
ROS production (sum LCL)

B. Pyloric caeca
PCB concentration (ng g -1 lipids; pg PCB169 g -1
lipids)
200
180
160
140
120
100
80
60
40
20
Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
PCB 77 PCB 126 PCB 169 Sum LCL NS Sum LCL S
0
0
0 5 10 15 20 25 30
Time (d)
Figure 14. Comparison between ROS production by nonstimulated (black dots) and
bacteria-stimulated (white dots) amoebocytes (total chemiluminescence, RLU) and
c-PCB 77 (white squares) 126 (light grey squares) and 169 (dark grey squares)
concentrations (ng g -1 total lipids; except PCB 169 in pyloric caeca: pg g -1 total
lipids) in (A) bodywall and (B) pyloric caeca of A. rubens exposed to spiked
sediments.
Most notably, the kinetic behaviour of ROS production matched perfectly that displayed by c-
PCB bioaccumulation in sea star tissues (see Figs 13 and 14, respectively). Therefore,
correlations were computed between ROS production and c-PCB tissue content expressed
either as TEQ or concentrations. Except for PCB 77 and PCB 169 in pyloric caeca, c-PCB
concentrations in the two body compartments were significantly correlated to ROS production
by nonstimulated and bacteria-stimulated coelomocytes, with correlation coefficients ranging
from 0.71 to 0.84.
DISCUSSION
The present study is a complement to two previous studies dealing with single PCB congener
accumulation in sea stars, in the form of two structurally-contrasting radiolabelled PCB
congeners, PCB 153 and c-PCB 77 (Danis et al. Chap. III.2, III.3, respectively). These
previous experiments, along with those performed in the present study, were designed to
obtain complementary information on PCB bioaccumulation in the sea star Asterias rubens
using different techniques (GC-ECD vs. radiotracer techniques), uptake routes (seawater,
food, sediments), and contaminants (10 different congeners), but always using contaminant
exposure concentrations that were on the same order of magnitude as those that can occur in
the marine environment.
54
100000
90000
80000
70000
60000
50000
40000
30000
20000
10000
ROS production (sum LCL)

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
Results of the present study clearly demonstrated that all PCB congeners tested were
accumulated efficiently in sea star tissues, with transfer factors from sediments of 120 in the
bodywall and 40 in the pyloric caeca. These values, although slightly lower, are comparable
to those found for sediment exposure in previous studies using the single congener PCB 153
(Danis et al. Chap. III.2); this suggests that the presence of a mixture of congeners does not
interfere with uptake efficiency of PCB 153 in sea star tissues.
When considering the sum of the 10 PCB congeners tested, uptake kinetics in bodywall and
pyloric caeca were best described by continuous increasing functions eventually reaching a
steady state. Between the two tissues, bodywall was found to be the most effective
accumulator which is in agreement with previous observations (Danis et al. Chap. III.2).
Because it is an efficient accumulator, easily dissected, and constitutes ca. 75% of the total
sea star body weight, bodywall could serve as an ideal target tissue for biomonitoring studies
using sea stars as bioindicators. Bodywall analysis of PCBs would complement the
information gathered through analysis of pyloric caeca which are often the only organs
examined in field studies (e.g. Everaarts et al. 1998, den Besten et al. 2001, Stronkhorst et al.
2003).
When examining individual congeners separately, all the non-coplanar PCBs followed the
same saturation kinetics as S 10 PCBs, although they attained different steady-state
concentration values. Surprisingly, the uptake kinetics for coplanar congeners, displayed
totally different patterns; for example, the concentrations of these PCBs reached a peak value
after 7 to 11 days of exposure, according to the c-congener, and then suddenly dropped to
initial values. This behaviour could be due either to a change in bioaccumulation parameters
(decrease in uptake rate or increase in loss rate) or, alternatively, to the progressive activation
of an efficient detoxification mechanism that triggers metabolization of the c-PCB
specifically. In view of existing literature, the second hypothesis appears to be the most
plausible one. Indeed, specific induction of P450 enzymatic activity has been reported in
echinoderms exposed to c-PCBs (e.g., den Besten et al. 1993; Danis et al. Chap. III.3, III.4).
In addition, the fact that ratios of the non-ortho substituted congeners (PCBs 77, 126 and 169)
to the PCB 153 increased during the experiment support the existence of a metabolic
transformation of these c-PCB congeners.
c-PCBs have close structural similarities with polychlorodibenzo-p-dioxins (PCDDs)
(Metcalfe et al 1994, Schweitzer et al. 1997, Walker & Peterson 1994). Those c-PCBs with
vicinal hydrogen atoms in o, m positions and 0-1 ortho chlorine atoms, such as PCBs 77 and
55

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
126, are generally thought to be targets of Cytochrome P450 (CYP) isozymes and
consequently metabolized (Hong et al. 1998). Regarding the effects of PCB 77, caution
should be exercised, as existing evaluations of the toxicity (and metabolization) are hampered
by variations in toxic effects and in different species (Safe 1994). Whereas the CYP enzyme
system and its inducibility have been extensively described in vertebrates, much less is known
for invertebrates despite the fact that the CYP1A system has been reported in four
invertebrate phyla: annelids, arthropods, echinoderms and molluscs (Lee 1981, Bucheli &
Fent 1995). In particular, the first echinoderm CYP genes were recently identified by Snyder
(1998) in digestive tissues of a sea urchin (Lytechinus anamesis), and evidence for the
presence of P450 enzymes belonging to the CYP1, CYP2, and CYP3 subfamilies has been
reported previously in sea stars (den Besten et al. 1993, Danis et al. Chap. III.3, III.4).
If, as strongly suggested by these observations, a metabolization of the c-PCBs actually
occurs, our results indicate that PCBs 77, 126 and 169 undergo similar metabolic processes in
sea stars, and that CYP enzymes play a pivotal role in these processes.
Most noteworthy, the immune biomarker (ROS production) that was measured during the
bioaccumulation experiment followed exactly the same pattern as c-PCB uptake; i.e., after an
initial increase to day 7, ROS production fell to control levels from day 16 until the end of the
experiment. This observation could be due to toxicity and subsequent impairment of
amoebocytes occurring when PCB concentrations reach a certain level within the organism’s
tissues. However, non-coplanar PCB congeners (viz., the ones showing increasing
concentrations over time) were shown to have no effect on ROS production in echinoderms
(Coteur et al. 2001, Danis et al. Chap. III.4). In contrast, c-PCBs are well-documented to
specifically stimulate ROS production by amoebocytes in echinoderms (Coteur et al. 2001,
Danis et al. Chap. III.3, III.4) as well as in other marine organisms (e.g., Wilbrink et al. 1991,
Duffy et al. 2002). Therefore, the return of ROS production to normal levels is clearly related
to the decrease in c-PCB concentrations in the sea stars. This is also suggested by the strong
correlation found between c-PCB concentrations and ROS production values. Such a drop in
ROS production was also reported in A. rubens during experimental exposure to the c-PCB 77
congener (Danis et al. Chap. III.3). The involvement of a congener-specific protective
mechanism through the cytochrome P450 system was also suggested, since a clear induction
of a CYP1A immunopositive protein (CYP1A IPP) was measured following c-PCB 77 and
126 exposure, but not following PCB 153 exposure (Danis et al. Chap. III.3, III.4).
56

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
The immunotoxic effects measured in the present study show the potential importance of
ROS production as a biomarker of c-PCB exposure, and constitute a good basis to argue for
the usefulness of echinoderms as bioindicator organisms for PCBs. In addition, this immune
biomarker is of high ecological relevance, since it furnishes direct information about the
health of the organisms. Indeed, high levels of ROS production can lead to the production of
large quantities of reduction products, such as the hydroxyl radical (OH • ) or the superoxide
anion ( • -
O2 ) which are extremely potent oxidants that react with cellular macromolecules (e.g.,
Babior et al. 1984, Winston & Di Giulio 1991, Chia and Xing 1996, Baier-Anderson &
Anderson 2000).
It is interesting to note that while the shape of the ROS production kinetics closely reflected
the observations of Danis et al. (Chap. III.3), the pattern of the PCB 77 uptake kinetics
observed here did not. In the present study PCB 77 was bioaccumulated in A. rubens tissues
following bell-shaped kinetics, whereas Danis et al. (Chap. III.3) described a continuous
increasing uptake finally reaching a steady state. In view of the above discussion, this kinetic
discrepancy is most certainly due to the different analytical methods used in both studies. The
present study used classical chemical analysis (GC-ECD) which unequivocally identifies each
congener, whereas Danis et al. (Chap. III.3) used liquid scintillation techniques to measure the
14 C activity of radiolabelled PCB 77. Although the use of a radiotracer has numerous
advantages in studying c-PCB biokinetics (e.g. increase in detection sensitivity), the major
drawback of this technique is that it detects only 14 C and not specifically the PCB congeners.
Hence, no differentiation can be made between the congener itself and possible degradation
products if the labelled molecule is metabolized. Therefore, based on the observations noted
in our study as well as those reported by Danis et al. (Chap. III.3, III.4) on ROS production
and CYP1A IPP induction, we suspect that the saturation kinetics reported for radiolabelled
PCB 77 bioaccumulation result from detection of both 14 C-PCB 77 and its 14 C-metabolite(s)
rather than being due to a different kinetic behaviour.
CONCLUSIONS
In general, the findings from this study demonstrate a strong relationships between the
bioaccumulation of coplanar PCB and resultant immunotoxic effects. Furthermore, these
results have provided evidence for congener-specific toxicity under complex, realistic
conditions (i.e. exposure to a mixture of PCB congeners bound to sediments containing
ecologically-relevant concentrations). These findings underscore the need to provide
57

Bioaccumulation and effects of a mixture of PCB congeners in the sea star Asterias rubens
information in natural environments about similar coplanar-specific biological effects in other
organisms, especially the “classical” bioindicator species such as bivalves. The
recommendations for PCB monitoring that are presently adopted by international
organisations (e.g. EU, ICES) and widely followed in the scientific community, generally
address a limited set of PCB congeners (viz. #28, 52, 101, 138, 153 and 180). The latter ones
are well-known to be the most relevant in terms of PCB abundance in biota and environment
(e.g. OSPAR 2000), but not at all with respect to their toxicity and as a threat to marine
ecosystems. Our results clearly show that coplanar congeners were the only ones responsible
for the observed immunotoxicity, despite the fact that they were much less abundant in the
experimental environment (as occurs in the natural environment) and far less bioconcentrated
than the congeners “recommended” for monitoring studies. Because PCBs affect the immune
system of sea stars in a congener-specific way, coplanar PCBs may represent a potential
threat to echinoderm populations, and hence to benthic communities in general. Therefore, we
propose that coplanar congeners be included on the list of congeners to be systematically
monitored in studies dealing with water quality assessment of the marine environment.
Furthermore, a generalization of our observations to other marine taxa would necessitate a
revision of the current international recommendation guidelines.
ACKNOWLEDGEMENTS
This work was performed at the IAEA-Marine Environment Laboratory which operates under
a bipartite agreement between the International Atomic Energy Agency and the Government
of the Principality of Monaco, and was partially supported by special funding from the
National Bank of Belgium (BNB), a NFSR Research fellowship to MW, and SSTC Contract
MN/11/30 of the Belgian Federal Research Programme. BD and GC were holders of a FRIA
doctoral grant, and MW is a Honorary Research Associate of the National Fund for Scientific
Research (NFSR, Belgium).
58

Delineation of PCB uptake pathways in the sea star Asterias rubens
III.2 Delineation of PCB uptake pathways in a benthic sea star using a
radiolabelled congener
Marine Ecology Progress Series 253:155-163
Danis B a , Cotret O b , Teyssié JL b , Fowler SW b , Bustamante P c & Warnau M b
a. Laboratoire de Biologie marine (CP 160/15), Université Libre de Bruxelles, Av. F.D.
Roosevelt 50, B-1050 Brussels, Belgium
b. International Atomic Energy Agency, Marine Environment Laboratory, 4 Quai Antoine Ier,
BP 800, MC-98012 Monaco cedex
c. Laboratoire de Biologie et d'Environnement Marins, UPRES-EA 3168, Université de La
Rochelle, 22 Avenue Michel Crépeau, F-17042 La Rochelle, France
59

ABSTRACT
Delineation of PCB uptake pathways in the sea star Asterias rubens
Asterias rubens, a common sea star in North Sea waters, was selected to study the
bioaccumulation of an important polychlorinated biphenyl congener, 14 C-labelled PCB 153,
from two contrasted sources: seawater and sediments. After 4 weeks of acclimation to
laboratory conditions, sea stars were exposed for 34 days to realistic concentrations (30 ng l -1
in seawater and 9.5 ng g -1 dry wt in sediments) of the contaminant during which time
bioaccumulation of PCB#153 was followed in 6 body compartments. Results showed that (1)
for each body compartment, PCB uptake kinetics were generally asymptotic and
bioaccumulation was far greater when A. rubens was exposed via seawater than via
sediments, (2) body wall and podia were the body compartments showing the greatest affinity
for the PCB congener making them ideal tissues for biomonitoring purposes, and (3) the
concentrations reached in body compartments were in the range of values reported in several
field studies. Because radioisotopic techniques are extremely sensitive, they allow taking into
account key organs which are sometimes too small for standard analysis of PCBs.
KEYWORDS
Polychlorinated biphenyls; PCB 153; bioaccumulation; kinetics; Asterias rubens; echinoderm
60

INTRODUCTION
Delineation of PCB uptake pathways in the sea star Asterias rubens
Polychlorinated biphenyls (PCBs) are strictly anthropogenic chemicals that constitute one of
the most problematic and widespread group of contaminants. These xenobiotics, represented
by 209 congeners, are extremely resistant to degradation (physico-chemical or biological), are
bioconcentrated by living organisms, and can cause various adverse effects depending on
their pattern and degree of chlorine substitution (Metcalfe 1994). For PCBs entering the
marine environment, bottom sediments are the ultimate repository where they may become a
source for uptake by marine organisms through direct or indirect contact or, for filter feeders,
by ingestion; however, information about their impact on benthic species is relatively scarce
(Chapman 1995, Carr et al. 1996, Wood et al. 1997).
According to different authors, the asteroid Asterias rubens qualifies as an excellent
bioindicator organism for monitoring heavy metal contamination in the North Sea and NE
Atlantic benthic ecosystems (Knickmeyer et al. 1992, den Besten et al. 1993, Everaarts et al.
1998, Temara et al. 1998b, Warnau et al. 1999). It is indeed a widely distributed and abundant
key species (sensu Lewis 1978) that is easy to collect, identify and maintain in the laboratory.
In addition, A. rubens is a top predator feeding mainly on mussels and living on or in the
proximity to bottom sediments which are the main reservoir of many contaminants, including
PCBs. The biological and ecological characteristics of A. rubens as well as its potential
economic impact (as a predator of commercially important mussels) have lead some authors
to use this species as a tool to assess the degree of PCB contamination in the North Sea (den
Besten et al. 1989, 1993, Everaarts et al.1998). However, to the best of our knowledge, no
study has investigated PCB bioaccumulation processes in A. rubens. The only two
experimental studies investigating PCB bioaccumulation in echinoderms concern sea urchins
exposed to contaminated sediments (Weisberg et al. 1996, Schweitzer et al. 2000), and only
Weisberg et al. (1996) examined the kinetic aspects of PCB uptake.
Such data are however needed to further assess the value of Asterias rubens as a bioindicator
of PCB contamination. Therefore, in the present study, we have investigated the kinetics of
PCB uptake in A. rubens exposed either to the contaminant in seawater or associated with
sediments, i.e. the two extreme pathways of contamination from the viewpoint of absolute
PCB concentrations. Indeed, the high hydrophobicity of PCBs result in a characteristic
partitioning with concentrations in seawater typically in the range of pg to ng l -1 while
sediment concentrations are in the range of µg to mg kg -1 (see Table 7).
61

Delineation of PCB uptake pathways in the sea star Asterias rubens
Table 7. Characteristics of the background and added concentrations of PCB 153. Background concentrations
were measured in sediments, seawater and sea stars (body wall and pyloric caeca) the day before starting the
experiment; added concentrations were measured in subsamples of sediments and seawater regularly collected in
the experimental microcosms throughout the experiment. Ranges of values of PCB 153 (unless specified)
reported for sediments and seawater in the field are given for comparison.
Compartment PCB concentration Location References
Background Sediments 0.017 ng g -1 dry wt (n = 6)
Sea water 0.026 ng l -1 (n = 6)
Pyloric caeca 522 ± 167 ng g -1 lipids (n = 6)
Body wall 559 ± 17 ng g -1 lipids (n = 6)
Added Sediments 9.49 ± 1.14 ng g -1 dry wt (n =
Sea water
(dissolved +
particulate)
12)
31.4 ± 15.6 ng l -1 (n = 36)
Field Values Sediments 22 - 4,060 ng g -1 dry wt
0.27 - 47 ng g -1 dry wt (sum7 PCB)
2.2 – 32 ng g -1 dry wt (sum7 PCB)
Sea water (dissolved) 0.1 - 67.2 pg l -1
Sea water
(dissolved +
particulate)
Sea water (extreme
hot spot)
0.8 – 8.7 ng l -1 (Aroclor 1260)
1.5 – 38.0 ng l -1 (Phenoclor
DP-5)
0.2 – 370 ng l -1 (Phenoclor
DP-5/DP-6)
0.34 – 4.93 ng l -1 (sum hexa-
CB)
dissolved: 1.8 ± 0.3 µg l -1
particulate: 14 ± 3.9 µg l -1
62
North Sea, German Bight
North Sea, Dutch coastal
zone
North Sea, Dutch coastal
zone
Stebbing et al 1992
Boon et al. 1985
Laane et al. 1999
Baltic Sea Shultz-Bull et al.
1995
Atlantic Harvey & Steinhauer
Mediterranean French 1976
coasts
Elder 1976
Med. and Atlantic Marchand et al. 1990
French coasts Telli-Karakoç et al.
Marmara Sea
2002
New Bedford Harbor,
USA
Bergen et al. 1996
The PCB congener IUPAC #153 (2,2’,4,4’,5,5’ hexachlorobiphenyl) was selected because it
is the most abundant in marine biota (Stebbing et al. 1992) and has been shown to be an
excellent indicator of total PCB contamination (Atuma et al. 1996).
MATERIALS AND METHODS
Sampling
The sea stars Asterias rubens (Linnaeus 1758) were collected in April 1999 in the intertidal
zone at Audresselles (Pas-de-Calais, France). Prior to experimentation, specimens were
acclimated to laboratory conditions for 1 month in constantly aerated closed circuit aquaria
(salinity: 36 ‰, T: 16 ± 0.5 °C, 12/12 h dark/light cycle).
In order to follow PCB 153 bioaccumulation under realistically simulated conditions, a 14 C-
labelled congener was used and measured using highly sensitive b spectrometry.

Radiotracer
Delineation of PCB uptake pathways in the sea star Asterias rubens
The 14 C-labelled 2,2',4,4',5,5' hexachlorobiphenyl (purity ≥ 95%) was purchased from Sigma
Chemicals, USA. Specific activity was 925 MBq mmol -1 . Stock solutions were prepared in
acetone at a concentration of 1 µg ml -1 .
Sample treatment and liquid scintillation counting
Water samples (2 ml) were directly transfered to 20 ml glass scintillation vials (Packard,
USA) and 10 ml of Ultima Gold XR ® (Packard Instruments) scintillation liquid were added.
Samples of sediment and sea star tissue (previously crushed) were placed in a vial containing
2 ml of Acetonitrile ® in an ultrasonic bath for 10 min. Acetonitrile ® was then collected and
replaced by another 2 ml of Acetonitrile ® and the ultrasonic operation was repeated a second
time. This treatment gave 4 ml of liquid phase (viz. the extraction) and a residue. The residue
was digested overnight at 70°C with 2 ml of Soluene ® , and 10 ml of Hionic Fluor ®
scintillation liquid were then added. The liquid phase (4 ml) was added to 16 ml of filtered
seawater and extracted twice using 2 ml of n-Hexane (Sigma, USA) under constant agitation.
The organic phase (4 ml) and the aqueous phase (20 ml) were treated separately. The entire
organic phase and 2 ml of the aqueous phase were each added separately to 10 ml of Ultima
Gold XR ® scintillation liquid.
14 C-radioactivity was then measured using a 1600 TR Liquid Scintillation Analyzer (Packard),
compared to standards of known activities, and corrected for quenching, background and
physical decay of the radiotracer. Counting times were adjusted to obtain counting rates with
relative propagated errors less than 5 %. PCB concentrations were expressed on a total lipid
content basis where lipids were determined according to the method of Barnes & Blackstock
(1973). A schematic diagram of the sample treatment is shown in Figure 15.
63

Delineation of PCB uptake pathways in the sea star Asterias rubens
Figure 15. Schematic representation of sample processing before ß-spectrometry analysis
Experimental procedures
Uptake from seawater
residue
20 ml aqueous
phase
(including
Acetonitrile)
Liquid scintillation counting
2x{2 ml Acetonitrile; ultrasonic bath 10’}
Asteroids (n = 24) were placed for 34 d in a 70 l glass aquarium (constantly aerated closed
circuit aquaria; salinity 36 ‰; 16 ± 0.5°C; 12/12 h dark/light cycle) containing natural
seawater spiked with 14 C-labelled PCB 153. One day prior to the experiments, four 5 l glass
beakers were filled with filtered seawater (36 ‰; 16 ± 0.5°C), spiked with the radiolabelled
PCB stock solution, and constantly stirred using an orbital agitation plate. Contaminated
water then was poured into the glass aquaria and uncontaminated seawater was added to
obtain a final volume of 70 l. Seawater and radiotracer were renewed every second day during
the entire experiment. Activity was checked before and after each renewal to assess the
stability of the labelled PCB concentration in seawater (Table 7). The sea stars were fed
unlabelled mussels (Mytilus edulis) every second day just before the seawater renewal. After
2 hours uningested mussels were removed to limit as much as possible PCB incorporation via
the food. Periodically (after 2, 4, 7, 11, 14, 21 and 34 days), sea stars (n = 3) were removed,
dissected into seven body compartments (oral and aboral body walls, pyloric caeca, gonads,
64
2 ml hexane (added twice)
16 ml sea water
4 ml Acetonitrile ®
2 ml 2 ml 4 ml
+ 10 ml
Hionic
Fluor ®
+ 10 ml Ultima
Gold XR ®
4 ml hexane

Delineation of PCB uptake pathways in the sea star Asterias rubens
rectal caeca, central digestive system, and podia), and radioanalyzed to determine uptake
kinetics and body distribution of the incorporated PCB.
Uptake from sediments
Sediments (2.5 kg dry wt) from the North Sea (Audresselles, Pas-de-Calais, France) were
contaminated for 4 days with the 14 C-labelled PCB using the rolling jar method (Murdoch et
al. 1997). Sea stars (n = 24) were placed in a 70 l glass aquarium (constantly aerated open
circuit aquarium; salinity 36 ‰; 16 ± 0.5°C; 12/12 h dark/light cycle) containing a 10 cm
layer of seawater running over a 2 cm layer of spiked sediments. A seperate group of 5 sea
stars were placed in the same aquaria, but in another compartment (not in contact with the
sediments), to serve as a control for possible cross-contamination through seawater. The sea
stars were fed every second day with mussels (M. edulis). Uningested food was removed after
2 hours. The radioactivity of the labelled PCB was measured weekly in the sediments to
check for possible leaching (Table 7). Periodically (after 2, 4, 7, 11, 14, 21, and 34 days), 3
individuals were removed, dissected as described above, and their tissues counted for
radioactivity.
Data analyses
Uptake of the PCB congener from seawater and sediments was expressed as change in PCB
concentration (ng g -1 total lipids) over time. Uptake kinetics were described either by using a
saturation exponential model (Equation 2), a single component exponential model (Equation
3), or a combined model (logistic and single component exponential) (Equation 4):
Equation 2: C(t) = Css (1-e -k.t )
Equation 3: C(t) = C 0 e k.t
Equation 4: Ct = Css (1-e -k.t ) / 1+e -k.(t-I)
where C(t), C 0, and Css are the PCB concentrations (ng g -1 total lipids), respectively, at time t
(d), at time 0 and at steady state, k is the rate constant (d -1 ), and I is the time (d) at the
inflexion point. The model showing the best fitting accuracy (based on the calculation of the
determination coefficient, R 2 , and examination of the residuals) was used.
Constants of the different models and their statistics were estimated by iterative adjustment of
the models and Hessian matrix computation, respectively, using the nonlinear curve-fitting
routines in the Systat ® 5.2.1 software (Wilkinson 1988). Differences between PCB
concentrations in the different sea star body compartments were tested by 1-way ANOVA and
65

Delineation of PCB uptake pathways in the sea star Asterias rubens
the multiple comparison test of Tukey (Zar 1996). Changes in PCB body distribution were
tested for significance using the G-test (adapted from the log-likelihood ratio test) for 2xk
contingency tables (Zar 1996). Prior to the latter test, data were arcsin-transformed using the
correction of Freeman-Tukey (1950) described by Zar (1996). The level of significance for
statistical tests was always set at a = 0.05.
RESULTS
The uptake of PCB#153 by Asterias rubens was investigated through separate exposures to
contaminated seawater or sediments. As differences between accumulation kinetics in aboral
and oral body walls were never found in any experiment (p always > 0.1), these two
compartments were pooled and are presented as a single compartment (body wall) throughout
the text. The uptake kinetics of PCB congener #153 in 6 different body compartments (body
wall, pyloric caeca, gonads, rectal caeca, central digestive system, podia) are shown in
Figures 17 and 18 for the seawater and sediment exposures, respectively.
ng g
15000
Bodywall
3000
Central Digestive System
-1 lipids
10000
5000
800
600
400
200
12.7 1-e -
C(t)=
0
0 5 10 15 20 25 30 35
0
0 5 10 15 20 25 30 35
2000
1000
1000
2000
Pyloric Caeca Gonads
819 1-e -0.18*time
1+e -0.18*(time-23)
C(t)=
0.46*time -0.46*(time-12)
1+e
1500
1000
500
C(t)= 98.8
e 0.09*time
0
0 5 10 15 20 25 30 35
1417 1-e -
C(t)=
0.21*time -0.21*(time-23)
1+e
0
0 5 10 15 20 25 30 35
Figure 16. Asterias rubens-Seawater experiment. Uptake of 14 C-PCB 153 from seawater
in different body compartments of the sea star (mean concentration in ng g -1 total lipids ±
SD, n=3)
66
10000
Podia
200
100
8000
6000
4000
2000
0
0 5 10 15 20 25 30 35
300 Rectal Caeca
C(t)= 4.50
e 0.11*time
6.58 1-e -
C(t)=
0.33*time -0.33*(time-12)
1+e
0
0 5 10 15 20 25 30 35

4000
3000
2000
1000
Delineation of PCB uptake pathways in the sea star Asterias rubens
Figure 17. Asterias rubens-Sediment experiment. Uptake of 14 C-PCB 153 from sediments
in different body compartments of the sea star (mean concentration in ng g -1 total lipids ±
SD, n=3).
Contamination via seawater
Depending on the body compartment, accumulation from seawater was best described by a
combined (logistic and exponential) model (viz. uptake in body wall, pyloric caeca, gonads
and podia) or a single component exponential model (viz. uptake in central digestive system
and rectal caeca) (Fig. 17, Table 8A).
Table 8. Asterias rubens. Parameters and statistics of the equations describing the uptake of 14 C-PCB #153 from
seawater and sediments in the body compartments of the sea star.
E (exponential model): C(t) = C 0.e k.t ;
S (saturation model): C(t) = Css.(1-e -k.t );
C (combined model): C(t) = Css.(1-e -k.t )/(1+e -k.(t-I) );
where C 0, C(t), Css: 14 C-PCB #153 concentrations (ng g -1 lipids) respectively at time 0, at time t (d) and at
steady-state; k: rate constant (d -1 ); I: time (d) at the inflexion point; ASE: asymptotic standard error; R 2 :
corrected determination coefficient.
A. Seawater
ng g
5000
Bodywall
2000
Central Digestive System
-1 lipids
1000
0
0 5 10 15 20 25 30 35
800
600
400
200
Pyloric Caeca
3537 1-e -
C(t)=
588 1-e -3.0*time
1+e -3.0*(time-16)
C(t)=
0.33*time -0.33*(time-12)
1+e
0
0 5 10 15 20 25 30 35
1500
1000
500
2000
1500
1000
992 1-e 2000
-4.1*time
1+e -4.1*(time-16)
C(t)=
0
0 5 10 15 20 25 30 35
500
Gonads
C(t)= 74 e 0.09*time
0
0 5 10 15 20 25 30 35
Body compartment Model C 0 (ASE) Css (ASE) k (ASE) I (ASE) R 2
Body wall C 12,665 (691) 0.46 (0.13) 11.7 (0.67) 0.92
Central digestive system E 98.8 (27.7) 0.093 (0.009) 0.77
Rectal caeca E 4.5 (2.0) 0.11 (0.01) 0.91
Pyloric caeca C 819 (398) 0.18 (0.17) 22.9 (8.2) 0.80
Gonads C 1,417 (231) 0.21 (0.11) 23 (2.8) 0.90
Podia C 6,584 (449) 0.33 (0.12) 12.4 (0.87) 0.93
67
10000
8000
C(t)= 7618 (1-e -0.03*time )
6000
4000
40
30
20
10
Podia
0
0 5 10 15 20 25 30 35
50 Rectal Caeca
31 1-e -0.22*time
1+e -0.22*(time-20)
C(t)=
0
0 5 10 15 20 25 30 35

B. Sediments
Delineation of PCB uptake pathways in the sea star Asterias rubens
Body compartment Model C 0 (ASE) Css (ASE) k (ASE) I (ASE) R 2
Body wall C 3,537 (206) 0.33 (0.08) 12 (0.92) 0.93
Central digestive system C 992 (81) 4.1 (29) 16 (2.0) 0.81
Rectal caeca C 31 (2.3) 0.22 (0.05) 20 (1.1) 0.94
Pyloric caeca C 588 (18) 3.0 (11) 16 (1.5) 0.97
Gonads E 74 (19) 0.085 (0.008) 0.89
Podia S 7,618 (4266) 0.034 (0.029) 0.57
Table 9 Asterias rubens. Concentration factors (CF; maximum, minimum and mean values) in the body
compartments of the sea star after 34 days of exposure via seawater (A). Transfer factors (TF; maximum,
minimum, and mean values) in the body compartments of the sea star after 34 days of exposure via sediments
(B). CFs are calculated as the ratio between PCB 153 concentration in the sea star body compartments (ng g -1
total lipids) and its concentration in seawater (ng g -1 ). TFs are calculated as the ratio between PCB 153
concentration in the sea star body compartments (ng g -1 total lipids) and its concentration in sediments (ng g -1 dry
wt).
A. Seawater
Body wall Central dig. syst. Gonads Rectal caeca Pyloric caeca Podia
Max. CF 3.91 10 5
9.16 10 4
6.01 10 4
7.90 10 3
4.75 10 4
2.43 10 5
Min. CF 3.52 10 5
5.44 10 4
2.96 10 4
4.58 10 3
1.05 10 4
1.72 10 5
Mean CF 3.74 10 5
7.50 10 4
4.62 10 4
6.76 10 3
2.31 10 4
2.17 10 5
B. Sediments
Body wall Central dig. syst. Gonads Rectal caeca Pyloric caeca Podia
Max. TF 417 109 150 3.43 70 863
Min. TF 286 81 111 2.91 55 258
Mean TF 343 94 137 3.10 61 479
Body wall was the compartment that concentrated 14 C-PCB 153 to the greatest degree, up to
two orders of magnitude higher than the rectal caeca (p Tukey test ≤ 0.0001; Table 9A).
Body distribution of incorporated 14 C-PCB 153 varied significantly along the timecourse of
the experiment (p G-test < 0.05). Initially, the contaminant was mostly present in the podia (74 ±
5 % of total body load after 2 days of exposure) and secondarily in the body wall (26 ± 5 %).
Progressively, the proportion of the PCB associated with body wall increased, reaching 69 ± 5
% of the total body burden after 34 days of exposure, while during the same time the podia
proportion had decreased to 7 ± 2 % (Table 10).
Table 10 Asterias rubens. PCB distribution (mean % ± SD, n = 3) in the different body compartments of the sea
star after 34 days of exposure via seawater or sediments.
Body compartments
14 C-PCB-153 distribution (% )
Seawater exposure Sediment exposure
Body wall 68.8 ± 1.4 20.7 ± 4.4
Central digestive System 13.9 ± 4.1 8.9 ± 2.3
Gonads 8.4 ± 2.5 12.7 ± 2.5
Rectal caeca 0.2 ± 0.1 0.3 ± 0.1
Pyloric caeca 1.8 ± 0.7 5.7 ± 1.4
Podia 6.8 ±1.9 39.9 ± 14.6
68

Contamination via sediments
Delineation of PCB uptake pathways in the sea star Asterias rubens
Frequent radioanalysis of the contaminated sediments indicated that the maximum difference
between measured 14 C-PCB 153 activities was 13.1% and that no significant decreasing
trends occurred; therefore, concentrations in labelled PCB remained relatively stable
throughout the 34 d-long experiment (9.5 ± 1.1 ng g -1 dry wt; see Table 7). Similarly,
radioactivity in the seawater and in control sea stars remained below the detection limit,
indicating that no significant 14 C-PCB was incorporated from suspended sediments possibly
ingested by the mussels on which they fed nor from seawater due to cross contamination.
Accumulation from contaminated sediments was best described either by a single component
exponential model (gonads), a saturation exponential model (podia), or a combined model
(body wall, rectal caeca, pyloric caeca and central digestive system) (Fig. 17, Table 8B). As
noted during the seawater exposure, body wall and podia were the body compartments that
accumulated 14 C-PCB 153 to the highest levels when exposed to labelled sediments (Table
9B).
The distribution of 14 C-PCB in sea star tissues was determined at different times during the
timecourse of the experiment. Relative transfers among body compartments appeared quite
different from those observed during the seawater uptake experiment. Indeed, the proportion
of contaminant in the body wall and podia remained relatively constant throughout the
experiment. Body wall and podia contained the major part (ca. 60%) of the total body burden
of 14 C-PCB, while the lowest percentage was found in the rectal caeca (≤ 0.3%) (Table 10).
DISCUSSION
The present study reports the first experimental data on the bioaccumulation kinetics of a key
PCB congener in the sea star Asterias rubens, a common species widely distributed in the
North Sea and NE Atlantic. The fact that organisms were also exposed to very low
background concentrations of stable PCB 153 (Table 7), showed that they were actually
exposed to a global concentration of PCB 153 that did not differ significantly from the 14 C-
PCB concentrations added experimentally to seawater or sediments (Table 7). Experimental
concentrations in seawater were higher than those usually reported for PCB 153 in the natural
North Sea waters. However, the latter concentrations most generally concern the dissolved
fraction whereas our measurements involve both dissolved and particulate fractions. Although
available PCB data on bulk seawater samples mostly concern the sum of congeners or PCB
mixture equivalents, it is noteworthy that the experimental concentrations used here are quite
69

Delineation of PCB uptake pathways in the sea star Asterias rubens
close (even much lower if considering extreme hot spots) to values reported for moderate to
highly contaminated marine locations (Table 7). In addition, the ratio between seawater and
sediment PCB concentrations added is similar to the ratio between the background PCB
concentrations measured in seawater and sediments used in the experiments (Table 7).
Therefore, the experimental exposures may be considered as acceptable simulations of
exposure situations that may actually occur in the field.
Data on PCB concentrations in Asterias rubens in the field are scarce, and even less are
available when looking for congener-specific data (e.g. Everaarts et al. 1998, den Besten et al.
2001). It is noteworthy that the total PCB 153 concentrations (background + incorporated)
reached in the pyloric caeca at the end of the experiments matched the concentrations reported
in the same organs of sea stars from moderate to highly contaminated North Sea locations
(Table 11).
Table 11 Asterias rubens. Comparisons among PCB #153 concentrations obtained in the present study
(background + incorporated concentrations) and previous field studies in the North Sea.
Body compartment PCB 153 concentration
(ng g -1 lipids)
70
Specifications References
Pyloric caeca 608 - 1,111 Experimental conditions;
seawater uptake
920 - 1,377 Experimental conditions;
sediment uptake
41 - 1,054 Pre-spawning period 1995;
southern North Sea
450 - 1,050 Pre-spawning period 1995;
Dutch coastal zone
40 - 125 Pre-spawning period 1995;
southern North Sea
Bodywall 728 - 4,360 Experimental conditions;
seawater uptake
1,215 - 14,068 Experimental conditions;
sediment uptake
Whole body 8,190 - 9,300 Experimental conditions;
seawater uptake
2,330 - 2,810 Experimental conditions;
sediment uptake
550 - 940 Spawning period 1986;
Dutch coastal zone
100 - 235 Spawning period 1986;
southern North Sea
4,300 (sum35 PCB) Spawning period 1989; German
Bight
2,400 (sum35 PCB) Post-spawning period 1989;
German Bight
Present study
Present study
den Besten et al.. 2001
Everaarts et al.. 1998
Everaarts et al.. 1998
Present study
Present study
Present study
Present study
Everaarts & Fischer 1989
Everaarts & Fischer 1989
Knickmeyer et al.. 1992
Knickmeyer et al.. 1992

Delineation of PCB uptake pathways in the sea star Asterias rubens
No field data were found concerning PCB concentrations in the body wall. Considering the
whole body, PCB concentrations reached in experimentally-exposed sea stars were 2 to 10
times higher than the few data available from the literature (Everaarts & Fischer 1989; Table
11). However, these comparisons should be made with caution, since the latter field values
are derived from sea stars collected during the spawning period. Indeed, it has been shown
that the whole-body content of extractable lipids is strongly dependent on the sexual state of
individuals, and may fluctuate by a factor of 2 to 3, particularly during the spawning period.
This may result in a similar range of variations of PCB concentrations occurring within a few
weeks (Knickmeyer 1992; Everaarts et al. 1998; Table 11).
Whether seawater or sediments were considered as a contamination source, a steady-state was
reached or tended to be reached in most body compartments during the course of the
experiments. This suggests either that target sites are rapidly saturated, or that a
metabolization mechanism is induced quite rapidly following PCB exposure. Although a
MFO-like system has been described in pyloric caeca of A. rubens by den Besten et al. (1990,
1993, 1998), it is well documented that PCB 153 is quite resistant to biological degradation
(Sipes & Schnellmann 1987, Letcher et al. 2000) due to its specific structure lacking
hydrogen atoms on the biphenyl molecule (Borlakoglu & Wilkins 1993). Therefore, the
hypothesis regarding target site saturation is considered to be the most plausible explanation.
It is also noteworthy that when a steady-state in uptake was observed, equilibrium
concentrations of PCB 153 were generally reached quite rapidly (around day 20), indicating
that the sea star could be used as a bioindicator to pinpoint a PCB contamination event soon
after its occurrence.
Concentrations of incorporated 14 C-PCB 153 at steady-state were much higher (up to 300
times) in body wall and podia than in any other compartment. Being easily dissected and
constituting 70-80% of the total body weight, body wall is of particular interest with respect
to field surveys, and it should be recommended as a body compartment to monitor
complementarily to pyloric caeca which are the only body compartment that has been used in
previous studies (e.g., Everaarts et al. 1998, den Besten et al. 1993, 2001).
Concentrations incorporated into the rectal caeca were always low, between one and two
orders of magnitude lower than all the other compartments. This is somewhat surprising but
could be related to the functions of the rectal caeca which are well known to play an essential
role in sea star digestion and excretion processes (Jangoux 1982, Warnau & Jangoux 1999).
Our results have shown that PCB uptake is far more efficient in sea stars exposed to spiked
seawater than to labelled sediments when compared to exposure concentrations. For a given
71

Delineation of PCB uptake pathways in the sea star Asterias rubens
body compartment, calculated concentration factors (CFs) based on seawater were between 2
and 3 orders of magnitude higher than transfer factors (TFs) from sediments (Tables 9A and
9B). Therefore, over the long term, despite the fact that sediments constitute the main
reservoir of PCBs in the marine environment and that seawater PCB concentrations are
comparatively extremely low, seawater would be an important route for PCB
bioaccumulation in this sea star as it has been suggested for certain benthic infauna (e.g.
Fowler et al. 1978). However, this does not imply that seawater would be the predominant
pathway for PCB uptake, since our results showed that final concentrations reached in the
different body compartiments following the two types of exposure were generally of the same
order of magnitude. In addition, direct trophic transfer was not addressed here and could also
contribute significantly to PCB bioaccumulation in the sea star.
While this work constitutes the first report on PCB bioaccumulation kinetics in a sea star,
several previous studies have used radiolabelled 14 C-PCB to examine bioaccumulation
kinetics in other aquatic organisms (e.g., Goerke et al. 1973, Gooch & Hamdy 1982,
Schweitzer et al. 1997). However, surprisingly, these studies mostly concern PCBs as Aroclor
equivalents (see e.g. Butcher et al. 1997). The main advantage of the 14 C approach to measure
PCB fluxes and transfers in aquatic biota is obviously the high sensitivity and the rapidity of
the detection, compared to analytical techniques using gas chromatography. It therefore
constitutes an interesting tool, since current research on the behaviour of PCBs in the
environment tends to focus on congener-specific information (Safe 1990, Metcalfe 1994,
Letcher et al. 2000). Furthermore, it allows working with low (realistic) PCB concentrations,
and assessing uptake in organs which are often too small to be analyzed by classical chemical
methodologies.
ACKNOWLEDGEMENTS
The IAEA Marine Environment Laboratory operates under a bipartite agreement between the
International Atomic Energy Agency and the Government of the Principality of Monaco. B.D.
is holder of a FRIA doctoral grant; M.W. is a Honorary Research Associate of the National
Fund for Scientific Research (NFSR, Belgium). Research was partially supported by a
Belgian Federal Research Programme (SSTC, Contract MN/11/30) and a NFSR fellowship to
M.W.
72

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
III.3 Coplanar PCB uptake kinetics in the common sea star Asterias rubens
and subsequent effects on ROS production and CYP1A induction
Marine Ecology Progress Series (submitted)
Danis B a , Cotret O b , Teyssié JL b , Fowler SW b & Warnau M b
a: Laboratoire de Biologie marine (CP 160/15), Université Libre de Bruxelles, Av. F.D.
Roosevelt 50, B-1050 Brussels, Belgium
b: International Atomic Energy Agency, Marine Environment Laboratory, 4 Quai Antoine I er ,
MC-98000 Monaco
73

ABSTRACT
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
The kinetic behaviour of a highly toxic PCB congener (IUPAC #77) was investigated in sea
stars experimentally exposed via seawater, sediments, or food. Simultaneously, biological
effects were assessed at the immune and subcellular levels, respectively, by measuring
reactive oxygen species (ROS) production and cytochrome P450 immunopositive protein
(CYP1A IPP) induction. Results indicate that sea stars efficiently accumulate the
contaminant, and that most organs bioconcentrate the congener according to saturation
kinetics. Biological effects were pronounced and affected essential functions of the sea star
biology (viz., immune and detoxification systems) following exposure to environmentally
realistic PCB concentrations. These findings stress the need to (1) obtain information about
similar coplanar-specific, biological effects in other organisms in the natural environment and
(2) include coplanar PCBs in the list of congeners to be systematically measured in marine
biomonitoring programmes.
KEYWORDS
Polychlorinated biphenyls; PCB 77; bioaccumulation; Asterias rubens; echinoderm;
CYP1A; reactive oxygen species; immune system
74

INTRODUCTION
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Polychlorinated biphenyls (PCBs) are a class of widespread stable and persistent
contaminants which have been widely used in industrial applications, such as electrical and
hydraulic equipment, plastics, lubricants, pesticides and flame retardants (Metcalfe 1994).
PCBs include 209 congeners, which differ by the chlorine substitution on the biphenyl rings.
These compounds are found in the environment (e.g. Metcalfe 1994, Schreitmüller et al.
1994, Bright et al. 1995, Wania & Daly 2002), and due to their inherent chemical, physical,
and toxicological properties, bioaccumulation of these compounds in marine biota is of
growing concern (Stebbing et al. 1992, OSPAR 2000). These compounds are readily
accumulated by organisms (Fowler et al. 1978, Boese et al. 1996, Koponen et al. 1998), and
are known to have deleterious effects on key biological processes including reproduction,
development, and immunity (Harding & Addison 1986, Zabel et al. 1995, Chapman 1996,
Krogenaes et al. 1998, Coteur et al. 2001).
From a toxicological point of view, some congeners appear to be more problematic than
others; indeed, the non-ortho- and mono-ortho-chlorinated congeners can display planar
configuration (c-PCBs). The toxicity of c-PCBs is produced through a receptor-mediated
response involving the binding of the contaminant to the cytosolic aryl hydrocarbon (Ah)
receptor followed by changes in gene expression (Kohn 1983, Shugart et al. 1992, Safe 1995,
Hahn 1998, Nebert et al. 2000). c-PCBs are known to induce CYP1A which is the major
enzyme responsible for the metabolic activation of promutagens and procarcinogens.
However, not all the adverse biological effects of PCBs are attributable to this mechanism. In
particular, there is increasing evidence that many PCBs can act both as endocrine disrupters
and immunosuppressants through more complex and diverse pathways such as alterations of
kinases and phospholipases, disturbance of Ca 2+ homeostasis, and modulation of gene
expression (Satar 2000, Chiu et al. 2000, Arukwe 2001).
In assessing the impending risk in the marine environment caused by these contaminants
monitoring biological effects is a fundamental issue (Suter 1993). To reach this goal,
biomarkers are precious tools as they offer early warning signals (den Besten 1998). A
biomarker is a variation of a biological response (ranging from molecular to cellular and
physiological responses) which can be related to exposure or to toxic effects of chemicals
(Bayne et al. 1988, Peakall 1992). In an environmental context, biomarkers are sensitive
indicators of the entrance of a toxicant in an organism, its distribution among organs and/or
tissues, and its toxic effect on critical processes (e.g. McCarthy & Shugart 1990).
75

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
The common NE Atlantic sea star, Asterias rubens, is considered as a valuable test organism
because of its key position in benthic ecosystems, and more particularly in the
"seston–mussel–sea star" food chain (den Besten et al. 2001). In previous studies, it has been
shown that the accumulation of PCB mixtures in this food chain can lead to adverse effects on
the reproduction of the sea star (e.g. den Besten et al. 1989). The effects of such contaminants
have been assessed at the subcellular level using biomarkers such as benzo[a]pyrene
hydroxylase (BaPH) activity (den Besten 1998b, den Besten et al. 1990b, 1993,) and hormone
synthesis rate (den Besten et al. 1991). Furthermore, in vitro or in vivo exposure of sea stars to
PCBs has been reported to affect DNA integrity, larval development, and immune and
detoxification systems (den Besten et al. 1990b, 1991, 1993, Everaarts & Sarkar 1995).
The aim of the present work was to describe (1) the accumulation and tissue-distribution of a
coplanar PCB congener (IUPAC#77) in the sea star Asterias rubens exposed via seawater,
sediments or food pathways and (2), in parallel, to assess the induced biological effects at the
immune and subcellular levels. Accumulation and loss kinetics were studied by means of a
14 C-labelled congener. This method has been used in a previous study (Danis et al. Chap.
III.2), and displays many advantages over conventional methods (viz. classical GC-ECD or
GC-MS detection methods) in that it allows working at low, realistic levels of contamination,
and with small organs of very low weight. Biological effects were assessed by measuring
reactive oxygen species (ROS) production by immune cells, and CYP1A immunopositive
protein (CYP1A IPP) induction in sea star pyloric caeca.
MATERIALS AND METHODS
Sampling
The sea stars, Asterias rubens (Linnaeus 1758), were collected in April 2002 in the intertidal
zone at Audresselles (Pas-de-Calais, France). Prior to experimentation, 250 specimens of
similar size (5-7 cm arm radius) and weight (36 ± 3.5 g) were acclimated to laboratory
conditions for 1 month (constantly aerated open circuit aquaria, 34 p.s.u., 16 ± 0.5 °C, 12/12 h
dark/light cycle). Mussels (Mytilus galloprovincialis) were collected off "la Pointe des
Douaniers" (Cap d'Ail, France), and all specimens were held under similar controlled
laboratory conditions until used in experiments.
Radiotracer
The 14 C-labelled 3,3',4,4' hexachlorobiphenyl (purity ≥ 95%) was purchased from Sigma
Chemicals, USA. Specific activity was 25 MBq mmol -1 . Stock solutions were prepared in
acetone at a concentration of 1 µg ml -1 .
76

Sample treatment and liquid scintillation counting
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Water samples (2 ml) were directly transferred to 20 ml glass scintillation vials (Packard,
USA) and 10 ml of Ultima Gold XR ® (Packard Instruments) scintillation liquid were added.
Samples of sediment, mussel or sea star tissues (previously ground) were placed in a vial
containing 2 ml of Acetonitrile ® in an ultrasonic bath for 10 min. The Acetonitrile ® was then
collected and replaced by another 2 ml of Acetonitrile ® and the ultrasonic operation was
repeated a second time. This treatment gave 4 ml of liquid phase (viz., the extract) and a
residue. The residue was digested overnight at 70°C with 2 ml of Soluene ® , followed by an
addition of 10 ml of Hionic Fluor ® scintillation liquid. The liquid phase (4 ml) was added to
16 ml of filtered seawater and extracted twice using 2 ml of n-Hexane (Sigma, USA) under
constant agitation. The organic phase (4 ml) and aqueous phase (20 ml) were treated
separately; the entire organic phase and 2 ml of the aqueous phase were each added separately
to 10 ml of Ultima Gold XR ® scintillation liquid.
14 C-radioactivity was then measured using a 1600 TR Liquid Scintillation Analyzer (Packard),
compared to standards of known activities, and corrected for quenching, background and
physical decay of the radiotracer. Counting times were adjusted to obtain counting rates with
relative propagated errors less than 5 %. PCB concentrations were expressed on a total lipid
content basis where lipids were determined gravimetrically (UNEP 1990). A schematic
diagram of the sample treatment is shown in Figure 18.
Figure 18. Schematic representation of sample processing before liquid scintillation counting
77

Experimental procedures
Uptake from seawater
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Sea stars (n = 25) were placed for 15 d in a 70 l glass aquarium (constantly aerated closed
circuit aquaria; 34 p.s.u.; 16 ± 0.5°C; 12/12 h dark/light cycle) containing natural seawater
spiked with 14 C-labelled PCB 77. One day prior to the experiments, four 5 l glass beakers
were filled with filtered seawater (34 p.s.u.; 16 ± 0.5°C), spiked with the radiolabelled PCB
stock solution, and constantly stirred using an orbital agitation plate. Spiked water was then
poured into the glass aquarium and natural seawater was added to obtain a final volume of
70l. Seawater and radiotracer were renewed every second day during the entire experiment.
Activity was checked before and after each renewal to assess the stability of the labelled PCB
concentration in seawater. The sea stars were fed unlabelled mussels (M. galloprovincialis)
every second day just before the seawater renewal. After 2 hrs, uningested mussels were
removed to limit as much as possible PCB incorporation via the food (PCB measurements in
uningested mussel soft parts were always below detection limit). Periodically, sea stars (n =
3) were removed, dissected into six body compartments (oral body wall, aboral body wall,
pyloric caeca, gonads, rectal caeca, and central digestive system), and radioanalyzed to
determine uptake kinetics.
Uptake from sediments
Sediments (2.5 kg dry wt) from the North Sea (Audresselles, Pas-de-Calais, France) were
contaminated for 4 days with the 14 C-labelled PCB using the rolling jar method (Murdoch et
al. 1997). Sea stars (n = 60) were placed for 34 d in a 70 l glass aquarium (constantly aerated
open circuit aquarium; flow 30 l hr -1 ; 34 p.s.u.; 16 ± 0.5°C; 12/12 h dark/light cycle)
containing a 10 cm layer of seawater running over a 2 cm layer of spiked sediments. A
separate group of 5 sea stars were placed in the same aquaria, but in another compartment
(not in contact with the sediments), to serve as a control for possible cross-contamination
from labelled PCB in seawater. The sea stars were fed every second day with fresh mussels
and any uningested food was removed after 2 hrs (PCB measurements in uningested mussel
soft parts were always below detection limit). The radioactivity of the labelled PCB was
measured weekly in the sediments to check for possible leaching. Periodically, 3 individuals
were removed, dissected as described above, and their tissues counted for radioactivity.
Uptake from food
Before the feeding experiment, mussels were exposed for 2 d in a glass aquarium containing 4
l of filtrated seawater spiked with 18 ng l -1 PCB 77. Radiolabelled seawater was changed daily
78

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
and mussels were regularly fed with phytoplankton (Isochrysis galbana). After 2 days of
exposure to the 14 C-PCB congener, mussels were fed to the sea stars. Mussel exposure was
carried out daily to obtain each day food that had been radiolabelled for 2 days. Sixty sea stars
were placed for 34 d in a 70 l glass aquarium (constantly aerated open circuit aquarium; flow
30 l hr -1 ; 34 p.s.u.; 16 ± 0.5°C; 12/12 h dark/light cycle). Sea stars were allowed to feed daily
for 1 hr on radiolabelled mussels (1 mussel per sea star) and then individuals were
periodically sampled to measure PCB uptake kinetics.
Reactive Oxygen Species (ROS) production measurements
Amoebocyte ROS production was measured by the peroxidase, luminol – enhanced method
optimized by Coteur et al. (2001). Briefly, 3 ml of coelomic fluid were collected in the same
volume of anticoagulant buffer. The cell concentration was measured by absorbance at 280
nm using a Tecan Spectrafluor Plus plate-reader. This suspension was then centrifuged for 10
min at 400 g and resuspended in Ca 2+ -, Mg 2+ -free artificial seawater (ASW), the volume of
which was adjusted to obtain a final amoebocyte concentration of 10 6 cells ml -1 . A stock
solution of luminol and horseradish peroxidase (HRP) in DMSO was freshly diluted 100 fold
in ASW (final concentrations of HRP and luminol were 500 µg ml -1 and 250 µg ml -1 ,
respectively). The reaction was begun by adding 200 µl of amoebocyte suspension in 100 µl
of luminol / HRP solution, and 20 µl of a Micrococcus luteus suspension containing 2.5x10 9
bacteria ml -1 (stimulated amoebocytes) or 20 µl of ASW (non-stimulated amoebocytes). The
chemiluminescence was measured every 10 min over a 2 h period using a Tecan Spectrafluor
Plus plate-reader placed in an incubator thermostabilized at 14 ± 0.5°C. Results were
expressed as the sum of all 10 min interval measurements for 10 6 cells ml -1 (total
chemiluminescence) of bacteria-stimulated amoebocytes or non-stimulated amoebocytes.
Cytochrome P450 immunopositive protein (CYP1A IPP) quantification
CYP1A IPP content was quantified using a competitive-ELISA method which has been fully
described elsewhere (Danis et al. Chap. III.4). Briefly, ELISA was carried out using
competition between the CYP1A IPP contained in the pyloric caeca of PCB-exposed sea stars
and a biotinylated CYP1A from ß-naphtoflavone (BNF)-injected trout (Oncorhyncus mykiss).
Multiwell plates (96 wells) were coated with Anti-CYP1A (rabbit anti-fish CYP1A peptide,
polyclonal antibody; Biosense, Norway). Wells were washed with phosphate-buffered saline
(PBS), and nonspecific binding sites were blocked with PBS-Bovine serum albumin (BSA).
Wells were washed again and biotinylated microsomes of BNF-injected trout were added
(except for the blank wells). Sea star samples or standards (with adjusted protein
concentration to 100 µg ml -1 ) were then added to the wells. Competition was allowed to take
79

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
place for 2 hrs, and after five washing steps extravidin-HRP was added to all the wells. The
plate was then incubated for 45 min and the wells washed again using PBS. Chromogen TMB
(Biosource, UK) was added to all the wells and the plate was incubated in the dark for 10
min. Sulfuric acid was then added to stop the reaction and absorbance was measured at 450
nm using a micro plate reader (Packard, Spectracount). Final results were expressed as
induction factors, viz., the ratio of CYP1A IPP levels between experimental and control
groups.
Data analyses
Uptake of the PCB congener from seawater, sediments or food was expressed as change in
PCB concentration (ng g -1 total lipids) over time. Uptake kinetics were described either by
using a saturation exponential model (Equation 5), or a linear model (Equation 6):
Equation 5: C(t) = Css (1-e -k e .t )
Equation 6: C(t) = k u.t
where C(t) and Css are PCB concentrations (ng g -1 total lipids) at time t (d) and steady state,
respectively, k u and k e are the respective biological uptake rate constants (d -1 ) (Whicker &
Schultz 1982). The model showing the most accurate fit (based on calculation of the
determination coefficient, R 2 , and examination of the residuals) was selected.
Constants of the models and their statistics were estimated by iterative adjustment of the
models and Hessian matrix computation, respectively, using the nonlinear curve-fitting
routines in the Systat ® 5.2.1 software (Wilkinson 1988). Differences among PCB
concentrations in the different sea star body compartments and between ROS production and
CYP1A IPP levels in the different exposure conditions were tested by 1-way ANOVA and the
multiple comparison test of Tukey (Zar 1996). Dose-response relationships were tested using
linear and non-linear regressions. The level of significance for statistical tests was always set
at a = 0.05.
RESULTS
Uptake from seawater
Regular monitoring of radiotracer activities in seawater allowed calculation of time-integrated
dissolved concentration of PCB 77; its value was 12.9 ± 0.45 ng l -1 . Uptake of the
contaminant in the different body compartments displayed saturation kinetics, except in the
80

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
gonads, in which the PCB was linearly accumulated (Fig. 19, Table 12). During the
experiment (15 d), the digestive organs (viz. pyloric and rectal caeca and central digestive
system) reached saturation, while saturation was almost reached in oral and aboral body
walls.
By order of decreasing bioaccumulation efficiency (Fig. 19, Table 12), the PCB 77
concentrations at the end of the exposure period were 286 ± 89.7 ng g -1 lipids in the gonads,
262 ± 55.2 in the rectal caeca, 240 ± 32.5 in the oral body wall, 157 ± 26.1 in the aboral body
wall, 53.1 ± 7.14 in the pyloric caeca and 11.9 ± 3.05 in the central digestive system.
400
300
200
100
200
150
100
0
0 5 10 15
50
0
0 5 10 15
Figure 19. Asterias rubens. Uptake of 14 C-PCB 77 from seawater in different body compartments
of the sea star (mean concentration in ng g –1 total lipids ± SD, n = 3).
Table 12 Asterias rubens. Parameters and statistics of the equations describing the uptake of 14 C-PCB #77 in
different body compartments of the sea star. L (linear model): C(t)=kt; S (saturation model): C(t)=Css.(1-e -kt );
where C(t) and Css: 14 C-PCB #77 concentrations (ng g -1 lipids) respectively at time t (d) and at steady-state; k:
rate constant (d -1 ); ASE: asymptotic standard error; R 2 : corrected determination coefficient.
Seawater exposure
500
Oral Bodywall Gonads Rectal Caeca
Aboral Bodywall
400
300
200
100
0
0 5 10 15
Time (d) Time (d)
Time (d)
100
80
60
40
20
0
0 5 10 15
Body Compartment Model Css (ASE) k (ASE) R 2
Oral body wall S 326 (53.3) 0.11 (0.03) 0.84
Aboral body wall S 190 (22.9) 0.14 (0.03) 0.84
Gonads L 19.2 (2.21) 0.53
Pyloric caeca S 52 (3.30) 0.77 (0.20) 0.37
Rectal caeca S 289 (18.7) 0.32 (0.06) 0.72
Central digestive
system
S 13.0 (0.95) 0.47 (0.13) 0.53
81
400
300
200
100
20
15
10
0
0 5 10 15
Pyloric Caeca Central Dig. System
5
Time (d)
0
0 5 10 15
Time (d) Time (d)

Sediments exposure
Food exposure
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Oral body wall S 125 (32.3) 0.05 (0.02) 0.59
Aboral body wall S 194 (14.9) 0.22 (0.05) 0.62
Gonads S 256 (67.6) 0.04 (0.02) 0.80
Pyloric caeca S 109 (7.09) 0.17 (0.03) 0.78
Rectal caeca S 300 (11.5) 0.11 (0.01) 0.95
Central digestive
system
S 11.7 (0.82) 0.37 (0.11) 0.57
Oral body wall S 76.7 (5.65) 0.16 (0.04) 0.72
Aboral body wall S 195 (18.2) 0.09 (0.02) 0.82
Gonads L 4.85 (0.34) 0.67
Pyloric caeca S 141 (6.31) 0.14 (0.02) 0.89
Rectal caeca S 190 (11.7) 0.16 (0.03) 0.77
Central digestive
system
S 43.1 (2.84) 0.16 (0.03) 0.76
Table 13 Asterias rubens. Concentration and Transfer factors, CF and TF (mean ± SD; n=24 for seawater
exposure and n=33 for sediments and food exposures) in body compartments at the end of exposure periods via
seawater, sediments or food. CFs calculated as ratio between PCB 77 concentration in body compartments (ng
g –1 total lipids) and its concentration in seawater (ng g -1 ). TFs calculated as ratio between PCB 77 concentration
in body compartments (ng g –1 total lipids) and its concentration in sediments (ng g -1 dry wt) or in food (ng g -1
total lipids). BW: bodywall, Pyl. Caec.: pyloric caeca, Rect. Caec.: rectal caeca, C.D.S.: central digestive system
Experiment Oral BW Aboral BW Gonads Pyl. Caec. Rect. Caec. C.D.S.
Seawater 18,600 ± 2520 12,200 ± 2020 22,100 ± 6950 4,110 ± 550 20,300 ± 4280 920 ± 237
Sediments 8.31 ± 1.74 14.1 ± 3.80 11.9 ± 1.80 7.37 ± 2.20 21.5 ± 2.06 0.97 ± 0.35
Food 2.84 ± 1.30 7.93 ± 2.03 5.90 ± 1.12 5.26 ± 0.73 6.16 ± 1.18 1.51 ± 0.22
Correlations, calculated between PCB 77 concentrations measured in the different body
compartments, showed that concentrations measured in oral body wall were significantly
correlated to those measured in aboral body wall (r = 0.68; Fig. 20), gonads (r = 0.50), rectal
caeca (r = 0.37) and central digestive system (r = 0.29). Significant correlation was also found
between PCB 77 concentrations measured in aboral body wall and those measured in rectal
caeca (r=0.50) and gonads (r = 0.43).
82

PCB concentration in Oral bodywall (ng g -1 lipids)
250
200
150
100
50
r=0.73
p

200
150
100
50
300
200
100
0
0 5 10 15 20 25 30 35
0
0 5 10 15 20 25 30 35
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Figure 21. Asterias rubens. Uptake of 14 C-PCB 77 from sediments in different body
compartments of the sea star (mean concentration in ng g –1 total lipids ± SD, n = 3).
The estimated mean transfer factors (TFs) indicated that bioaccumulation in the body
compartments was ranked differently than in the case of seawater-exposed sea stars (Table
13), with rectal caeca showing the highest TF, followed by aboral body wall and gonads. In
addition, TF values were 3 orders of magnitude lower than CFs calculated in the seawater
experiment in which sea stars were exposed for only half the time.
Significant correlations were calculated between PCB concentrations measured in the
different body compartments at the end of the exposure period; the strongest one was found
for pyloric vs rectal caeca (r=0.73; Fig. 22), followed by weaker correlations for gonads vs
oral body wall (r=0.51), and aboral body wall vs rectal caeca (r=0.34).
300
Oral Bodywall Gonads Rectal Caeca
Aboral Bodywall
200
100
200
150
100
0
0 5 10 15 20 25 30 35
50
0
0 5 10 15 20 25 30 35
84
400
300
200
100
20
15
10
0
0 5 10 15 20 25 30 35
Pyloric Caeca Central Dig. System
5
0
0 5 10 15 20 25 30 35

PCB concentration in Rectal caeca (ng g -1 lipids)
450
400
350
300
250
200
150
100
50
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
85
y = 2.23x + 1.70
0
0 20 40 60 80 100 120 140 160 180 200
PCB concentration in Pyloric caeca (ng g -1 lipids)
Figure 22. Linear regression between 14 C-PCB 77 concentrations (ng g –1 total lipids)
measured in rectal and pyloric caeca of sea stars exposed for 34 d to contaminated
sediments. r: correlation coefficient.
Uptake from food
r=0.67
p

100
80
60
40
20
300
200
100
0
0 5 10 15 20 25 30 35
0
0 5 10 15 20 25 30 35
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Figure 23. Asterias rubens. Uptake of 14 C-PCB 77 from food (Mytilus galloprovincialis) in
different body compartments of the sea star (mean concentration in ng g –1 total lipids ± SD, n
= 3).
TFs from food were of the same order of magnitude than those calculated for sediment-
exposed sea stars; they reached the maximal value of 7.93 in the aboral body wall after 34 d
of exposure (Table 13).
Correlations were found between PCB concentrations measured in the different body
compartments at the end of the feeding period. The highest correlation was found between the
pyloric and the rectal caeca (r=0.67; Fig. 24).
PCB concentration in Rectal caeca (ng g -1 lipids)
300
250
200
150
100
50
200
Oral Bodywall Gonads Rectal Caeca
Aboral Bodywall
r=0.67
p

ROS production
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Reactive oxygen species (ROS) production was measured in amoebocytes collected from the
sea stars sampled during the seawater, sediments, or food experiments; these results are
presented in Fig. 25. In the case of seawater-exposed sea stars, the measurement of ROS
production in non-stimulated amoebocytes showed no clear trends. However, bacteria-
stimulated amoebocytes were induced to produce more ROS during the first five days, while
afterwards ROS production decreased again to control levels until the end of the experiment
(Fig. 25A). During the sediment experiment (Fig. 25B), an increasing trend was observed
during the first 9 days of exposure, at which time ROS levels dropped dramatically. The same
trends were also observed in amoebocytes from sea stars exposed to labelled food; in this
experiment maximum ROS production was reached after 5 days of exposure (Fig. 25C).
A.
ROS production (sum LCL)
B.
ROS production (sum LCL)
60000
50000
40000
30000
20000
10000
400000
350000
300000
250000
200000
150000
100000
Non-stimulated Stimulated
0
0 5 10 15
50000
87
Time (d)
Non-stimulated Stimulated
0
0 5 10 15 20 25 30 35
Time (d)

C.
Ros production (sum LCL)
300000
250000
200000
150000
100000
50000
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Non-stimulated Stimulated
0
0 5 10 15 20 25 30 35
Time (d)
Figure 25. Asterias rubens. ROS production (sum LCL; non-stimulated
and bacteria-stimulated amoebocytes; mean±SD; n = 3) measured in
sea stars exposed to 14 C-PCB 77 via (A) seawater, (B) sediments, or (C)
food.
Correlations between non-stimulated and bacteria-stimulated ROS production were found for
all exposure routes, but with variable correlation coefficient values: r seawater=0.72,
r sediments=0.94, r food=0.50.
CYP1A immunopositive protein induction
Cytochrome P450 (CYP1A) immunopositive protein (CYP1A IPP) induction was measured
in pyloric caeca collected from experimental sea stars exposed to PCB 77 via seawater,
sediments, or food (Fig. 26). Induction of CYP1A IPP followed saturation kinetics for
whichever exposure route was considered (Table 14). Saturation values were higher in the
case of sediments and food exposures, but the rate constant was almost 3 times faster in the
case of seawater exposure.
88

CYP1A immunopositive protein (induction fold)
3
2,5
2
1,5
Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
Sea water Sediments Food
1
0 5 10 15 20 25 30 35
Time (d)
Figure 26 Asterias rubens. CYP1A IPP induction (ratio of experimental response to
control group; mean±SD; n = 3) measured using competitive ELISA in sea stars
exposed to 14 C-PCB 77 via seawater, sediments or food.
Table 14 Asterias rubens. Parameters and statistics of the equations describing the induction of CYP1A
immunopositive protein during the exposure experiments. Kinetics were described using a saturation model:
C(t)=Css(1-e -lt ) where l is the rate constant (d -1 ). Other symbols as in Table 12.
Exposure route Css (ASE) l s (ASE) R 2
Seawater 1.57 (0.04) 1.16 (0.18) 0.50
Sediments 1.91 (0.06) 0.40 (0.09) 0.42
Food 1.98 (0.06) 0.42 (0.08) 0.37
For each exposure route, significant correlations were found between CYP1A IPP induction
and PCB concentrations measured in seastars body compartments. The highest correlations
were found for gonads (r=0.55) during sediment exposure, aboral body wall (r=0.45)
during seawater exposure, and oral body wall (r=0.29) during food exposure.
DISCUSSION
Although sea stars have been reported to efficiently concentrate PCBs, available literature
mainly concerns field measurements of these contaminants in the tissues and organs, and
more specifically concerns non-coplanar congeners (e.g. Stebbing et al. 1992, Everaarts et al.
1998, den Besten et al. 2001, Stronkhorst et al. 2003). One previous experimental study
focused on bioaccumulation kinetics of the non-coplanar congener #153 following exposure
89

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
via sediments or seawater (Danis et al. Chap. III.2). To the best of our knowledge, no
previous information is available regarding the relative importance of the different exposure
routes on uptake of coplanar congeners in marine invertebrates.
In the present study, biokinetic experiments were performed using a 14 C-labelled PCB
coplanar congener (IUPAC#77). The procedure was designed in order to have added
contaminant concentrations on the same order as those found in contaminated natural marine
environments. In general, the coplanar PCB congener was readily accumulated by A. rubens
exposed via seawater (CFs ranging from 672 to 22,100 according to the body compartment
after 15 d of exposure), via sediments (TFs ranging from 1 to 22 after 34 d of exposure) or via
food (TFs ranging from 1.5 to 7.9 after 34 d of exposure). In a previous study using the non-
coplanar congener PCB 153, maximum TFs and CFs were, respectively, up to 1 and 2 orders
of magnitude higher than in the present study (Danis et al. Chap. III.2).
When considering CFs and TFs, the relative bioaccumulation efficiency was up to 3 orders of
magnitude higher in seawater-exposed animals than when sea stars were exposed via
sediments or food. The same observation was made for PCB 153 (Danis et al. Chap. III.2);
i.e. PCB uptake was far more efficient in sea stars exposed to spiked seawater than in those
exposed to labelled sediments when related to exposure concentrations. Therefore, over the
long term, despite the fact that sediments constitute the main reservoir of PCBs in the marine
environment and that comparative seawater PCB concentrations are extremely low, seawater
could be a non-negligible route for PCB bioaccumulation in the sea star, as has been
suggested for other benthic infauna (e.g., Fowler et al. 1978). However, this does not imply
that seawater would be the predominant pathway for PCB uptake, since results show that final
concentrations reached in the different body compartments following the three types of
exposure were generally similar.
In contrast with what was observed by Danis et al. (Chap. III.2) for PCB 153, concentrations
of PCB 77 incorporated into the rectal caeca were in the same range as those measured in
other tissue compartments. Although we do not have a clear explanation for this observation,
it is nevertheless interesting, since PCB 153 concentrations in rectal caeca were between 1
and 2 orders of magnitude lower than in all the other compartments.
In most body compartments, accumulation kinetics were described using saturation models.
In seawater-exposed individuals, the highest saturation concentrations were reached in the
gonads, rectal caeca, and oral body wall, but the compartment reaching saturation at the
fastest rate was the pyloric caeca. In sediment-exposed sea stars, rectal caeca displayed the
highest saturation concentration, but the central digestive system had the highest PCB
90

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
accumulation rate. Food-exposed animals displayed the highest saturation concentrations in
the rectal caeca and aboral body wall, and the different body compartments showed similar
accumulation rates. Considering the different accumulation routes, the highest uptake rates
were found in seawater-exposed sea stars, followed by sediments and food, attesting to the
importance of seawater as a PCB exposure pathway.
Particular concern arises from the PCB behaviour in the gonads of the sea star. Indeed,
relatively high concentrations were incorporated in these organs according to linear uptake
kinetics, suggesting that a steady state in this compartment would take a very long time to be
reached under natural conditions. This may have an important impact on sea star population
survival, because of the endocrine disrupting effect of coplanar PCBs (den Besten et al. 1989,
Suedel et al. 1997, Chiu et al. 2000).
In parallel to bioaccumulation kinetics, two different sublethal biological effects were
measured: the ROS production by amoebocytes and CYP1A immunopositive protein
induction in pyloric caeca. Significant stimulation of ROS production by amoebocytes was
observed in PCB 77-exposed sea stars regardless of the exposure pathway. In addition, ROS
production stimulation was observed in bacteria-stimulated amoebocytes as well as in non-
stimulated cells. This is in contrast with the observations reported by Coteur et al. (2001) who
did not observe any significant ROS production by non-stimulated coelomocytes of PCB 77-
exposed sea urchins, Paracentrotus lividus. This is most probably due to differences in
composition of the coelomocyte population between sea urchins and sea stars. Indeed,
amoebocytes which are responsible for ROS production are the sole, free-circulating
coelomocyte type in A. rubens, whereas there are six different coelomocyte types co-existing
in sea urchins (Chia & Xing 1996). As ROS production is measured on a «per cell» basis, the
difference in coelomocyte population composition could result in an underestimated response
when ROS production is measured in sea urchins compared to the same response in
A.rubens.
Surprisingly, after a strong stimulation during the first 4 to 10 days, ROS production fell to
control levels after a variable period of time. This observation could be due to a toxic effect
on amoebocytes occurring when PCB concentrations reach a certain level within the
organism’s tissues. This toxicity could impair or inhibit the ROS production capacity of the
amoebocytes. Indeed, it has been shown in both sea stars and sea urchins that when coplanar
PCBs are injected into the coelomic cavity, ROS production is stimulated proportionally to
the injected dose up to a certain injected concentration at which it drops to ROS production
control levels (Coteur et al. 2001, Danis et al. Chap. III.4). However, this hypothesis may not
91

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
be sufficiently valid in order to explain the observations reported here, since PCB
concentrations incorporated in sea stars were very similar during the first days of exposure
regardless of the exposure pathway (seawater, food or sediments), whereas the time at which
ROS production began to decrease varied between days 4 (seawater exposure) and 9
(sediment exposure).
Alternatively, the decrease in ROS production could be due to the progressive activation of an
efficient detoxification mechanism (e.g. components of the cytochrome P450 system,
CYP1A) that would limit and then eliminate the impact of PCB 77 on amoebocyte functions,
and thus on the stimulation of these immune cells to produce ROS. Induction of P450
enzymatic activity has been reported in echinoderms exposed to coplanar PCB congeners
(den Besten et al. 1993, Danis et al. Chap. III.4). In the present study, very clear and
significant induction of a CYP1A immunopositive protein (CYP1A IPP) was measured in the
pyloric caeca of PCB 77-exposed sea stars, regardless of the exposure pathway. CYP1A IPP
induction increased with time according to a saturation model; however, the maximum
induction was relatively low (approx. two-fold).
Most noteworthy, from Figure 9 it is evident that at the time corresponding to the ROS
production maximum in sea stars exposed via seawater or their food (i.e. day 5), the CYP1A
IPP induction factor is identical (1.6) for both groups of sea stars. In contrast, this factor is
lower for individuals exposed to PCB 77 from sediments. Furthermore, the time (9 days)
when the CYP1A induction factor reaches the same value (1.6) in sediment-exposed sea stars
matches perfectly the time at which ROS production reaches its maximum value in this group
of sea stars. This suggests that CYP1A enzymatic activity reaches an effective efficiency at an
induction factor of ca. 1.6, above which the toxic action of PCB 77 on the immune system
would be efficiently limited, resulting in a ROS production at more normal levels. This is of
prime importance from a biological point of view, since immunomodulation induced by
coplanar PCBs can lead to dysfunctions in defence against infections (see, e.g., Livingstone et
al. 2000).
In conclusion, Asterias rubens efficiently took up the coplanar PCB congener #77 which was
distributed in all the tissues examined. Consequent sublethal biological effects were
pronounced and affected essential sea star physiological functions at environmentally realistic
levels of PCB. Indeed, the present work showed that both the immune and P450 enzymatic
systems were impacted. In this sea star species, exposed under the experimental conditions
considered here, the immune effects appeared to be reversible. The ROS production
parameter could thus be a very interesting short-term biomarker of the presence of coplanar
92

Coplanar PCB uptake and subsequent effects in the sea star Asterias rubens
PCBs, and probably of other dioxin-like organic contaminants as well. Furthermore, its use in
conjunction with the CYP1A-induction measurement would allow gathering highly
informative data about timescales of exposure of animals, since this parameter has different
stimulation kinetics than ROS production; i.e. the CYP1A signal is still observable in
organisms when ROS production has returned to control levels.
Nowadays the recommendations related to PCB monitoring in the environment that are
adopted by international organisations such as the EU or ICES, address a limited number of
PCB congeners, viz. #28, 52, 101, 118, 138, 153 and 180. These congeners are well-known to
be the most relevant regarding PCB abundance in biota and the environment (e.g. Metcalfe
1994, OSPAR 2000), but not at all regarding their toxicity to organisms and as a threat to
marine ecosystems. In contrast, an increasing data set for a variety of organisms demonstrates
that coplanar congeners, even if occurring at very low concentrations, are the key ones with
respect to PCB toxicity (e.g. Wilbrink et al. 1991, Michel et al. 1993, Schweitzer et al. 1997,
den Besten 1998a, Coteur et al. 2001, Duffy et al. 2002, Danis et al. Chap. III.1, III.3, III.4).
These observations underscore the urgent need to provide further information in natural
marine environments about coplanar-specific biological effects, especially in the most
commonly used bioindicator species (i.e. bivalves), in order to assess whether coplanar PCBs
should be included in the list of congeners to be systematically monitored.
ACKNOWLEDGEMENTS
The IAEA Marine Environment Laboratory operates under a bipartite agreement between the
International Atomic Energy Agency and the Government of the Principality of Monaco. B.D.
is holder of a FRIA doctoral grant, and M.W. is an Honorary Research Associate of the
National Fund for Scientific Research (NFSR, Belgium). The research was partly supported
by a travel grant from the Belgian Ministry of the French Community (B.D.’s travel to and
stay at IAEA-MEL) and by a NFSR equipment funding (Tecan Spectrafluor Plus plate-
reader).
93

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
III.4 Contrasting effects of coplanar vs non-coplanar PCB congeners on
immunomodulation and CYP1A levels (determined using an adapted ELISA
method) in the common sea star Asterias rubens L.
Aquatic Toxicology (submitted)
Danis B a , Goriely S b , Dubois Ph a , Fowler SW c , Flamand V b & Warnau M c
a. Laboratoire de Biologie marine (CP 160/15), Université Libre de Bruxelles, Av. F.D.
Roosevelt 50, B-1050 Brussels, Belgium.
b. Laboratoire d’immunologie expérimentale (IMMEX, CP615), Université Libre de
Bruxelles, Campus hospitalo-universitaire d'Anderlecht, route de Lennik 808, B-1070
Brussels, Belgium.
c. International Atomic Energy Agency - Marine Environment Laboratory, 4 Quai Antoine
I er , MC-98000 Monaco.
95

ABSTRACT
Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
Biological effects of two structurally-contrasting PCB congeners (coplanar 77 and non
coplanar 153) were investigated by measuring the induction of CYP1A immunopositive
protein (CYP1A IPP) in the pyloric caeca and the production of reactive oxygen species
(ROS) by amoebocytes in the common sea star Asterias rubens. CYP1A IPP was quantified
using a specially designed ELISA which uses competitive binding between sea stars and trout
CYP1A IPPs. Only the coplanar congener had a significant effect on the two considered
biological responses. Intensity of the effects was dose-dependent. However, the highest dose
of PCB 77 induced a dramatic decrease of ROS production. It is concluded that coplanar
PCBs straightforwardly affect key biological processes such as the immune system and
mixed-function oxidase (MFO) system.
KEYWORDS
CYP1A; ELISA; PCBs; echinoderms; ROS; immune system; Asterias rubens
96

INTRODUCTION
Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
Biological effect monitoring is essential in determining the spatial and temporal distribution
of risks caused by contaminants in the marine environment (Huggett 1992, Stegeman et al.
1992). In the marine environment, biomarkers can offer early warning signals of exposure to
contaminants and may be used to diagnose and interpret observed effects, for example by
testing sublethal toxicity in water or sediment samples in laboratory bioassays using selected
test species (den Besten 1998a). In this study, biological effects of PCBs were investigated in
the sea star Asterias rubens, a widely distributed and abundant key benthic species (sensu
Lewis 1978) in the North Sea and NE Atlantic, which has been used as bioindicator for
monitoring contamination in the field (e.g., den Besten et al. 1993, 2001, Everaarts et al.
1998, Temara et al. 1998b, Coteur et al. 2003a, Stronkhorst et al. 2003). Two biological
responses were specifically addressed: reactive oxygen species (ROS) production by
amoebocytes and cytochrome P450 immunopositive protein (CYP1A IPP) induction in the
pyloric caeca.
ROS production is one of the main immune responses of echinoderms and plays a key role in
the destruction of microorganisms through cytotoxic mechanisms (Chia & Xing 1996). This
immune response is triggered by amoebocytes which are the most active, free-circulating cells
found in echinoderm coelomic cavities. Its amplitude was reported to be modulated by
contaminants in A. rubens (Coteur et al. 2003a,b).
CYP1A induction is probably one of the most commonly studied biochemical markers in
environmental monitoring (e.g., Bucheli & Fent 1995, Stegeman 1995, Hahn 2002). The main
function of the CYP1A-dependent monooxygenase system is to convert relatively insoluble
organic compounds to soluble metabolites; however, the resulting products are often more
toxic than the parent compounds (Walker & Peterson 1994). Total cytochrome P450
concentrations were reported in A. rubens using the carbon monooxide (CO) difference
spectra of sodium dithionite-reduced samples (den Besten et al. 1991 1993). During the last
decade, a shift from analyzing total cytochrome P450 towards measuring amounts of dioxin-
inducible isoenzyme of CYP1A has taken place (Bucheli & Fent 1995). More specifically, in
the analysis of fish cytochrome P450, isoenzyme content has been measured using Enzyme-
Linked Immuno Sorbent Assays (ELISAs) (Celander & Förlin 1991, Goksøyr 1991, Förlin et
al. 1992). However, this method has never been used before for such measurements in
echinoderms. Because of the high conservation of CYP1A sequence throughout the animal
kingdom, the ability of antipeptide antibodies (developed with a high affinity for trout
CYP1A) to cross react with sea star cytochrome P450-like proteins is worthwhile testing in
97

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
order to optimize CYP1A IPP measurement in these invertebrates.
In this study, we set up, test and apply a customized ELISA for the determination of CYP1A
IPP in sea stars. Intracoelomic injections were used as an exposure route in order to test the
toxicity of PCBs. The doses considered were chosen to include the higher range of those
reported in previous bioaccumulation experiments (Danis et al. Chap. III.2). The effect of two
structurally-contrasting PCB congeners (viz. the coplanar congener 77 and the non-coplanar
congener 153) on the immune system and on CYP1A IPP induction in the sea star A. rubens
was then determined.
MATERIALS AND METHODS
Organisms
The sea stars Asterias rubens (Linnaeus 1758) were collected in April 2002 from the intertidal
zone at Audresselles (Pas-de-Calais, France). Prior to experimentation, 50 specimens of
similar size (5-7 cm arm radius) and weight (36 ± 3.5 g) were acclimated to laboratory
conditions for 1 month (constantly aerated closed circuit aquaria; salinity 34 p.s.u.; 16 ± 0.5
°C; 12/12 hrs dark/light cycle).
Chemicals and solutions
Stock powders of 2,2’,4,4’,5,5’ hexachlorobiphenyl (non-coplanar PCB 153), 3,3',4,4',5
pentachlorobiphenyl (coplanar PCB 126) and 3,3',4,4' tetrachlorobiphenyl (coplanar PCB 77)
purchased from Promochem (Germany) were dissolved in ultrapure acetone (Sigma) to reach
a concentration of 1 µg ml -1 . Stock solutions were stored at –20°C until final dilutions were
prepared. A few minutes before injecting the animals, successive dilutions were made in glass
haemolysis tubes using ultrapure acetone as a solvent.
The polyclonal rabbit anti-trout CYP1A antibody (0.4 mg ml -1 ) and the ß-naphtoflavone
(BNF)-induced trout microsomal fraction were purchased from Biosense (Norway).
Phosphate-buffered saline (PBS), bovine serum albumin (BSA), horseradish peroxidase
(HRP) coupled to Extravidin and Coomassie blue were obtained from Sigma (USA). The
substrate chromogen (tetramethylbenzidine, TMB) was purchased from Biosource (UK).
Preliminary PCB exposure
In order to setup and test the ELISA, four batches of sea stars (n = 3) were injected into the
coelomic cavity with 50 µl of either the coplanar PCB 126 (6 ng g -1 whole body FW), the non
coplanar PCB 153 (15.2 ng g -1 whole body FW), seawater (control 1), or acetone (control 2),
using a 50 µl Hamilton glass syringe. Animals were sampled 48 h after being injected and
98

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
pyloric caeca were dissected and immediately frozen in liquid nitrogen (-196°C). Samples
were then stored at –80°C.
PCB exposure
For the toxicity test, five batches of 3 sea stars were injected with 50 µl of PCB solutions (at 5
different concentrations) using a 50 µl Hamilton glass syringe. In addition, a control group
(n=5) was injected with acetone alone, and another one (n = 5) with seawater; a last group
of blank animals (n = 5) was not injected. Injected doses were 15.2, 2.82, 0.56, 0.15 and 0.02
ng PCB 153 g -1 whole body (FW) and 5.96, 1.51, 0.18, 0.04 and 0.008 ng PCB 77 g -1 whole
body (FW). Animals were sampled 48 h after being injected and pyloric caeca were dissected,
frozen in liquid nitrogen and stored at -80°C until making biomarker measurements.
Protein normalisation
Frozen pyloric caeca were homogenized in PBS buffer using an ultraturrax (22,500 rpm; 20 s)
and the homogenate was centrifuged (12,500 g; 4°C; 15 min). The supernatant (post-
mitochondrial supernatant, PMS) was transferred to new centrifugation tubes and re-
centrifuged using the same protocol. In order to quantify the total protein content, a sub-
sample of the PMS was transferred to 96-well plates and mixed with Coomassie blue
(Bradford 1976). Absorbance at 620 nm was measured using a Tecan Spectrafluor Plus plate-
reader. Total protein concentration of the samples was then normalised to 100 µg ml -1 .
Protein biotinylation
Protein biotinylation of the trout microsomal extract was performed using Pierce Endogen
(UK) biotinylation kit. Sulfo-NHS-Biotin was added to protein solutions to give a 20-fold
molar excess of Sulfo-NHS-Biotin in a 10 mg ml -1 protein solution. After incubating at room
temperature for 30 min, a desalting column was equilibrated with 30 ml PBS. After
application of the trout microsome solution, an aliquot of buffer was applied and fractions
containing variable protein concentrations were collected in separate test tubes. Fractions
containing the highest protein concentration (as determined using the Coomassie blue assay,
Bradford 1976) were then pooled together. The biotin incorporation efficiency in this pool
was tested using the HABA method (Green 1965).
Western Blot
Western blot was performed to test the cross-reactivity of the anti-trout CYP1A antibody with
sea star homologue proteins. After being cooled at 4°C, 0.1% bromophenol blue was added to
the samples. Protein samples were combined with equal parts of treatment buffer (0.125 M
Tris(hydroxymethyl)aminomethane (Tris)-Cl, 2% Sodium dodecyl sulphate (SDS), 10%
glycerol, 5% 2-mercaptoethanol) and heated in a water bath (60-70°C) for 2 min. SDS
99

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
polyacrylamide gel electrophoresis (SDS-page, 10% polyacrylamide) was used for this assay.
After electrophoretic migration (200 V, constant voltage, 2 hrs), proteins were transferred to
nitrocellulose sheets using a Genie ® system (BioRad). Transfer papers were rinsed and
blocked for 30 min using a blocking buffer (10 -2 M Tris-SDS). Anti-trout CYP1A antibody
was added to a fresh change of blocking buffer at a 1:1000 dilution, and left overnight at 4°C.
After thorough rinsing, transfer papers were placed in blocking buffer (Tris-
ethyldiaminetetraacetic acid (EDTA)-BSA) containing anti-rabbit IgG-alkaline phosphatase
(1:1000 dilution). The papers were then washed, rinsed and transferred to 100 mM Tris-HCl
with 1 mM MgCl 2. Enzyme-substrate chromophore cocktail (0.15 M bicarbonate/carbonate-
BiCP-nitroblue tetrazolium) was finally added.
Enzyme-linked immuno sorbent assay (ELISA)
Cytochrome P450 IPP content was quantified using competitive-ELISA (Fig. 27). The ELISA
was carried out using competitive binding to an anti-trout CYP1A antibody between the
CYP1A IPP contained in the sea star samples and the biotinylated CYP1A from BNF-injected
trout. Because the concentration of biotinylated trout CYP1A remains constant while the
concentration of sea star CYP1A IPP varies between samples, the amount of bound
biotinylated trout CYP1A is inversely proportional to the concentration of CYP1A IPP in sea
star samples.
Anti-CYP
Plate
Starfish CYP
Figure 27. Schematic representation of CYP1A IPP competitive ELISA method.
In order to determine the optimal quantity of biotinylated trout CYP1A to be used in our
setting, a sigmoid-shaped fixation curve was resolved experimentally. The optimal
biotinylated trout CYP1A quantity to be used was chosen in the linear portion of the fixation
100
Biotin
Avidin-HRP
Biotinylated
Trout CYP

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
curve, corresponding to a concentration of 1mgml -1 total protein. For the ELISA, multiwell
plates (96 wells) were coated at 4°C with 100 µl of anti-CYP1A (rabbit anti-fish CYP1A
peptide, polyclonal antibody; dilution 1:1000 in PBS-BSA 0.1%). After 12 hrs, wells were
washed 5 times with PBS and nonspecific binding sites were blocked with 200 µl PBS-BSA
2% agitated for 2 hrs at room temperature. Wells were washed again, and 100 µl of
biotinylated trout were added along with protein-normalised samples. This operation was
performed as fast as possible, using a template multiwell plate (with low sorption
characteristics) containing the samples. Competition between the two antigens was allowed to
take place for 2 hrs at room temperature on an agitation plate. At this stage, five washing
steps with PBS were performed and Extravidin-HRP (Sigma) in PBS-BSA 0.1% (dilution
1:2000) was added to all the wells. The plate was then incubated at room temperature for 45
min and the wells washed again using PBS. TMB chromogen cocktail (dilution 1:105 in
substrate buffer; Biosource, UK) was added to all the wells and the plate was left in the dark
for approximately 10 min at room temperature. The reaction was stopped by adding 50 µl of
95-97% sulfuric acid (Sigma) to the wells. Absorbance was measured in the 96-well plates at
450 nm using a microplate reader (Spectracount, Packard), assuming that absorbance is
proportional to the amount of biotinylated trout CYP1A bound to the antibody.
Reactive Oxygen Species (ROS) production measurements
Amoebocyte ROS production was measured by the peroxidase, luminol-enhanced method
optimized by Coteur et al. (2002). Briefly, 3 ml of coelomic fluid were collected in the same
volume of anticoagulant buffer. The cell concentration was measured by absorbance at 280
nm using a Tecan Spectrafluor Plus plate-reader. This suspension was then centrifuged for 10
min at 400 g and resuspended in Ca 2+ -, Mg 2+ -free artificial seawater (ASW). ASW volume
was adjusted to obtain a final amoebocyte concentration of 10 6 cells ml -1 . A stock solution of
luminol and HRP in dimethylsulphoxide (DMSO) was freshly diluted 100 fold in ASW (final
concentrations of HRP and luminol were 500 µg ml -1 and 250 µg ml -1 , respectively). The
reaction was begun by adding 200 µl of amoebocyte suspension in 100 µl of luminol/HRP
solution and 20 µl of a Micrococcus luteus suspension (2.5x10 9 bacteria ml -1 ) (stimulated
amoebocytes) or 20 µl of ASW (non-stimulated amoebocytes). The chemiluminescence was
measured every 10 min over a 2 hr period using a Tecan Spectrafluor Plus plate-reader
previously placed in an thermostabilized incubator (14 ± 0.5°C). Results were expressed as
the sum of all 10 min interval measurements for 10 6 cells ml -1 (total chemiluminescence) of
bacteria-stimulated amoebocytes or non-stimulated amoebocytes.
101

Data analysis and statistics
Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
Experiments were performed with at least three biological replicates. Data was expressed as
arithmetic means ± standard deviation (SD). CYP1A induction was determined using dilution
curves performed with highly responsive samples (PCB 126-injected sea star). Dilution
curves were best modelled using exponential equations (Equation 7). These data were then
expressed as CYP1A inhibition percentages (Equation 8). Finally, inhibition percentages were
converted to CYP1A inhibition percentages measured in control (acetone-injected) sea stars
(Equation 9) to obtain CYP1A induction folds in sea star samples.
Equation 7: y = a . e b.CYPi
Equation 8: CYP i = 100 – (100 . A/A max)
Equation 9: CYP I = (a . e b.CYPi )/(CYP ic)
where CYP i is CYP1A inhibition percentage, A is absorbance at 450 nm in wells, A max is
maximum absorbance in the absence of competitor (sea star samples), CYP I is the measure of
CYP1A induction (induction folds), a and b are the parameters of the exponential equation
fitting the dilution curve, and CYP ic is the inhibition percentage in control sea stars (viz.
acetone-injected animals).
Dose-response relationships were tested by linear and non-linear regressions, using the Systat
5.2.1 software (Wilkinson 1988). Differences between ROS production and P450 levels in the
different exposure conditions were tested for significance using 1-way ANOVA followed by
the multiple comparison test of Tukey (Zar 1996). The level of significance was always set at
a =0.05.
RESULTS
ELISA setup
In the setup phase of the novel ELISA for measuring CYP1A IPP in sea stars, a first step was
to carry out western blot immunoassays to check the cross-reactivity of anti-trout CYP1A
antibody with sea star samples. For this purpose, post-mitochondrial supernatant (PMS)
extracts from sea stars injected with PCB 126, PCB 153, seawater or acetone were run on
gels. One of these gel (PCB 126- and seawater-injected sea stars) is shown in Fig. 28.
102

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
Figure 28. Western blot SDS-PAGE gel. Lanes are: (A) molecular
weight standards (kD), (B) BNF-injected trout microsomes, (C) PCB
126-injected sea star PMS extract and (D) seawater-injected sea star
PMS extract
BNF-induced trout microsomes were used as positive controls (lane B) and molecular weight
standards were also run in the wells (lane A). Results show that a band appears in the case of
PCB 126-exposed sea stars (lane C). The sea star band is in the range of CYP1A molecular
weights reported in other species (between 45 and 60 kD; Lewis 1996) and was induced only
in individuals exposed to the coplanar PCB 126: control sea stars (lane D) showed no band.
Therefore, the antibody was used for further testing and ELISA procedures.
The competitive binding-based ELISA was carried out on PMS extracts of sea star pyloric
caeca using biotinylated BNF-induced trout microsomes as a competitor. The linear zone of
the fixation curve was between 1 and 10 mg total protein ml -1 , justifying the use of a 1 mg
total proteins ml -1 solution for the competitor (being in the linear zone and representing the
lesser quantity to be used for the assay). The test for specificity (values of a representative
run) are presented in Table 15.
103

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
Table 15. Optical densities measured in a representative ELISA (mean ± SD). Ab = rabbit antitrout CYP1A
antibody; SW = seawater injected sea stars (100 µg ml -1 ); AC = acetone injected sea stars (100 µg ml -1 ); Tµ =
trout microsomes (100 µg ml -1 ); B * Tµ = biotinylated trout microsomes; Av-HRP = avidine-coupled horse
radish peroxidase; O.D. = optical density (mean ± SD, biological triplicates); - = no sample/treatment
Ab Sample/treatment B * Tµ Av-HRP O.D. n
+ - - + 0.06
+ - + + 0.93
+ SW + + 0.75 ± 0.05 5
+ AC + + 0.71 ± 0.13 5
+ PCB 153 + + 0.65 ± 0.09 15
+ PCB 126 + + 0.26 ± 0.05 15
+ Tµ + + 0.16 ± 0.05 5
The optical density (OD) measurement in blank wells was 0.06. Maximum absorbance,
measured in the absence of sample, produced an OD of 0.93. Weak competition occurred with
seawater and acetone-injected sea star PMS extract (OD = 0.75 and 0.71, respectively). PCB
153 exposure induced a competition of medium intensity (OD = 0.65) while samples from
PCB 126-exposed sea stars induced significant competition for the anti-CYP1A antibody (OD
= 0.26). The most intense competition was found for BNF-injected trout microsomes (OD =
0.16). As mentioned in the methods section, the dilution curve was used to calculate CYP1A
IPP induction folds (as compared to controls) in the different samples using equation 3.
Seawater and acetone controls showed CYP induction factors of 1.01 ± 0.24 and 1.92 ± 0.84,
respectively, while exposure to the non coplanar PCB 153 resulted in a CYP induction factor
of 2.67 ± 0.75. Significant variation of CYP1A IPP content was observed only in animals
injected with the dioxin-like, coplanar congener 126 (CYP induction factor: 67 ± 8.60) (Fig.
29).
CYP1A IPP induction (time fold)
100
90
80
70
60
50
40
30
20
10
0
PCB 153 PCB 126
Figure 29 CYP1A IPP induction (induction fold, mean ± SD, n = 3) in
sea stars injected with coplanar (126) or non-coplanar (153) PCB.
104

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
Reproducibility among different experiments was good when results were converted to
induction factors, with a 73-fold mean increase in CYP1A IPP levels. The sensitivity of the
test, defined as the lowest protein concentration at which a significant variation of CYP1A
IPP can be detected, is 1.005-fold (5 ‰) in the case of PCB 126-exposed sea stars (calculated
as 3.SD of blank measurements).
Sea stars were injected with increasing doses of structurally contrasting congeners (PCB 153
vs 77). CYP1A IPP induction was measured in the different groups of individuals. Control
groups (acetone-injected, seawater-injected and non-injected) displayed no significant
differences (p = 0.36) among each other. Thus, all control individuals were considered as a
single control group for the purpose of comparison.
For PCB-exposed sea stars, the results showed a contrasting response between individuals
exposed to the coplanar vs the non-coplanar congener (Fig. 30). PCB 153 did not significantly
induce CYP1A IPP compared to controls, whereas PCB 77 induced a steep increase of
CYP1A IPP content which displayed a significant, linear dose-response relationship
(R 2 =0.87, p = 0.01).
CYP1A IPP induction (ratio experimental:control levels)
50
40
30
20
10
PCB 77 PCB 153
105
y = 5.68x + 4.21
R 2 = 0.87
p = 0.01
0
0.001 0.01 0.1 1 10 100
Injected dose (ng g -1 FW)
Figure 30. CYP1A IPP induction (ratio of experimental response to
control group) as a function of injected PCB dose (ng g -1 FW; log scale)
for two structurally contrasting congeners. Curve fitting: linear
regression of CYP1A induction as a function of injected dose.
Effects of PCB congeners on ROS production
ROS production was measured in the different control groups (non-injected, seawater-injected
or acetone-injected sea stars) and no significant differences were found among these three
groups (p = 0.59 and 0.16 for non-stimulated and bacteria-stimulated amoebocytes,
respectively). Therefore, all control individuals were pooled together as a single global
control group for further comparisons.

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
ROS production by non-stimulated amoebocytes is presented in Fig. 5. While PCB 153
induced some increase in ROS production in a few individuals compared to controls, no
statistically significant relationship was observed between ROS production and injected PCB
153 dose. In contrast, coplanar congener PCB 77 induced a significant increase in ROS
production, i.e. up to 20 times the control values in individuals exposed to a dose of 1.51 ng g -
1 FW. At the highest PCB 77 dose (5.96 ng g -1 FW), the level of ROS production dramatically
dropped to control values (baseline in Fig. 31).
ROS production (RLU/10 6 cells)
50000
40000
30000
20000
10000
PCB 77 PCB 153
y = 18800x + 9700
R 2 = 0.86
p = 0.001
0
Baseline
2950±1350
0.001 0.01 0.1 1 10 100
Injected dose (ng g -1 FW)
Figure 31. ROS production (total chemiluminescence) by nonstimulated
amoebocytes as a function of injected PCB dose (ng g -1 FW;
log scale) for two structurally contrasting congeners. Curve fitting:
linear regression of ROS production as a function of injected dose
(excluding data of the highest dose); Baseline: ROS production value in
control group (n = 15).
The same behaviour was observed for bacteria-stimulated amoebocytes exposed to PCB 77
(Fig. 32). Considering only the results at doses lower than the highest one, a very clear, linear
dose-response relationship was found between ROS production and the dose of PCB 77
(R 2 =0.86, p = 0.001 and R 2 = 0.95, p = 0.01 for non-stimulated and bacteria-stimulated
amoebocytes, respectively).
106

ROS production (RLU/10 6 cells)
250000
200000
150000
100000
50000
Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
PCB 77 PCB 153
y = 125000x + 27900
R 2 = 0.95
p = 0.01
0
Baseline
14000 ± 10800
0.001 0.01 0.1 1 10 100
Injected dose (ng g -1 FW)
Figure 32. ROS production (total chemiluminescence) by bacteriastimulated
amoebocytes as a function of injected PCB dose (ng g -1 FW;
log scale) for two structurally contrasting congeners. Curve fitting:
linear regression of ROS production as a function of injected dose
(excluding data of the highest dose); Baseline: ROS production value in
control group (n = 15).
In the case of PCB 153, no significant differences were found for ROS production by
stimulated or non-stimulated amoebocytes in control vs exposed sea stars.
Finally, in order to check for a relationship between the responses of the two biological
effects (viz. ROS production and CYP1A IPP induction) in PCB 77-exposed sea stars, both
factors were plotted as a linear function of either non-stimulated or bacteria-stimulated
amoebocytes (Fig. 33). Very strong determination coefficients were found for linear
regressions in both cases, i.e. R 2 = 0.95 , p = 0.03 and R 2 = 0.96 , p = 0.02 respectively. The
strongest relationship was obtained between ROS production and CYP1A induction in the
presence of bacteria-stimulated coelomocytes.
107

DISCUSSION
ROS production (RLU/10 6 cells)
250000
200000
150000
100000
50000
Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
Non-stimulated amoebocytes Bacteria-stimulated amoebocytes
108
y = 14100x - 14200
y = 2200x + 2700
0
0 5 10 15 20
CYP1A IPP induction (ratio experimental:control levels)
r 2 = 0.95
p = 0.03
r 2 = 0.96
p = 0.02
Figure 33. Regressions between ROS production (total chemiluminescence)
and CYP1A IPP induction (ratio of experimental response to control group)
in the case of non-stimulated and bacteria-stimulated amoebocytes from
PCB 77-injected sea stars. Curve fitting: linear regression of CYP1A IPP
induction as a function of ROS production; R 2 : determination coefficient.
In this paper, a novel, rapid and sensitive competitive ELISA for assessing CYP1A IPPs in
sea stars is described. This assay does not require long ultracentrifugation steps, or cell
culture maintenance and is less time-consuming than other widespread methods such as
Western blot, or the CO-difference spectrum for determining total P450 contents. The only
pre-requisite of the method is to determine the optimal quantity of biotinylated competitor to
be used in the test. Another advantage of the method is the need of only minute quantities of
tissues for analysis, as the relationship between CYP1A levels and sample dilution remained
linear over a wide range of concentrations. The number of samples that can be processed
simultaneously is also greatly enhanced, allowing fast and efficient screening.
Purified sea star CYP1A protein is not available commercially and a true calibration curve
could not be established. Therefore, for further testing, it is recommended to establish a
dilution curve using PMS extracts of individuals exposed to a CYP1A-inducing compound,
such as coplanar PCBs, dioxins, furans or polyaromatic hydrocarbons. Obviously, using this
curve implies comparing treated samples to control individuals. Nevertheless, overall, the
method is very satisfactory for fast screening of CYP1A IPP variations. In addition, the high
degree of inductibility of the CYP1A (up to 73-fold mean increase) is encouraging with
respect to applying this method to samples coming from low-contrasting field situations. The
exact affinity of the antibody used in this study for sea star CYP1A could not be determined,
since no pure sea star CYP1A is available. Also, absolute values of absorbance at 450 nm

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
varied among the different experiments, but this observation is generally considered as typical
of ELISA testing (Kennedy 1991).
The induction of CYP1A is one of the most frequently used biomarkers in the field of
environmental contamination (Bucheli & Fent 1995). Quantification of CYP1A protein levels
in field and laboratory samples can provide important and complementary information about
the regulation of this protein and toxicity mechanisms in general. Fast and simple methods are
required for this purpose and the ELISA described here seems to be a very promising tool.
Provided that the cross-reactivity of the antibody and that the responsiveness to the compound
is tested, the present method can probably be applied to a certain range of marine invertebrate
species.
In the present study, the adapted ELISA method allowed the observation of PCB congener-
dependent behaviour for CYP1A IPP induction in sea stars. Exposure of congener 153 did not
result in any significant induction of CYP1A IPP whereas coplanar PCB 77 induced a sharp
increase in this protein content. In a previous study, den Besten et al. (1993) measured total
cytochrome P450 in sea stars that received single injections of different PCB congeners
(PCBs 118, 126 and 153). In this work, total cytochrome P450 was determined using the
carbon monoxide difference spectrum of dithionite-reduced samples. Morever, PCB 126 was
used at doses ranging up to 2 µmol kg -1 (approx. 650 ng g -1 ), which is 2 orders of magnitude
greater than the highest dose used in our experiments. No dose-dependent effects of PCBs
were found on the cytochrome P450 content by den Besten et al. (1993). The discripancy
between the latter results and those obtained in the present study can be explained by the
lower specifity of measuring total cytochrome P450 compared to measuring only CYP1A IPP
isoforms. Other workers found results that were similar to ours by exposing cell cultures of
trout hepatocytes to different PCB congeners: only coplanar congeners 77 and 126 induced
significant CYP1A responses (Bruschweiler et al. 1996). As suggested by Coteur et al.
(2001), similarities found in responsiveness to dioxin-like compounds measured in sea stars
and vertebrates could indicate the existence of a similar detoxification mechanism in
echinoderms and vertebrates. The intensity of CYP1A IPP induction measured in this study
was comparable to that described in human cells (Jones & Anderson, 1999) and in mussels
(Livingstone et al. 2000). The occurrence of a CYP1A-like protein that can be induced by
PAHs and PCBs in molluscs has been shown in various studies (Livingstone & Goldfarb
1998, Peters et al. 1998), and its induction in mussels has been shown experimentally (Michel
et al. 1993). However, there is still a lot of discussion about the mechanisms underlying this
induction in invertebrates (see Hahn 2002a for a review). In vertebrates, CYP1A induction is
109

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
known to be linked to the exposure to dioxin-like compounds. Mechanisms underlying this
induction have been investigated and many studies have drawn the Aryl hydrocarbon
Receptor (AhR) in the production of dioxin-like toxicity (e.g., Fernandez-Salguerro et al.
1996, Schmidt et al. 1996). Although the occurrence of the AhR has not been demonstrated in
echinoderms, our data and those of Coteur et al. (2001) strongly suggest that a vertebrate-like,
CYP1A-triggering mechanism may actually occur in echinoderms. The strongest arguments
for this occurrence are: (i) the conservative evolution of the AhR sequence (Lewis 1996), (ii)
a similar structure-induction relationship, and (iii) the dose-response relationship found with
cPCBs. Nevertheless, further research is needed to demonstrate the existence of the AhR in
echinoderms, by addressing e.g. the induction of echinoderm CYP1A by vertebrate AhR
ligands, the relative structure-activity relationship (SAR) of the potential inducers, and finally
2,3,7,8-tetrachloro-dibenzo-p-dioxin (TCDD) interaction with the receptor leading to its
potency to induce CYP1A expression.
The effect of the two structurally contrasting PCBs on the immune system activity was
assessed by measuring ROS production in sea stars. In these organisms, ROS production by
amoebocytes represents the main defence line against non-self material, but the efficiency of
this process has been shown to be modulated by xenobiotic exposure (Anderson et al. 1997,
Coteur et al. 2001). For example, immunomodulation induced by coplanar PCBs can lead to
altered defence against infections (Livingstone et al. 2000). One of the key actors in ROS
production is the activation of a membrane-associated enzyme, the NADPH-oxidase, which
reduces molecular oxygen to superoxide anion ( • -
O2 ), which leads to the production of other
oxidants (Babior et al. 1984). In our study, non-coplanar PCB 153 did not induce significant
variation in ROS production whichever dose was applied (except at one concentration). In
contrast, coplanar PCB 77 induced a significant increase in ROS production at all doses tested
except the highest one, where ROS production dramatically dropped to control levels. This
type of response has also been observed in sea urchins (Coteur et al. 2001). Sea urchins
injected with increasing doses of selected PCBs (congeners 77, 126 and 153) showed no
significant effect in ROS production with PCB 153, whereas PCB 77 and PCB 126 caused a
large increase in ROS production. As in the present experiment, ROS production in sea
urchins also dropped at the highest PCB concentration (12.5 ng l -1 ) tested by Coteur et al.
(2001). In our study, induction of ROS production was observed in bacteria-stimulated
amoebocytes as well as in non-stimulated cells, an observation which is in contrast with that
reported by Coteur et al. (2001) in the case of non-stimulated amoebocytes. This is most
probably due to differences in composition of the coelomocyte population between sea
110

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
urchins and sea stars. While amoebocytes (which are responsible for ROS production) are the
sole free-circulating coelomocyte type in A. rubens, there are at least six different
morphological and functional coelomocyte types co-existing in sea urchins (Chia & Xing
1996). As ROS production is measured on a «per 10 6 cells ml -1 » basis, this difference in
coelomocyte population composition could result in an underestimated response when ROS
production is measured in sea urchins compared to the same response in A. rubens.
In the present study, ROS production increased with injected cPCB doses, which indicates
immunomodulation by this xenobiotic, but could also be due partly to the production of ROS
as by-products of other mechanisms related to cPCB exposure, viz. detoxification
mechanisms. For instance, PCB metabolites produced by CYP1A (quinones, hydroquinones)
can lead to the formation of superoxide anion radicals (Aryoshi et al. 1993). Therefore,
CYP1A induction can trigger an oxidative stress, which is also a major signal in apoptosis
processes (Zahner et al. 1998, White & Prives 1999). In our study, the possible occurrence of
ROS sources other than those produced in amoebocytes could be supported by the fact that
CYP1A IPP induction was directly proportional to ROS production (Fig. 33), at least up to
the second highest injected dose, but also by the fact that ROS production increased
regardless amoebocytes were bacteria-stimulated or not. Whether the measured ROS were
produced only by amoebocytes or were also the result of PCB metabolisation processes (both
sources probably coexist), it is however a fact that ROS production dropped to control levels
in sea stars exposed to the highest dose of PCB 77. This drop could be explained by the death
of amoebocytes either through direct cytotoxicity of PCB 77 at high concentration or through
indirect PCB cytotoxicity via a ROS-triggered oxidative stress. Massive amoebocyte death
directly geopardizes the efficiency of sea stars immune system, leaving them vulnerable to
microbial attack. However, as ROS production did not drop down to zero but to control
levels, one could alternatively suggest that the drop in ROS production could also be due to
the inhibition of this immune response once a certain treshold of PCB 77 concentration is
reached in the coelomic fluid (this treshold would be comprised between the two highest
doses tested, viz. 1.51 and 5.96 ng g -1 whole-body fresh weight).
We conclude that cPCBs have the potency to induce detoxification mechanism and to impair
immune system of A. rubens, and therefore represent a threat for sea star populations and in
extenso for benthic ecosystems. However, observed effects at the immune level could be also
indirectly enhanced by CYP1A IPP activity, which can potentially produce toxic cPCB
metabolites, as occurs in the vertebrates. Measuring CYP induction and ROS production in
111

Contrasting effects of coplanar vs non-coplanar PCBs in the sea star Asterias rubens
parallel provides valuable information on the organisms health status and thus could serve as
an undissociable tool in environmental monitoring studies.
ACKNOWLEDGEMENTS
The IAEA Marine Environment Laboratory operates under a bipartite agreement between the
International Atomic Energy Agency and the Government of the Principality of Monaco. B.D.
and S.G. are respectively holders of a FRIA and FNRS doctoral grant. V.F. and Ph.D. are
Research Associates and M.W. is a Honorary Research Associate of the National Fund for
Scientific Research (NFSR), Belgium. Research is partly supported by an Agathon De Potter
(Belgian Royal Academy of Sciences) grant, a NFSR equipment funding (Tecan Spectrafluor
Plus) and a Belgian Federal Research Programme (SSTC, Contract MN/11/30 and
EV/11/23A). Special thanks are due to Geoffroy Coteur (Laboratoire de Biologie Marine,
ULB) for fruitful discussions on ROS mechanisms.
112

IV. Field Conditions
113

114

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
IV.1 Contaminant levels in sediments and asteroids (Asterias rubens L.,
Echinodermata) from the Belgian coast and Scheldt estuary: polychlorinated
biphenyls and heavy metals.
Science of the Total Environment (in press)
Danis B a , Wantier P b , Flammang R b , Dutrieux S a , Dubois Ph a & Warnau M a*
a : Laboratoire de Biologie Marine (CP 160/15), Université Libre de Bruxelles, 50 avenue
F.D.Roosevelt, B-1050 Bruxelles, Belgium
b : Laboratoire de Chimie Organique, Université de Mons-Hainaut, 19 avenue Maistriau, B-
7000 Mons, Belgium
* Present address: International Atomic Energy Agency, Marine Environment Laboratory, 4
Quai Antoine Ier, MC-98000 Monaco
115

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
ABSTRACT
The Southern Bight of the North Sea is particularly exposed to anthropogenic contamination,
due to its demographic and industrial situation. The present work focuses on PCB and metal
contamination of the marine environment along and off the Belgian coast. Its objectives were
to compare the concentrations of seven PCB congeners and four heavy metals in the
sediments (a sink for anthropogenic contaminants) and in the asteroid Asterias rubens (a
recognized bioindicator species). Nineteen sampling stations were considered between the
mouth of the Scheldt estuary and the southern limit of the Belgian coast (asteroids were found
in 10 out of the 19 stations). PCB and metal concentrations measured in sediments and
asteroids were in the range of values reported in previous studies. Stations under direct
influence of the Scheldt were the most impacted by the considered contaminants. Metal
concentrations varied according to the grain-size fraction considered. In asteroids, PCBs and
metals were found to be selectively distributed among body compartments and pyloric caeca
were found to discriminate sampling stations the most efficiently. The two strategies used in
this study were complementary: PCB and metal analysis of sediments provided a physico-
chemical evaluation of the contamination, whereas analysis of asteroids introduced a
biological dimension by taking into account bioavailability of the contaminants.
KEYWORDS
Echinoderms; Asterias rubens; heavy metals; polychlorinated biphenyls; sediments; North
Sea
116

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
INTRODUCTION
The Southern Bight of the North Sea is a highly urbanized and industrialized region, with
important rivers (viz. major vectors of anthropogenic contaminant fluxes to the sea) running
through its catchment area (NSTF 1993a,b). These rivers carry considerable amounts of
contaminants of domestic and industrial origins (NSTF 1993a,b, Bayens 1998) which
eventually flow into the marine environment.
When reaching marine waters, contaminants mainly concentrate in the sediments due to their
generally low solubility in seawater and their tendency to adsorb onto particles. Sediments
therefore constitute the bulk of anthropogenic contaminants in the coastal environment and
may be a major source of contamination for numerous organisms living in or close to them
(Stebbing et al. 1992, NSTF 1993a, Karbe et al. 1994, Alzieu & Michel 1998). Among the
latter organisms, the common asteroid Asterias rubens is a widely distributed and abundant
top-predator species in the North Sea which is known to influence the structure of benthic
communities (Menge 1982, Hayward & Ryland 1990, Hostens & Hammerlink 1994). This
species has proved to be an efficient tool for biomonitoring of several anthropogenic
contaminants (e.g., PCBs, metals, organometals) in laboratory and/or field studies (e.g.
Bjerregaard 1988, Everaarts & Fischer 1989, Temara et al. 1997a, 1998a,b, Warnau et al.
1999).
The present study is part of the "Sustainable Development of the North Sea" Program of the
SSTC (Belgian Government - Prime Minister Services) and focuses on anthropogenic
contamination of the Belgian marine environment. This marine region is directly exposed to
large amounts of anthropogenic contaminant carried by the Scheldt river which ends up
between the northern limit of the Belgian coast and the southern limit of the Dutch coast (see
e.g., Bayens 1998). Among these contaminants, polychlorinated biphenyls (PCBs) and metals
are of particular concern (NSTF 1993a, Mommaerts et al. 1994, Alzieu & Michel 1998,
Laane et al. 1999).
The aim of the present paper was to determine the concentrations of seven PCB congeners
(IUPAC# 28, 52, 101, 118, 138, 153, 180) and four heavy metals (Zn, Cd, Cu, and Pb) in
sediments (i.e. the main reservoir of anthropogenic contaminants in the marine environment)
and in the asteroid Asterias rubens (i.e. a dominant macrobenthic species whose bioindicative
value is recognized) in order to (i) compare and integrate the information gathered through
these two indicators and (ii) assess the quality status of the considered region.
117

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
MATERIALS AND METHODS
Sampling and sample preparation
Samples (sediment and asteroids) were collected between February 4th and March 31st
during a cruise with the RV “Belgica” and by seashore fishing and scuba diving. Nineteen
sampling stations were considered (Table 16, Fig. 34); 18 of them are located along the coast
or in the open sea (these stations may be arranged according to 7 transects parallel or
perpendicular to the coast) and one is located in a closed estuary (Scharendijke,
Grevelingenmeer, The Netherlands).
Figure 34. Sampling stations and transects along and off the Belgian coast.
Table 16. Positions and characteristics of the sampling stations
Station code
Coordinates
(N)(O)
Depth
(m)
118
Salinity
(p.s.u.)
Date of
sampling
(mm/dd/yy)
Sediment
collection
Asteroid
collection
120 51°11.10 2°42.07 11.6 33.4 02/05/98 yes no
130 51°16.25 2°54.30 8.32 32.0 02/05/98 yes no
140 51°20.00 3°24.00 5.73 31.2 02/05/98 yes no
150 51°25.00 3°24.00 13.7 32.1 02/05/98 yes no
230 51°18.50 2°51.00 11.5 32.2 02/13/98 yes yes
250 51°31.00 3°19.00 9.65 32.0 02/05/98 yes yes
330 51°26.00 2°48.50 24.0 33.4 02/06/98 yes yes
700 51°22.60 3°13.20 10.9 31.1 02/11/98 yes no
710 51°26.45 3°08.32 12.2 32.6 02/11/98 yes yes
Breskens 51°24.40 3°30.00 - 31.0 03/05/98 yes yes
Knokke 51°20.80 3°17.80 - 32.0 03/20/98 yes yes

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
Nieuwpoort 51°08.80 2°42.80 - 32.5 03/06/98 yes yes
Oostende 51°13.80 2°54.40 - 32.0 03/10/98 yes yes
Perkpolder 51°24.80 4°02.60 - 27.0 03/09/98 yes no
S01 51°25.00 3°34.20 17.0 29.9 02/05/98 yes no
Terneuzen 51°20.70 3°48.30 - 28.0 03/09/98 yes no
Wenduine 51°17.80 3°04.40 - 32.7 03/19/98 yes yes
ZG03 51°15.70 2°40.00 18.6 32.6 02/11/98 yes no
Scharendijke 51°44.50 3°50.70 5 - 15 29.0 03/04/98 yes yes
Sediments (6 replicates per station) were collected either using a Reineck box-corer during
the cruise or by hand in the intertidal zone during seashore collection. In both cases, only the
upper 5 cm layer of the sediments was considered and immediately frozen (-20°C) in 400 ml
acid-washed polypropylene containers. Asteroids (n = 15 per station) were collected either
using a beam trawl (during the cruise; trawling time ranged from 10 to 30 min, depending on
the sampling conditions i.e. currents, bottom nature, etc.) or by hand (seashore fishing) or
scuba diving. Only specimens belonging to the same size-class were considered. Immediately
after collection, asteroids were dissected to five compartments (oral body wall, aboral body
wall, gonads, gut and pyloric caeca). Dissected body compartments were pooled by 3 (leading
thus to 5 pools per station, each pool coming from 3 individuals) and frozen at 20°C). Tissues
to be analyzed for PCB content (all except gut, due to too little tissue quantities) were
wrapped in aluminum foils immediately after dissection and frozen at -20°C until analysis.
Prior to metal analyses, sediment and asteroid samples were dried at 800°C for 72 h. In
addition, dried sediments were sieved using an Endicots Octagon 200 siever in order to
determine grain-size distributions. The different fractions (>4 mm, 2-4 mm, 1-2 mm, 500-
1000 µm, 250-500 µm, 125-250 µm, 63-125 µm, and

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
dichloromethane. Extracts were then gathered. A fraction of the extracts was used to
determine the total lipid content (gravimetrically). The remaining part was evaporated under
nitrogen and 3 ml of isooctane were added to each sample. Lipids were eliminated by
precipitation with sulfuric acid (3 times). For sediment analysis, an additional purification
step using mercury to precipitate sulfur was performed at this stage. A final step of
purification on Fluorisil was then performed (40 ml hexane + 20 ml hexane:CH 2Cl 2, 95:5).
The eluate was evaporated, 10µl of internal standard (PCB 155) was added to 70 µl of the
sample and 2µl of this solution was injected in the GC/MS.
Samples were analyzed using a Finnigan GC/MS GCQ equipped with an AS9000
autosampler and a CP-Sil 8 capillary column (50 m length, 0.25 mm id, 0.25 µm film
thickness). Initial temperature of the column was 90°C. Temperature program was: (i)
increase to 180°C at 15°C min -1 and hold for 6 min, (ii) increase to 220°C at 4°C min -1 and
hold for 2 min and (iii) increase to 275°C at 5°C min -1 . The carrier gas was helium at a flux
rate of 30 cm sec -1 . Injection mode was “splitless”. Mass spectra were acquired in electron
impact in mode “multiple reaction monitoring”.
Concentration of congeners #28, 52, 101, 118, 153, 138 et 180 were measured in each
sediment sample and in the different body compartments of the asteroids. Accuracy of the
method was tested using certified material reference from the BCR (sediments from the
harbour “Nova Scotian” in East of Canada). Analysed CRM were always within 15% of the
mean certified values (except for congener 101 analyses that fell within 30% of the mean
certified value) (Table 17). The detection limits were between 0.01 and 0.1 ng g -1 dry wt,
depending on the PCB congener.
Table 17. Certified and measured PCB concentrations (ng g -1 dry wt ± sd, n = 4) in a certified material reference
(sediments from the harbour “Nova Scotian” in East of Canada)
Metal analyses
IUPAC # Certified value Analysis
101 1.62 ± 0.21 2.1 ± 0.49
138 1.98 ± 0.28 1.67 ± 0.60
153 2.27 ± 0.28 1.97 ± 0.43
180 1.17 ± 0.15 1.06 ± 0.28
The concentrations of Zn, Cu, Cd and Pb were measured in each of the 5 selected grain-size
fractions and in each asteroid body compartment, according to the method described by
Warnau et al. (1995). Briefly, samples (0.4 g for sediments, 1 g for asteroid body wall, and
0.5 g for other asteroid body compartments) were digested with 65% HNO 3 (Merck, p.a.
120

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
grade). Acid digestions were carried out at 20, 40, 60 and 80°C during respectively 24, 6, 6
and 12 h. Digests were diluted in milli-Q water (Millipore) and filtered on glass microfiber
filters (Whatman GF/A). Metal concentrations were determined by atomic emission
spectrometry using a Jobin-Yvon 38+ ICP-AES. Accuracy of the method was tested using
certified reference material (Mytilus edulis tissues, CRM n°278, BCR) (Table 18). Detection
limits for Zn, Cu, Cd, and Pb were respectively 0.002, 0.002, 0.001, 0.014 µg of metal per ml
of digested sample.
Table 18. Certified (C; mean value ± 95% CI) and measured (M; min and max values, n = 6) metal
concentrations (mg g -1 dry wt) of certified reference material (Mytilus edulis tissues, CRM n°278; BCR).
Data analyses
Zn Pb Cd Cu
C 76.0 ± 2.0 1.91 ± 0.04 0.34 ± 0.02 9.60 ± 0.16
M 77.0 – 82.0 1.91 - 1.95 0.31 - 0.37 9.60 - 10.7
For PCB analysis, it was assumed that the loss was the same for the surrogate (PCB 103) than
for the other analyzed congeners, and the PCB concentrations were corrected accordingly.
PCB 28 is not reported because of unexplained interference with other compounds. Everaarts
et al. (1999) have encountered the same type of analytical problems.
For metal analysis, emission intensities were converted to concentrations (µg g -1 ). Statistical
differences among metal concentrations measured in different grain-size fractions of
sediments, asteroid body compartments, and sampling stations were performed using 1- or 2-
way analysis of variance (ANOVA) followed by a multiple comparison test of means (Tukey
test) (Zar 1996). The variability explained by each considered factor was derived from the
sum of squares. Correlations between concentrations of different metals measured in
sediments or asteroids were tested using simple linear correlation procedures (Zar 1996).
Concordance of ranking of the stations according to the contamination level determined
through sediment and asteroid analyses were tested using Kendall's coefficient of
concordance (Zar 1996). The level of significance for statistical analyses was always set at
a=0.05.
121

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
RESULTS
PCBs in sediments
PCB concentrations were measured in the bulk sediment. Data was analyzed using 1-factor
ANOVA and are presented in Table 19.
Table 19. PCB concentrations (mean ± sd; mg g -1 dry wt, n = 6) in the bulk fraction of the sediments. Superscripts
indicate ranking of stations according to decreasing concentrations of a given PCB congener (a>b>c>…).
Stations with concentrations sharing a common superscript are not significantly different from each other
(multiple comparison test of Tukey; a = 0.05). nm = not measured
Station ∑ 6PCB #52 #101 #118 #138 #153 #180
120 4.79 bc ±0.56 0.20 ab ± 0.01 0.53 b ± 0.13 0.85 b ± 0.27 0.94 b ± 0.12 1.09 b ± 0.17 0.66 a ± 0.11
130 10.6 abc ±1.01 0.44 a ± 0.05 1.16 b ± 0.09 2.84 a ± 1.36 2.03 ab ± 0.31 2.12 ab ± 0.40 1.60 a ± 0.71
140 0.83 c ±1.03 0.07 b ± 0.05 0.14 b ± 0.10 0.28 b ± 0.35 0.15 b ± 0.19 0.21 b ± 0.18 0.14 a ± 0.14
150 1.06 bc ±0.62 nm nm 0.34 b ± 0.16 0.25 b ± 0.13 0.18 b ± 0.13 0.19 a ± 0.06
230 0.86 bc ±0.35 0.05 b ± 0.01 0.11 b ± 0.05 0.15 b ± 0.10 0.21 b ± 0.07 0.23 b ± 0.08 0.07 a ± 0.01
250 0.90 bc ±0.62 0.05 b ± 0.02 0.08 b ± 0.06 0.25 b ± 0.16 0.18 b ± 0.15 0.23 b ± 0.12 0.16 a ± 0.18
330 0.23 c ±0.09 0.05 b ± 0.02 0.03 b ± 0.01 nm 0.08 b ± 0.03 0.06 b ± 0.01 nm
700 1.93 bc ±0.74 0.23 ab ± 0.18 0.36 b ± 0.25 0.27 b ± 0.19 0.31 b ± 0.09 0.45 b ± 0.22 0.26 a ± 0.22
710 1.90 bc ±1.48 0.15 b ± 0.11 0.24 b ± 0.16 0.51 b ± 0.63 0.40 b ± 0.19 nm 0.25 a ± 0.08
Breskens 7.70 abc ±4.84 0.16 b ± 0.08 1.48 ab ± 0.471.35 ab ± 1.34 2.29 ab ± 1.65 1.72 ab ± 0.91 0.63 a ± 0.66
Knokke 2.93 bc ±3.64 0.05 b ± 0.01 0.31 b ± 0.18 0.53 b ± 0.01 1.10 b ± 1.48 0.80 b ± 1.05 nm
Nieuwpoort 4.13 bc ±1.38 0.08 b ± 0.03 1.15 b ± 0.38 0.79 b ± 0.32 0.94 b ± 0.45 1.09 b ± 0.37 0.14 a ± 0.05
Oostende 1.05 bc ±0.19 0.07 b ± 0.05 0.31 b ± 0.13 0.31 b ± 0.33 0.24 b ± 0.09 0.34 b ± 0.15 nm
Perkpolder 14.4 ab ±13.8 0.16 b ± 0.08 1.75 ab ± 1.411.22 ab ± 1.35 4.59 ab ± 4.59 3.90 ab ± 4.13 2.92 a ± 3.72
S01 7.46 bc ±1.52 0.41 a ± 0.23 0.92 b ± 0.161.35 ab ± 0.32 1.36 b ± 0.49 1.80 ab ± 0.39 1.07 a ± 0.12
Scharendijke 3.94 bc ±0.29 0.24 ab ± 0.11 0.79 b ± 0.23 0.73 b ± 0.13 0.71 b ± 0.10 0.88 b ± 0.02 0.59 a ± 0.30
Terneuzen 21.1 a ±11.3 0.16 b ± 0.06 3.09 a ± 1.531.83 ab ± 0.51 7.17 a ± 5.04 5.80 a ± 3.33 3.06 a ± 2.72
Wenduine 2.98 bc ±2.36 0.03 b ± 0.01 0.86 b ± 0.25 0.67 b ± 0.24 0.99 b ± 0.90 0.92 b ± 0.56 nm
ZG03 1.36 bc ±0.99 0.04 b ± 0.03 0.12 b ± 0.09 0.31 b ± 0.34 0.32 b ± 0.22 0.33 b ± 0.21 0.15 a ± 0.12
PCB concentrations varied significantly among the stations for all the congeners except #180.
Sum of 6 PCB concentrations varied between 0.23 and 21.1ng g -1 dry weight in sediment for
the 19 stations. The lowest concentrations were found in stations 330 and 140. Stations close
to the estuary and in front of Oostende (Terneuzen, Perkpolder, 130) were the most
contaminated. PCB patterns appeared identical for all the stations: PCBs #153 and #138 were
the most abundant whereas PCBs #52 and #180 were detected in lower concentrations.
Analyzing data in stations located along transects (Fig. 34) helped in obtaining more
discriminative information. 1-way ANOVAs resulted in significant differences in PCB
concentrations among stations for each investigated congener. In Transect I, Terneuzen
exhibited significantly higher concentrations of PCB 101 and of the sum of PCBs. Transect II
showed 130 and S01 were significantly more contaminated than other stations (all congeners
except PCB 52). No significant differences were found in Transect III. In Transect IV, station
122

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
120 was significantly more contaminated in most case. Station 130 was always significantly
more contaminated than the other stations in Transect V. Transect VI showed that Knokke
was significantly more contaminated (PCB 180). Finally, in Transect VII, it appeared that S01
was the most contaminated station (PCB 52) and that Terneuzen was significantly more
contaminated in all the other cases.
PCBs in asteroids
Asteroids were found in 10 out of the 19 investigated stations. Individuals were pooled by
groups of 3 (n = 3 to 5 pools per sampling site). Concentrations of 6 PCB congeners (PCB 28
is not reported because of interference with other compounds) were determined in four body
compartments (oral and aboral body walls, pyloric caeca and gonads) (Table 20).
Table 20. PCB concentrations (mean ± sd; ng g -1 total lipids, n = 3 pools of 3) in the different body
compartments of the asteroid Asterias rubens. Superscripts indicate ranking of stations according to decreasing
concentrations of a given congener (a>b>c>…). Stations with concentrations sharing a common superscript are
not significantly different from each other (multiple comparison test of Tukey; a = 0.05). nm = not measured.
A. Oral body wall
Station ∑ 6PCB #52 #101 #118 #138 #153 #180
230 431 a ± 101 19.3 a ± 16.1 63.4 a ± 24.7 46.7 a ± 10.6 169 a ± 40.5 118 a ± 11.6 nm
250 590 a ± 222 28.4 a ± 12.8 79.9 a ± 31.5 76.6 a ± 39.9 228 a ± 85.0 176 a ± 61.2 19.0 a ± 7.59
330 836 a ± 836 23.7 a ± 23.7 119 a ± 118 100 a ± 100 314 a ± 314 234 a ± 234 36.0 a ± 36.0
710 901 a ± 233 16.5 a ± 4.08 106 a ± 13.4 148 a ± 35.8 351 a ± 122 280 a ± 56.6 nm
Breskens nm nm nm nm nm nm nm
Knokke 1484 a ± 594 63.9 a ± 41.4 207 a ± 54.1 148 a ± 61.4 612 a ± 250 443 a ± 235 38.2 a ± 7.56
Nieuwpoort 1128 a ± 275 32.2 a ± 13.5 157 a ± 37.7 116 a ± 21.9 437 a ± 157 354 a ± 124 38.5 a ± 11.6
Oostende 1463 a ± 719 39.0 a ± 13.5 216 a ± 86.3 130 a ± 72.6 552 a ± 310 445 a ± 248 62.4 a ± 39.2
Scharendijke 1202 a ± 333 27.9 a ± 11.8 150 a ± 8.1 124 a ± 32.3 484 a ± 233 389 a ± 164 51.2 a ± 9.42
Wenduine 1045 a ± 331 78.9 a ± 45.3 178 a ± 100 119 a ± 24.4 377 a ± 51.9 298 a ± 110 29.9 a ± 3.29
B. Aboral body wall
Station ∑ 6PCB #52 #101 #118 #138 #153 #180
230 778 a ± 312 39.7 a ± 28.9 130 a ± 98.4 62.7 a ± 11.8 298 a ± 107 219 a ± 75.5 27.7 a ± 9.78
250 1015 a ± 55 45.2 a ± 16.9 105 a ± 52.0 116 a ± 75.4 410 a ± 208 296 a ± 160 55.0 a ± 51.1
330 nm nm nm nm nm nm nm
710 nm nm nm nm nm nm nm
Breskens nm nm nm nm nm nm nm
Knokke 1046 a ± 231 31.9 a ± 17.2 154 a ± 62.6 76.7 a ± 39.3 475 a ± 124 299 a ± 52.3 28.6 a ± 1.26
Nieuwpoort 919 a ± 272 53.4 a ± 16.3 70.7 a ± 49.4 80.9 a ± 23.6 393 a ± 137 287 a ± 94.4 62.7 a ± 49.2
Oostende 843 a ± 126 30.8 a ± 9.97 124 a ± 2.82 74.2 a ± 19.1 335 a ± 109 231 a ± 29.4 46.6 a ± 38.5
Scharendijke 1125 a ± 132 62.5 a ± 6.31 54.2 a ± 6.25 100 a ± 10.0 490 a ± 119 361 a ± 58.0 nm
Wenduine nm nm nm nm nm nm nm
123

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
C. Pyloric caeca
Station ∑ 6PCB #52 #101 #118 #138 #153 #180
230 562 b ± 102 31.3 b ± 0.47 102 b ± 21.9 54.1 a ± 32.8 189 b ± 7.17 131 b ± 51.8 27.5 a ± 20.5
250 735 b ± 233 29.0 b ± 4.76 107 b ± 32.3 110 a ± 50.2 262 b ± 103 197 b ± 69.3 15.6 a ± 9.43
330 592 b ± 27.5 22.2 b ± 9.79 61.0 b ± 5.51 59.6 a ± 13.9 223 b ± 20.2 182 b ± 19.4 24.7 a ± 4.54
710 695 b ± 440 42.5 ab ± 20.7 136 b ± 62.5 87.4 a ± 54.5 227 b ± 135 191 b ± 110 17.0 a ± 11.4
Breskens nm nm nm nm nm nm nm
Knokke 1695 a ± 658 88.6 a ± 29.5 317 a ± 119 150 a ± 44.9 614 a ± 270 463 a ± 215 35.9 a ± 20.8
Nieuwpoort 634 b ± 261 22.8 b ± 5.79 104 b ± 47.4 78.3 a ± 32.9 203 b ± 101 162 b ± 64.5 54.1 a ± 32.7
Oostende 689 b ± 338 24.9 b ± 16.4 108 b ± 65.9 74.9 a ± 32.3 230 ab ± 140 205 b ± 105 66.3 a ± 71.9
Scharendijke 501 b ± 246 24.9 b ± 8.52 90.8 b ± 44.3 53.9 a ± 33.9 169 b ± 106 120 b ± 52.3 26.3 a ± 15.4
Wenduine 396 b ± 20.0 45.7 ab ± 11.6 62.5 b ± 9.58 38.3 a ± 20.2 134 b ± 42.6 104 b ± 11.4 nm
D. Gonads
Station ∑ 6PCB #52 #101 #118 #138 #153 #180
230 655 b ± 164 56.9 a ± 19.3 121 b ± 42.6 71.9 a ± 4.94 200 b ± 70.6 169 b ± 47.3 23.6 b ± 4.37
250 561 b ± 88.3 23.5 a ± 8.48 83.3 b ± 14.6 63.7 a ± 32.1 213 b ± 50.5 158 b ± 27.4 16.1 b ± 5.12
330 nm nm nm nm nm nm nm
710 356 b ± 117 18.1 a ± 5.45 50.5 b ± 13.5 37.8 a ± 8.74 145 b ± 75.2 94.8 b ± 39.4 nm
Breskens nm nm nm nm nm nm nm
Knokke 1473 a ± 307 72.7 a ± 65.9 294 a ± 99.0 149 a ± 44.4 546 a ± 105 359 a ± 36.2 50.9 ab ± 15.9
Nieuwpoort 727 b ± 169 32.5 a ± 14.5 129 b ± 59.5 71.1 a ± 28.5 237 b ± 100 176 b ± 71.6 75.7 a ± 35.1
Oostende 688 b ± 257 27.1 a ± 15.0 117 b ± 46.2 80.6 a ± 46.1 219 ab ± 99.4 192 b ± 88.2 nm
Scharendijke 653 b ± 218 28.7 a ± 14.7 136 b ± 61.7 46.6 a ± 21.1 251 b ± 83.8 161 b ± 73.6 nm
Wenduine 1325 ab ± 166 59.5 a ± 4.76 229 ab ± 37.3 132 a ± 22.1 519 a ± 53.4 358 ab ± 72.9 37.1 ab ± 0.23
Regarding sea stars, sum of PCB concentrations were the highest in the pyloric caeca, varying
between 396 and 1695 ng g -1 total lipids. Oral and aboral body walls were generally more
contaminated than gonads. Pyloric caeca were also the most discriminative body
compartment. Sea stars collected in stations Knokke, Breskens, and Nieuwpoort revealed the
highest concentration levels. Individuals were less contaminated in stations Wenduine and
230.
The PCB contamination profile was similar in sediments and sea stars: PCBs #153 and #138
exhibited the highest concentrations, while congeners #52 and #180 were the less
concentrated.
PCB concentrations in the body compartments and their lipid contents were compared
(Fig.35). These two factors were highly correlated (r ≥ 0.99, p ≤ 0.01), confirming that PCB
concentration in the organism mainly depends on its lipid content.
124

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
∑6PCBs (ng g -1 dry wt.)
∑6PCBs (ng g -1 dry wt.)
300
250
200
150
100
50
r = 0.99
p = 0.01
Station Nieuwpoort
0
0 5 10 15 20 25 30 35 40
200
150
100
50
Lipid content (g 100g -1 dry wt.)
r = 0.99
p = 0.009
Station 250
0
0 5 10 15 20 25
Lipid content (g 100g -1 dry wt.)
Figure 35 Correlation between ∑ 6PCBs (ng g -1 dry weight) and lipid
content (g 100g -1 dry wt) in four body compartments of the asteroid A.
rubens for stations Nieuwpoort and 250.
Possible correlations were also assessed between PCB levels in sediments and asteroids, but
no significant correlation could be found. A striking difference between PCB concentration in
sediments and asteroids is exemplified by stations Knokke and 330, which both exhibit high
PCB concentrations in asteroids while being among the less impacted station when
considering sediments alone.
The two latter stations show a very different granulometry from the one of the other stations
(Table 21): fraction 250-500 µm is predominant in Knokke and 330 whereas fraction 125-250
µm is generally the most represented fraction in the other stations.
125

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
Table 21. Grain-size distribution (mean % ± sd, n = 6) in the dried sediments for the 19 sampling stations.
Predominant grain-size fractions are indicated in bold.
Station Grain-size fraction
500-1000µm 250-500 µm 125-250 µm 63-125 µm

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
Metals in sediments
The grain-size distributions of the different sediments are presented in Table 21. The fraction
size "125-250 µm" was the dominant one in 14 stations, fraction "250-500 µm" in 3 stations
and fractions "63-125 µm" and "500-1000 µm" only in one station.
Metal concentrations were measured in the different sediment fractions (Table 22). Since
some of the investigated metals have close physico-chemical properties, correlations between
their concentrations were determined in each grain-size fraction (Table 23). Highly significant
correlations (p≤0.0001) were found between all metals in all sediment fractions, except
between Cu vs Pb and Cu vs Cd in the c>…). Stations with concentrations sharing a common superscript are not significantly different from each
other (multiple comparison test of Tukey; a = 0.05)
A. Fraction

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
150 32.7 ± 7.20 cdefg
230 19.1 ± 2.10 efgh
250 16.8 ± 2.10 fgh
330 25.3 ± 8.10 defgh
700 37.4 ± 6.08 cde
710 31.5 ± 20.5 cdefgh
Breskens 34.3 ± 10.1 cdef
Knokke 14.3 ± 1.66 gh
Nieuwpoort 15.6 ± 1.60 fgh
Oostende 15.6 ± 2.40 fgh
Perkpolder 33.8 ± 4.80 cdef
S01 103 ± 6.00 a
Scharendijke 26.2 ± 4.90 defgh
Terneuzen 12.8 ± 1.40 h
Wenduine 16.5 ± 3.53 fgh
ZG03 42.1 ± 13.3 cd
15.8 ± 3.50 cde
11.4 ± 1.40 de
11.5 ± 0.80 de
14.6 ± 2.40 de
25.8 ± 3.36 bc
18.1 ± 7.10 bcd
17.0 ± 4.00 cde
16.4 ± 4.55 cde
11.0 ± 0.40 de
11.8 ± 1.50 de
19.0 ± 4.00 bcd
50.6 ± 3.50 a
13.0 ± 8.00 de
7.00 ± 1.00 e
17.3 ± 5.40 cde
28.9 ± 6.20 b
128
0.60 ± 0.13 de
0.50 ± 0.04 def
0.30 ± 0.02 ef
0.50 ± 0.06 def
1.23 ± 0.25 b
0.60 ± 0.18 def
0.55 ± 0.10 def
0.41 ± 0.01 ef
0.40 ± 0.02 ef
0.40 ± 0.02 ef
0.48 ± 0.07 def
1.60 ± 0.09 a
0.30 ± 0.02 ef
0.26 ± 0.02 f
0.43 ± 0.04 ef
0.80 ± 0.20 cd
3.40 ± 1.50 cdef
1.50 ± 0.40 ef
1.30 ± 0.30 f
2.20 ± 0.50 def
5.80 ± 1.28 c
3.20 ± 2.00 cdef
3.30 ± 1.30 cdef
1.50 ± 0.62 ef
1.00 ± 0.01 f
1.20 ± 0.01 f
3.60 ± 0.80 cdef
13.4 ± 1.80 a
4.50 ± 1.70 cde
1.00 ± 0.10 f
2.10 ± 1.92 def
5.40 ± 2.00 c
C. Fraction 125-250 µm
Station Zn Pb Cd Cu
120 28.7 ± 7.90 c
130 106 ± 7.90 a
140 15.6 ± 4.00 def
150 5.80 ± 0.70 f
230 14.3 ± 1.70 ef
250 15.0 ± 0.90 def
330 8.30 ± 0.80 f
700 34.0 ± 5.05 c
710 11.4 ± 2.00 f
Breskens 27.8 ± 9.90 cd
Knokke 5.80 ± 1.49 f
Nieuwpoort 11.4 ± 0.90 f
Oostende 7.40 ± 0.60 f
Perkpolder 35.0 ± 5.70 c
S01 63.1 ± 20.1 b
Scharendijke 26.6 ± 4.00 cde
Terneuzen 8.90 ± 0.90 f
Wenduine 6.00 ± 0.79 f
ZG03 12.8 ± 2.00 f
23.4 ± 4.50 cd
64.9 ± 3.30 a
15.8 ± 7.50 def
3.80 ± 0.80 g
4.30 ± 0.40 g
8.70 ± 0.60 efg
5.00 ± 0.20 g
25.9 ± 3.43 c
7.50 ± 1.50 fg
14.0 ± 4.00 ef
3.30 ± 0.44 g
7.40 ± 0.70 fg
4.00 ± 0.20 g
17.0 ± 2.00 de
43.1 ± 14.3 b
7.00 ± 1.00 fg
5.00 ± 0.01 g
4.80 ± 1.11 g
9.70 ± 0.70 efg
0.60 ± 0.15 c
2.50 ± 0.17 a
0.30 ± 0.07 de
0.10 ± 0.01 e
0.30 ± 0.01 de
0.30 ± 0.02 de
0.30 ± 0.01 e
1.03 ± 0.15 b
0.20 ± 0.04 e
0.47 ± 0.12 cd
0.13 ± 0.01 e
0.30 ± 0.02 de
0.20 ± 0.01 e
0.46 ± 0.07 cd
1.20 ± 0.28 b
0.20 ± 0.03 e
0.23 ± 0.02 e
0.16 ± 0.01 e
0.30 ± 0.03 de
3.20 ± 1.10 d
13.7 ± 1.00 a
1.10 ± 0.50 ef
0.40 ± 0.10 f
0.60 ± 0.01 f
0.90 ± 0.10 ef
0.70 ± 0.10 f
5.60 ± 1.16 bc
0.90 ± 0.30 f
2.80 ± 1.30 de
0.30 ± 0.05 f
0.60 ± 0.10 f
0.70 ± 0.40 f
3.80 ± 0.90 cd
7.40 ± 3.10 b
2.30 ± 0.50 def
0.60 ± 0.10 f
0.40 ± 0.03 f
1.10 ± 0.10 ef
D. Fraction 250-500 µm
Station Zn Pb Cd Cu
120 37.0 ± 11.3 cd
130 120 ± 13.2 ab
140 26.7 ± 12.8 cd
150 9.0 ± 1.70 cd
230 13.0 ± 2.60 cd
250 11.8 ± 1.40 cd
330 7.80 ± 1.10 cd
26.0 ± 7.10 de
66.6 ± 4.90 a
15.5 ± 7.50 efg
6.70 ± 3.30 gh
4.20 ± 1.40 h
6.90 ± 0.70 gh
7.40 ± 5.30 h
0.80 ± 0.21 cde
2.10 ± 0.30 a
0.70 ± 0.47 cdef
0.20 ± 0.03 gh
0.30 ± 0.05 gh
0.10 ± 0.03 h
0.30 ± 0.02 gh
4.70 ± 1.70 d
16.3 ± 2.30 b
2.40 ± 2.00 d
0.70 ± 0.20 d
0.90 ± 0.20 d
0.70 ± 0.10 d
0.70 ± 0.30 d

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
700 36.1 ± 8.65 cd
710 18.1 ± 5.30 cd
Breskens 95.1 ± 45.0 b
Knokke 3.50 ± 0.85 d
Nieuwpoort 19.4 ± 1.50 cd
Oostende 6.30 ± 0.70 d
Perkpolder 149 ± 51.1 a
S01 34.9 ± 5.00 cd
Scharendijke 39.4 ± 15.5 c
Terneuzen 20.3 ± 9.40 cd
Wenduine 4.70 ± 0.85 d
ZG03 31.2 ± 6.30 cd
34.5 ± 5.61 cd
10.4 ± 2.90 gh
45.0 ± 15.0 bc
3.30 ± 0.24 h
16.1 ± 1.80 efg
5.30.0 ± 1.60 gh
50.0 ± 10.0 b
22.8 ± 4.50 def
15.0 ± 6.00 efg
11.0 ± 6.00 fg
4.50 ± 0.41 gh
22.9 ± 3.80 de
129
1.10 ± 0.25 c
6.80 ± 0.89 cd
0.30 ± 0.08 fgh
1.30 ± 0.80 d
1.12 ± 0.41 c
13.2 ± 7.40 bc
0.13 ± 0.01 h
0.30 ± 0.02 d
0.50 ± 0.04 defg
1.50 ± 0.20 d
0.20 ± 0.01 gh
0.40 ± 0.01 d
1.62 ± 0.26b 27.2 ± 14.9 a
0.90 ± 0.07 cd
2.90 ± 0.70 d
0.34 ± 0.16 efgh
4.50 ± 2.20 cd
0.47 ± 0.16 defgh
0.18 ± 0.02 gh
0.80 ± 0.15 cdef
2.20 ± 1.30 d
0.30 ± 0.03 d
3.60 ± 1.00 d
E. Fraction 500-1000 µm
Station Zn Pb Cd Cu
120 40.1 ± 10.9 bcd
42.3 ± 4.30 bc
1.20 ± 0.22 a
6.20 ± 1.60 bcd
130 99.0 ± 17.2 ab
85.0 ± 34.5 ab
1.20 ± 0.32 a
13.4 ± 3.00 b
140 53.3 ± 49.9 bcd
28.7 ± 24.5 bc
0.88 ± 0.81 ab
7.10 ± 6.10 defgh
150 65.7 ± 17.5 abc
34.6 ± 7.90 bc
1.30 ± 0.32 a
9.30 ± 2.90 bc
230 28.2 ± 3.60 cd
15.8 ± 3.20 c
0.60 ± 0.09 ab
2.80 ± 0.80 efgh
250 31.9 ± 5.50 bcd
20.7 ± 1.60 c
0.62 ± 0.06 ab
3.10 ± 0.75 defgh
330 25.6 ± 5.10 cd
24.5 ± 2.60 bc
1.30 ± 0.24 a
1.80 ± 0.30 fgh
700 29.8 ± 14.3 bcd
37.0 ± 8.55 bc
0.88 ± 0.16 ab
5.70 ± 1.84 cdef
710 37.8 ± 8.10 bcd
32.3 ± 3.20 bc
1.20 ± 0.12 a
5.40 ± 1.40 cdefg
Breskens 53.3 ± 26.8 d
105 ± 104 a
0.75 ± 0.41 b
7.20 ± 2.40 cde
Knokke 9.80 ± 0.82 cd
15.2 ± 0.78 c
0.58 ± 0.04 ab
1.00 ± 0.07 gh
Nieuwpoort 25.8 ± 17.2 cd
23.7 ± 4.20 bc
0.87 ± 0.17 ab
1.80 ± 0.50 fgh
Oostende 16.0 ± 1.30 cd
26.2 ± 2.60 bc
0.98 ± 0.09 a
1.90 ± 0.27 fgh
Perkpolder 104 ± 29.7 a
38.0 ± 7.00 bc
1.12 ± 0.21 a
18.6 ± 5.50 a
S01 65.3 ± 33.4 abc
45.2 ± 24.4 bc
1.40 ± 0.32 a
7.50 ± 3.10 cd
Scharendijke 34.8 ± 15.7 24.0 ± 16.0 0.27 ± 0.10 21.3 ± 45.7
Terneuzen 2.10 ± 19.6 cd
28.0 ± 8.00 c
0.59 ± 0.07 ab
1.80 ± 0.90 h
Wenduine 14.3 ± 1.39 cd
23.1 ± 1.01 bc
1.04 ± 0.08 a
1.30 ± 0.05 gh
ZG03 28.2 ± 9.50 cd
31.7 ± 4.4 bc
1.30 ± 0.16 a
3.80 ± 1.40 defgh
Table 23 Correlation coefficients (r) between metal concentrations measured in the different grain-size fractions
of the sediments (pcorrelation always < 0.0001 except when ns -non significant correlation- is indicated)
A. Fraction 500-1000 µm
Pb Cd Cu
Zn 0.66 0.51 0.89
Pb 0.62 0.79
Cd 0.59

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
B. Fraction 250-500 µm
C. Fraction 125-250 µm
D. Fraction 63-125 µm
E. Fraction

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
followed by the multiple comparison test of Tukey. Results of the 1-way ANOVAs showed
significant differences in metal concentrations among stations for each investigated element
and each sediment grain-size fraction (Table 22). Though the contamination level of several
stations was not well contrasted, some stations (viz. Perkpolder, SO1 and 130) generally
exhibited significantly higher metal concentrations than the other ones, whereas other stations
(Knokke, Terneuzen, Wenduine, Oostende and Nieuwpoort) were generally ranked among the
less contaminated ones.
When data was analyzed among stations located along transects parallel or perpendicular to
the coast (see Fig. 34), a better discrimination between stations was obtained. In Transect I,
Perkpolder and Breskens exhibited significantly higher concentrations of Cu (grain-size 500-
1000µm), of Cd (grain-size 250-500µm) and of Zn (grain-size 250-500µm). Transect II
showed 130 and S01 were significanly more contaminated than other stations (Cu: 250-
500µm, 125-250µm and 63-125µm; Cd: 250-500µm and 125-250µm; Pb: all except the
finest fraction; Zn: all except 500-1000µm). ZG03 was the most contaminated station of
Transect III, showing the highest concentrations in Cu (250-500µm and

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
Table 25 Metal concentrations (mean ± sd; mg g -1 dry wt, n = 5 pools of 3) in the different body compartments of
the sea star Asterias rubens. Superscripts indicate ranking of stations according to decreasing concentrations of a
given metal (a>b>c>…). Stations with concentrations sharing a common superscript are not significantly
different from each other (multiple comparison test of Tukey; a = 0.05)
A. Oral body wall
Station Zn Pb Cd Cu
230 38.1 ± 25.8 c
250 86.9 ± 19.9 ab
330 59.1 ± 8.7 bc
710 73.3 ± 13.1 abc
Breskens 98.5 ± 13.5 a
Knokke 73.1 ± 5.9 abc
Nieuwpoort 44.4 ± 23.3 c
Oostende 68.9 ± 16.1 abc
Scharendijke 54.6 ± 19.2 bc
Wenduine 82.2 ± 13.0 ab
B. Aboral body wall
0.85 ± 0.63 bc
1.47 ± 0.44 ab
2.11 ± 0.97 a
1.62 ± 0.07 ab
1.77 ± 0.43 ab
2.00 ± 0.34 ab
0.16 ± 0.02 c
1.77 ± 0.18 ab
0.46 ± 0.58 c
1.37 ± 0.13 ab
132
0.10 ± 0.06 cde
0.16 ± 0.04 cd
0.21 ± 0.02 bc
0.17 ± 0.03 bcd
0.31 ± 0.06 a
0.25 ± 0.02 ab
0.03 ± 0.02 e
0.19 ± 0.01 bc
0.09 ± 0.08 de
0.17 ± 0.02 cd
0.38 ± 0.22 abc
0.51 ± 0.19 abc
0.91 ± 0.14 a
0.42 ± 0.06 bcd
0.55 ± 0.30 abc
0.75 ± 0.14 ab
0.02 ± 0.00 e
0.25 ± 0.24 de
0.07 ± 0.11 e
0.12 ± 0.09 de
Station Zn Pb Cd Cu
230 109 ± 65.8 d
250 156 ± 34.7 ab
330 91 ± 24.6 e
710 189 ± 35.5 ab
Breskens 161 ± 28.7 ab
Knokke 135 ± 37.8 b
Nieuwpoort 127 ± 26.5 b
Oostende 120 ± 26.2 b
Scharendijke 210 ± 50.0 a
Wenduine 108 ± 21.1 c
C. Pyloric caeca
0.36 ± 0.19 bc
0.26 ± 0.03 ab
1.15 ± 0.16 c
0.52 ± 0.12 bc
0.91 ± 0.13 abc
1.16 ± 0.53 ab
0.52 ± 0.28 bc
0.92 ± 0.72 abc
0.73 ± 0.20 abc
1.43 ± 0.32 a
0.22 ± 0.11 bc
0.27 ± 0.06 bc
0.31 ± 0.10 abc
0.28 ± 0.06 bc
0.47 ± 0.10 ab
0.51 ± 0.19 a
0.25 ± 0.08 c
0.26 ± 0.03 c
0.25 ± 0.06 c
0.27 ± 0.05 bc
0.54 ± 0.67 c
0.04 ± 0.00 c
0.04 ± 0.00 c
1.45 ± 0.62 a
0.02 ± 0.00 c
0.13 ± 0.17 c
0.03 ± 0.01 c
0.03 ± 0.01 c
0.03 ± 0.01 c
0.26 ± 0.19 b
Location Zn Pb Cd Cu
230 53.0 ± 64.5 b
250 45.1 ± 8.80 b
330 58.3 ± 9.30 b
710 149 ± 12.4 ab
Breskens 89.4 ± 10.6 b
Knokke 73.1 ± 4.90 b
Nieuwpoort 57.2 ± 12.0 b
Oostende 119 ± 27.9 b
Scharendijke 282 ± 186 a
Wenduine 88 ± 9.3 b
0.95 ± 0.73 abc
1.27 ± 0.25 abc
1.19 ± 0.23 bc
0.63 ± 0.16 c
2.03 ± 0.32 a
1.51 ± 0.13 ab
1.16 ± 0.70 bc
1.18 ± 0.10 bc
1.22 ± 0.58 bc
1.47 ± 0.16 abc
1.31 ± 1.15 ab
0.83 ± 0.52 b
0.60 ± 0.23 b
0.92 ± 0.25 ab
1.06 ± 0.64 ab
2.36 ± 1.40 a
1.11 ± 0.38 ab
1.62 ± 0.78 ab
0.56 ± 0.18 b
0.62 ± 0.13 b
5.12 ± 4.15 c
5.45 ± 0.64 c
19.0 ± 9.70 bc
7.63 ± 3.37 c
30.3 ± 17.6 b
16.4 ± 2.97 bc
9.00 ± 1.44 c
11.6 ± 2.18 c
51.1 ± 14.4 a
11.8 ± 0.75 bc

D. Gut
Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
Location Zn Pb Cd Cu
230 37.6 ± 37.5 c
250 79.9 ± 4.60 c
330 70.2 ± 8.60 c
710 68.8 ± 2.70 c
Breskens 72.9 ± 6.30 c
Knokke 67.5 ± 4.60 c
Nieuwpoort 59.3 ± 1.80 c
Oostende 58.6 ± 30.1 c
Scharendijke 81.7 ± 16.4 c
Wenduine 70.9 ± 26.0 c
E. Gonads
0.92 ± 0.33 c
1.09 ± 0.75 c
2.11 ± 2.09 c
1.15 ± 0.66 c
1.36 ± 0.24 c
0.92 ± 0.09 c
1.08 ± 0.50 c
1.20 ± 0.46 c
1.86 ± 2.00 c
1.68 ± 0.76 c
133
0.19 ± 0.08 c
0.21 ± 0.08 c
0.24 ± 0.13 c
0.22 ± 0.08 c
0.26 ± 0.09 c
0.19 ± 0.07 c
0.19 ± 0.04 c
0.16 ± 0.05 c
0.12 ± 0.02 c
0.19 ± 0.04 c
3.76 ± 3.19 ab
4.92 ± 1.06 ab
4.84 ± 1.72 ab
4.25 ± 1.67 ab
4.92 ± 2.18 ab
2.33 ± 0.28 b
7.11 ± 2.70 a
2.72 ± 1.22 ab
6.52 ± 3.48 a
3.52 ± 1.65 ab
Location Zn Pb Cd Cu
230 37.6 ± 15.9 c
250 55.0 ± 16.7 c
330 37.2 ± 25.9 c
710 23.0 ± 25.3 c
Breskens 32.3 ± 9.1 c
Knokke 42.4 ± 29.8 c
Nieuwpoort 56.7 ± 7.3 c
Oostende 39.7 ± 25.7 c
Scharendijke 78.1 ± 16.7 c
Wenduine 40.4 ± 17.7 c
0.59 ± 0.37 c
1.02 ± 0.03 c
1.44 ± 1.36 c
0.96 ± 0.10 c
0.58 ± 0.06 c
1.48 ± 0.48 c
1.01 ± 0.45 c
0.76 ± 0.20 c
1.75 ± 1.87 c
0.88 ± 0.13 c
0.09 ± 0.07 ab
0.06 ± 0.01 b
0.12 ± 0.08 ab
0.05 ± 0.01 b
0.11 ± 0.02 ab
0.08 ± 0.04 b
0.18 ± 0.04 a
0.11 ± 0.03 ab
0.12 ± 0.01 ab
0.11 ± 0.04 ab
0.74 ± 0.73 b
1.25 ± 0.21 b
0.90 ± 1.11 b
1.52 ± 1.19 b
2.04 ± 0.58 b
2.07 ± 0.96 b
6.66 ± 2.31 a
1.78 ± 0.85 b
6.18 ± 3.24 a
1.93 ± 0.71 b
Table 26. Correlation coefficients (r) between metal concentrations measured in the different sea star body
compartments (p correlation always

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
Data were analyzed using 2-way ANOVA (Table 27). These analyses showed that metal
concentrations differed significantly among stations and among body compartments for all
investigated elements.
Table 27. Variability (%) in metal concentrations measured in Asterias rubens explained by the factors
considered (station, body compartment) and their interaction.
Factor Zn Pb Cd Cu
Station 15.5 10.1 6.2 11.1
Body compartment 29.8 7.8 50.1 39.8
Station x Body compartment 26.1 22.8 18.2 35.0
Residual 28.6 59.3 25.4 14.1
In particular, the multiple comparison test of Tukey (data not shown) indicated that Cd and
Cu were always more concentrated in pyloric caeca, while Zn tended to be preferentially
concentrated in the aboral body wall; no clear trends was found for Pb. Two-way ANOVA
also showed that the two factors considered (station sampled and body compartment) and
their interaction were responsible for 41-86% of the variability in metal concentrations
measured in asteroids. Except for Pb, the largest part of variation was always associated to the
"body compartment" factor. Interaction between "station" and "body compartment" factors
was always significant and accounted for an important part of the observed variablity (18-
35%), indicating that differences in station contamination varied according to the considered
body compartment. Therefore, data had to be re-analyzed using 1-way ANOVA, considering
separately each body compartment. The latter analyses showed that, except for Cd, Pb and Zn
in the digestive tract and for Pb and Zn in the gonads, significant differences in metal
concentrations occurred among stations (p ANOVAalways

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
significant correlations were found between metal concentrations measured in sediments and
in asteroid body compartments (oral and aboral body walls and pyloric caeca). However,
correlation coefficients were generally low (maximum value of r = 0.74).
Station rankings using sediment and asteroid data were compared using Kendall's coefficient
of concordance. Results of these tests did not show any significant concordance. This lack of
ranking concordance was probably mainly due to several intermediary-contaminated and
poorly contrasted stations whose ranking was quite variable.
DISCUSSION
Determination of the grain-size distribution in collected sediments showed that the fraction
125-250 µm (fine sand) was predominant in most of the considered stations. Its degree of
predominance varies significantly from one station to another, suggesting that local
hydrodynamics is variable.
PCB analyses in bulk sediments showed that the lowest concentration levels were found in
the offshore stations (e.g. stations 330 and 250) while the highest levels were measured in
stations under direct influence of the Scheldt river (Terneuzen, Perkpolder, 130, S01). These
results are in agreement with those found in other studies (Laane et al. 1999) in the Dutch
coastal zone. Station 130 is localized where the Scheldt gyre meets North-Atlantic currents,
giving place to locally favorable conditions for the sedimentation of heavily contaminated
particles swepped by the Scheldt. Stations Perkpolder, Terneuzen and S01 are situated in the
Scheldt and consequently receive highly contaminated flows. Considering PCB congeners
separately, #101 was the most discriminating (although not the most abundant). As found in
previous studies (e.g. Stebbing et al. 1992), the most abundant congeners in sediments were
#153 and #138.
The use of transects (parallel or perpendicular to the coast, Fig. 34) brought a more
discriminative approach, as a smaller number of stations were considered. This helped in
enlightening more subtle differences between contamination levels in the sampling stations.
Perpendicular transects (Transects IV, V, and VI) were particularly useful in showing the
contamination gradient direction: intermediary stations were always more contaminated.
Highest PCB concentrations were measured along the halfway transect (Transects II), which
includes stations under direct influence of the Scheldt (Lizen 1990, Bayens 1992). This shows
that PCB contamination from the Scheldt spreads out southerly, along the coast.
PCB concentrations measured in asteroids were comparable to those measured in previous
studies (Everaarts et al. 1998, den Besten et al. 2001). PCBs were distributed selectively
135

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
between body compartments of the sea stars. Generally, pyloric caeca exhibited the highest
concentrations. No significant difference was found between contaminant levels in aboral and
oral body walls although only the oral body wall is in direct contact with sediments.
Sampling sites were ranked according to their PCB concentration level in sediments and in
asteroids. No rank concordance was found, probably because stations such as Knokke, 330 or
Wenduine presented surprising differences between relative contamination of asteroids and
sediments. These three stations were the only ones where fraction 250-500 µm was dominant
(together with station S01, where no sea stars were found). The grain-size distribution varied
from one station to another and bioavailability of contaminants is known to vary according to
grain-size (Luoma & Carter 1991). Therefore the relative importance of these fractions might
influence the exposure of asteroids to the contaminants. A significant correlation (r=0.85)
was found between asteroid concentration ratios (CRs) and the percentage of the 250-500 µm
grain size fraction in sediments. This suggests that contaminants associated to this grain-size
fraction are more readily available for asteroids.
Another cause for the lack of concordance between ranking of stations obtained via sediment
and asteroid analysis could be that asteroid specimens were found only in some sampling
stations, and particularly that none were found along Transect II, where sediments were the
most contaminated. The sediment type (muddy) along this transect is not appropriate for
asteroid populations (Hayward & Ryland 1990). The fact that sea stars were found in stations
where sediments contamination level was not very well contrasted might also have affected
statistical tests (i.e. rank concordance). Other authors have encountered the same problem
(e.g. Guns et al. 1999).
Metal analysis in the different grain-size fractions of the sediments showed selective
distribution. The finest fraction (

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
Atlantic currents. A good explanation for its high contamination by heavy metals is that this
station is the most saline considered in the Scheldt. Indeed, it is a well described phenomena
that organic matter tends to flocculate when salinity increases and that metals have a tendency
to complex with suspended organic matter (NSTF 1993a, Verlaan et al. 1998). Station 130 is
located in a low-energy area, where the Scheldt gyre contacts the Northwards Atlantic
currents: their interaction leads to local conditions allowing sedimentation of very fine
particles swept by the Scheldt (Lizen 1990). Coastal stations, especially Knokke, Oostende
and Wenduine exhibit relatively low metal concentrations: they are probably partly protected
from the Scheldt particular contribution.
As in the case of PCBs, the use of transects for metal analysis enlightened more subtle
differences between contamination levels in the sampling stations. As for PCBs, the highest
metal concentrations were measured along Transect II.
The metal measurements in asteroids showed that investigated elements are selectively
distributed among body compartments. Cu and Cd are more concentrated in the pyloric caeca,
Zn is more efficiently concentrated in the aboral body wall, while Pb does not seem to have a
preferential "target" compartment. The latter observation was somehow unexpected. Indeed,
data available in the literature generally report that Pb tends to be more efficiently
incorporated in calcitic skeletons (Kersten & Kroncke 1991, Temara et al. 1997b, 1998b).
Differences in metal contamination among stations were found in each asteroid body
compartment. Among these compartments, three of them (pyloric caeca, oral body wall, and
aboral body wall) showed better discrimination among sampling stations. Station ranking was
obtained from sediment and tissue analyses. As in the case of PCBs, no concordance was
found between ranking obtained for sediments and asteroids, enlightening that using a
physico-chemical approach (sediments) might not be sufficient to characterize a
contamination as it does not take into account bioavailability.
Significant correlations were found between metal concentrations measured in each sediment
grain-size and in each asteroid body compartment. The highest correlations were found for
grain-size fraction

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
body compartments. These two sources coexist: some authors have shown that A. rubens
concentrates heavy metals from seawater (den Besten et al. 1990, Temara et al. 1996, 1998a,
Warnau et al. 1999), and translocation processes are well-documented in echinoderms
(Temara et al. 1996, 1998b, Warnau et al. 1995c, 1996b, 1999).
Using Asterias rubens in this study allowed us to take into account the multiple contamination
sources and the variation of bioavailability of metals depending on sediments. Investigating
biota brings complementary information that is necessary to tackle the ecological
consequences of a contamination. A. rubens helped in putting forward a group of more
contaminated stations, different from that which would have resulted from a study limited to
sediments analysis. Analysis of biota and sediments is however more demanding than the
latter approach alone because it implies finding organisms. This constraint can be solved -for
example- by using active monitoring or bioassays (den Besten et al. 2001).
In sediments, concentrations of Zn, Pb, Cd and Cu are in the range of concentrations
generally reported (e.g. Bayne et al. 1988, Everaarts & Fischer 1989, Stebbing et al. 1992,
NSTF 1993a,b). However, Cd concentrations along Transect II are in the superior part of the
ranges generally reported in sediments from the North Sea. This transect can therefore be
considered as relatively contaminated, at least by Cd. Concentrations measured in the
asteroids are also comprised in the range of concentrations found in literature for the same
species (Riley & Segar 1970, Bryan 1984, Everaarts & Fischer 1989, Temara et al. 1997).
Sampling stations where asteroids were found are not particularly contaminated by heavy
metals, implying a certain weakening of statistical tests (no concordance between sediment
and asteroid contamination levels).
CONCLUSIONS
Variations in contaminant levels in sediments from the Southern Bight of the North Sea are
mainly explained by hydrodynamic factors: stations with low residual currents are often the
most contaminated ones. Sediments are a source of contamination for asteroids, although it
may not be considered as the main one. The use of A. rubens has shown that this species helps
in distinguishing the contamination state of selected stations while taking into account
bioavailability. Some body compartments appear as better bioindicators, ie. pyloric caeca.
Information brought by the use of asteroids does not compare with that brought by the use of
sediments, and this highlights the observable differences between an "abiotic" indicator
(sediments) and a biotic indicator (asteroids), the latter one integrating the bioavailability of
contaminants.
138

Contaminant levels in sediments and in the sea star Asterias rubens from the Belgian coast and Scheldt estuary
ACKNOWLEDGEMENTS
Research supported by a Belgian Federal Research Program (SSTC, Contract MN/11/30).
Grateful thanks are due to Ch. De Ridder, C. De Amaral, G. Coteur (ULB), P. Gosselin
(UMH) and G. Radenac (Univ. La Rochelle, France) for helpful assistance in offshore and sea
shore samplings. M. Warnau and Ph. Dubois are respectively Honorary Research Associate
and Research Associate of the National Fund for Scientific Research (NFSR, Belgium).
Contribution of the "Centre Interuniversitaire de Biologie Marine" (CIBIM).
139

140

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
IV.2 Bioaccumulation and effects of PCBs and heavy metals in sea stars
(Asterias rubens, L.) from the North Sea: a small scale perspective
Marine Pollution Bulletin (submitted)
Danis B a , Wantier P b , Flammang R b , Pernet Ph a , Chambost-Manciet Y a , Coteur G a ,
Warnau M a,c & Dubois Ph a
a : Laboratoire de Biologie Marine (CP 160/15), Université Libre de Bruxelles, 50 avenue
F.D. Roosevelt, B-1050 Bruxelles, Belgium
b : Laboratoire de Chimie Organique, Université de Mons-Hainaut, 19 avenue Maistriau, B-
7000 Mons, Belgium
c: Present address: Marine Environment Laboratory, International Atomic Energy Agency, 4
quai Antoine 1er, MC-98000 Monaco
141

ABSTRACT
Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
Sea stars (Asterias rubens L.) were collected in different stations distributed in the Southern
Bight of the North Sea. Concentrations of four heavy metals and six PCB congeners were
measured in two body compartments (body wall and pyloric caeca). In order to assess the
potential harm of these contaminants, two biological effects were measured in sea stars, viz.
reactive oxygen species (ROS) production by amoebocytes and cytochrome P450
immunopositive protein (CYP1A IPP) induction in pyloric caeca. Sea stars from stations
located in the plume of the Scheldt river showed the highest contamination levels. Other
stations, similarly located, displayed lower levels. No simple relationship could be established
between ROS production by sea star amoebocytes and contaminant levels measured in sea
star tissues. CYP1A IPP induction displayed more contrasted responses, and highly
significant regressions were found between PCB concentrations measured in pyloric caeca
and CYP1A IPP. Both key biological processes were found to be significantly affected. On
the whole, data indicated that contamination levels and subsequent effects in sea stars were
comparable to those described in previous large-scale studies, but that working at a smaller
scale highlighted the existence of patterns of contamination which can blur general trends due
to major contamination sources like contaminated rivers.
KEYWORDS
PCBs; Heavy metals; North Sea; Echinoderms; CYP1A; ROS
142

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
INTRODUCTION
Harmful substances, such as heavy metals or polychlorinated biphenyls (PCBs), are carried
by rivers or the atmosphere and eventually reach and contaminate the marine environment
(Bayens 1998, OSPAR 2000). These contaminants are of particular concern due to their
persistence, bioavailability and toxicity to marine species (Mommaerts et al. 1994, Alzieu &
Michel 1998, Laane et al. 1999, OSPAR 2000). Because of their low solubility in seawater,
heavy metals and PCBs tend to adsorb on particles and concentrate in sediments, ultimately
affecting organisms living in or close to them (Stebbing 1992, Alzieu & Michel 1998,
OSPAR 2000). The delay between their introduction in the environment and the first
observable effects on organisms can vary between a few hours and decades, depending on the
species and/or biological level considered (Everaarts et al. 1998).
The common sea star, Asterias rubens (Echinodermata, L.), has proven to be a valuable
organism for ecotoxicological testing (Bjeregaard 1988, Everaarts & Fisher 1989, Temara et
al. 1997a, 1998a,b, Everaarts et al. 1998, Warnau et al. 1999, den Besten et al. 2001). Its
position as a predator in the food chain “seston-mussels-seastars”, its wide distribution and its
key position in North Sea benthic communities contribute to its high potential as a sentinel
organism (Menge 1982, Hayward & Ryland 1990, Hostens & Hammerlink 1994, Temara et
al. 1998b, Den Besten et al. 2001, Coteur et al. 2003a,b,c). Different biomarkers have been
used in A. rubens, among which the levels of cytochrome P450 in the digestion and storage
organs (the so-called pyloric caeca) (den Besten et al. 1991, 1993, Danis et al. Chap. III.4)
and the production of reactive oxygen species (ROS) by amoebocytes (the main immune cell
type circulating in coelomic cavities) (Coteur et al. 2002a,c, 2003a,c). ROS production
participates in the cytotoxic destruction of microorganisms (Chia & Xing 1996, Gross et al.
1999). CYP1A is an enzyme family whose main function is to convert insoluble organic
compounds to soluble metabolites that can be excreted (Bucheli & Fent 1995). However,
some of these metabolites can in turn reveal more toxic than the original compounds
(Stegeman 1995): CYP1A-dependent oxidation is found to activate compounds such as
polyaromatic hydrocarbons (PAHs) or PCBs to various reactive intermediates that ultimately
cause toxicity, mutagenicity and carcinogenicity (Walker & Peterson 1994).
A broad scale study by den Besten and collaborators (2001) focused on the relation between
biomarkers and contaminant accumulation in sea stars collected along pollution gradients in
the central and central North Sea. These authors showed that contamination by heavy metals
and PCBs as well as some biomarker responses were relatively high in stations located close
to estuaries. The goal of the present study is to determine contaminant concentrations and
143

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
biomarker responses in the same sea star species at fine scale, focusing on stations located in
or close to the plume of the Scheldt river, one of the highly contaminated estuaries of the
North Sea. Considered contaminants were 4 heavy metals (Zn, Cd, Cu and Pb), 6 PCB
congeners (IUPAC #52, 101, 118, 138, 153, 180); recorded biological effects were CYP1A
immunopositive protein induction and ROS production.
MATERIALS AND METHODS
Field trips and sample preparation
Asteroids were collected in February 1999, during cruise n°9905 on the RV “Belgica” and by
seashore fishing in 10 sampling stations (Table 28).
Table 28. Positions and characteristics of the sampling stations
Station code
Coordinates
(N)(E)
144
Depth
(m)
Salinity
(‰)
Date of sampling
(dd/mm/yy)
230 51°18.50 2°51.00 16 29.2 24/02/99
340 51°29.94 2°59.45 26 31.3 25/02/99
710 51°26.45 3°08.32 11 30.0 24/02/99
Breskens 51°24.40 3°30.00 Intertidal 31.0 03/03/99
GC1 51°34.40 3°24.80 11 31.2 24/02/99
Knokke 51°20.80 3°17.80 Intertidal 32.0 23/03/99
Nieuwpoort 51°08.80 2°42.80 Intertidal 32.5 11/03/99
Oostende 51°13.80 2°54.40 Intertidal 32.0 11/03/99
S01 51°25.00 3°34.20 27 27.9 23/02/99
Scharendijke 51°44.50 3°50.70 5-10 29.0 15/03/99
Sea stars were collected using a beam trawl (mesh: 30 mm; counter-current trawling for 10 to
30 min), by hand (intertidal stations) or by SCUBA diving (Scharendijke, Grevelingenmeer:
closed estuary system). Only specimens belonging to the same size-class (5-7 cm from the tip
of the arm to the center of the mouth) were considered. After coelomic liquid was withdrawn
by bleeding for ROS measurement, two compartments (body wall and pyloric caeca) were
immediately isolated by dissection. A part of the pyloric caeca (ca. 5 ml) was immediately
frozen in liquid nitrogen for cytochrome P450 immunopositive protein (CYP1A IPP)
analysis. Remaining pyloric caeca and the integument intended for metal analyses were
placed in PET recipients immediately after dissection , while those intended for PCB analysis
were pooled by 5 (3 pools per station) and wrapped in aluminum foil. Sampled compartments
were frozen at -20°C until analysis.

Metal analyses
Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
The concentrations of Zn, Cu, Cd and Pb were measured in each asteroid body compartment
according to the method described by Warnau et al. (1995). Briefly, tissues were dried (60°C,
72 hrs) and samples (1 g of body wall, 0.5 g of pyloric caeca) were digested with 65% HNO 3
(Merck, p.a. grade). Acid digestions were carried out at 20, 40, 60 and 80°C during
respectively 24, 6, 6 and 12 h. Digests were diluted in Milli-Q water (Millipore) and filtered
on glass microfiber filters (Whatman GF/A). Metal concentrations were determined by atomic
emission spectrometry using a Jobin-Yvon 38+ ICP-AES. Accuracy of the method was tested
using certified reference material (Mytilus edulis tissues, CRM n°278, Community Bureau of
Reference, Table 29). Detection limits for Zn, Cu, Cd, and Pb were respectively 0.002, 0.002,
0.001, 0.014 µg of metal per ml of digested sample. Calibration was carried out by adding
known metal doses to matrix standards.
Table 29. Certified (mean value ± 95% CI) and measured metal (min and max values, µg g -1 DW, n=43) and
PCB concentrations (mean ± sd, ng g -1 total lipids, n=4) of certified reference materials (Mytilus edulis tissues,
CRM n°278, BCR for metals and mackerel oil; CRM n°350, BCR for PCBs).
PCB analyses
Contaminant Certified value Measured
Cd 0.35 ± 0.01 0.33 - 0.39
Cu 9.45 ± 0.13 8.62 - 10.7
Pb 2.00 ± 0.04 1.88 - 2.21
Zn 83.1 ± 0.10 81.1 - 90.1
PCB 52 62.0 ± 9.00 64.6 ± 0.90
PCB 101 164 ± 9.00 197 ± 5.70
PCB 118 142 ± 20.0 119 ± 2.60
PCB 138 274 ± 27.0 274 ± 8.30
PCB 153 317 ± 20.0 311 ± 9.40
PCB 180 73.0 ± 13.0 68.1 ± 1.80
All solvents and reagents were pesticide grade. PCB congeners and reference material were
bought from Promochem (Germany). Samples were insufficient to analyze PCBs in sea stars
from S01 and in the body wall for sea stars from station 230.
Part of the frozen sample was taken and dried overnight at 103°C to deduce the dry weight.
The remaining sub-sample was homogenized and a surrogate (PCB 103) was added. Samples
were then extracted three times with a modified Bligh and Dyer method (MeOH: CH 2Cl 2:
H 2O, 10:5:4) (Booij & Van den Berg, 1994). The organic phase was separated and the
aqueous phase was extracted three times with dichloromethane. Dichloromethane extracts
were then pooled. A fraction of the extracts was used to determine the total lipid content
(gravimetrically). The remaining part was evaporated under nitrogen and 3 ml of isooctane
145

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
were added to each sample. Lipids were eliminated by precipitation with sulfuric acid (three
times). A final purification on Florisil® was performed (40 ml hexane + 20 ml
hexane:CH 2Cl 2, 95:5). The eluate was evaporated, 10 µl of internal standard (PCB 155) was
added to 70 µl of the sample and 2 µl of this solution was injected into the GC/MS.
Samples were analyzed using a Finnigan GC/MS GCQ equipped with an AS9000 auto
sampler and a CP-Sil 8 capillary column (50m length, 0,25mm id, 0,25 µm film thickness).
Initial temperature of the column was 90°C. Temperature program was: (i) increase to 180°C
at 15°C/min and hold for 6 min, (ii) increase to 220°C at 4°C/min and hold for 2 min and (iii)
increase to 275°C at 5°C/min. The carrier gas was helium, at a flux rate of 30 cm sec -1 .
Injection mode was “splitless”. Mass spectra were acquired in electron impact in mode
“multiple reaction monitoring”.
Concentration of congeners #52, 101, 118, 153, 138 and 180 were measured in the different
body compartments of the asteroids. Accuracy of the method was tested using certified
material reference from the BCR (mackerel oil; CRM n°350) (Table 29). It was assumed that
the loss was the same for the surrogate (PCB 103) as for the other analyzed congeners, thus
the PCB concentrations were corrected accordingly.
Measure of Reactive Oxygen Species (ROS) production
Amoebocytes ROS production was measured by the peroxidase, luminol – enhanced method
developed by Coteur et al. (2002). Briefly, 3 ml of coelomic fluid were collected in the same
volume of anticoagulant buffer. The cell concentration was measured by absorbance at 280
nm using a Tecan Spectrafluor+ plate reader. The cell suspension was then centrifuged and
resuspended in Ca 2+ , Mg 2+ -free artificial seawater (ASW) to obtain a final amoebocyte
concentration of 10 6 cells ml -1 . A stock solution of luminol and horseradish peroxidase (HRP)
in DMSO was then freshly diluted 100 fold in ASW (final concentrations of HRP and luminol
were respectively 5x10 -1 mg ml -1 and 2.5x10 -1 mg ml -1 ). The reaction was started by adding
200 µl of amoebocyte suspension in 100 µl of luminol/HRP solution and 20 µl of a
Micrococcus luteus suspension (Sigma, 2.5x10 9 bacteria ml -1 ; stimulated ROS production) or
20 µl of ASW (unstimulated ROS production). The chemiluminescence was measured every
10 min over a 2 h period using a Tecan Spectrafluor+ plate reader. During this period, the
plates were placed at 14°C. Results were expressed as the sum of all 10 min interval
measurement for 10 6 cells ml -1 (total chemiluminescence) for stimulated amoebocytes (in the
presence of bacteria) or unstimulated amoebocytes (without bacteria).
146

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
Cytochrome P450 immunopositive protein quantification
Cytochrome P450 immunopositive protein (CYP1A IPP) induction was quantified using
competitive ELISA as described by Danis et al. (Chap. III.4). The ELISA was carried out
using competition for anti-trout CYP1A antibodies between the CYP1A IPP contained in the
sea star samples and a biotinylated CYP1A from ß-naphtoflavone (BNF)-injected trouts
(Oncorhyncus mykiss). Multiwell plates were coated with Anti-CYP1A (rabbit anti-fish
CYP1A peptide, polyclonal antibody purchased from Biosense, Norway). Wells were washed
with phosphate-buffered saline (PBS, Sigma), and nonspecific binding sites were blocked
with PBS containing bovine serum albumin (BSA, Sigma). Wells were washed again, and
were added with biotinylated microsomes from BNF-injected trouts (except for the blank
wells). Samples or standards (with adjusted protein concentration) were then added to the
wells. Competition was allowed to take place and after five washing steps, Extravidin-HRP
(Sigma) was added to all the wells. The plate was then incubated and the wells were washed
again using PBS. Chromogen TMB (Pierce, UK) was added to all the wells and the plate was
incubated in obscurity. Sulfuric acid (Sigma) was added to the wells to stop the reaction.
Optical density was measured in the 96-well plates at 450 nm using a Packard Spectracount
microplate reader.
Data analyses
Statistical differences between measured concentrations in different asteroid body
compartments and sampling stations were performed using 1-way analysis of variance
(ANOVA) followed by a multiple comparison test of the means (Tukey test) (Zar 1996).
Relationships between the various factors (measured concentrations and biological effects)
were tested using simple linear regression procedures (Zar 1996). The level of significance
for statistical analyses was always set at a = 0.05.
RESULTS
Raw data are accessible at: http://www.mumm.ac.be/datacentre/Databases/IDOD/index.php
Heavy metal concentration in asteroids
Concentrations of four heavy metals (Zn, Cd, Cu and Pb) were measured in sea stars sampled
in the Southern Bight of the North Sea. Two body compartments were considered for analysis
(body wall and pyloric caeca). No significant differences were found between stations
regarding Cu concentrations measured in the body wall (p ANOVA=0.31, mean concentration for
all stations was 1.65±0.27µg Cu g -1 ). Significant differences were found for the other metals
147

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
(p ANOVA≤0.0001). Metal concentrations measured in sea star body wall are presented in Fig.
37.
Figure 37. Asterias rubens. Concentrations (µg g -1 ; mean+SD; n=5) of heavy metals (Zn, Cu, Cd, Pb) measured
in the body wall of sea stars sampled in the southern North Sea. Histogram bars sharing the same superscript do
not differ significantly. Cu concentrations do not differ significantly between stations.
Zn concentration in the body wall separated sea stars from Breskens and Knokke (displaying
higher concentrations) from those sampled in Scharendijke (intermediary concentrations) and
the remaining stations. Regarding Cd concentrations measured in the same body
compartment, sea stars from Breskens was significantly more contaminated than all the other
stations (p ANOVA≤0.0001). Sea stars from Breskens also displayed the highest Pb
concentrations in the body wall, followed by those from an intermediary group (Oostende,
230 and Knokke) and by the remaining stations. Heavy metal concentrations measured in
pyloric caeca significantly differed according to the sampling station (p ANOVA≤0.0001; Fig 38).
148

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
Figure 38. Asterias rubens. Concentrations (µg g -1 ; mean+SD; n=5) of heavy metals (Zn, Cu, Cd, Pb) measured
in the pyloric caeca of sea stars sampled in the southern North Sea. Histogram bars sharing the same superscript
do not differ significantly.
Ranking of the stations according to metal concentrations measured in sea star tissues varied
from one metal to another. For Zn, three groups of stations were formed: the most
contaminated group included Breskens and Knokke, the least contaminated group included
S01 and Scharendijke, while sea stars from the remaining stations displayed lower Zn
concentrations (p ANOVA≤0.0001). For Cd, sea stars sampled in 230 and Nieuwpoort displayed
the highest contaminant levels, followed by sea stars from Knokke, Oostende, and Breskens
and lower levels were determined in the remaining stations. For Cu, three groups of stations
can be distinguished: the most contaminated sea stars were smapled in Breskens and
Scharendijke, intermediate concentrations were found in sea stars from Knokke, S01 and 710
and the less contaminated were stations GC1, Oostende, Nieuwpoort, 340 and 230. For Pb,
the three groups were respectively [Oostende, Knokke and Breskens], [S01, 710, 230 and
Nieuwpoort] and [Scharendijke, GC1 and 340].
PCB concentrations in asteroids
The concentrations of PCBs measured in the two considered body compartments are
presented in Fig 39.
149

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
Figure 39. Asterias rubens. Concentration (ng g -1 total lipids; mean+SD; n=3) of PCB 153 and the sum of 6 PCB
congeners measured in the pyloric caeca and body wall of sea stars sampled in the southern North Sea.
Histogram bars sharing the same superscript do not differ significantly.
Significant differences were found between the stations when considering PCB
concentrations measured in both body compartments. However, the pyloric caeca
(p ANOVA≤0.0001) were found to be more discriminant than the body wall (p ANOVA≤0.00067)
When considering the sum of PCBs in the pyloric caeca, sea stars sampled in Knokke
displayed the highest concentrations. Sea stars from Breskens and 230 were also among the
most contaminated, while lower PCB concentration were measured in sea stars from the
remaining stations. As expected, a similar ranking, although less discriminant, was
established when considering PCB 153 concentrations alone. A strong and highly significant
regression was calculated between PCB 153 and remaining congeners (PCB 52, 101, 118,
138, 180, ∑ 5PCB) concentrations in body wall and pyloric caeca (Fig. 40). The contribution
of PCB 153 to the total burden of PCBs accounted for 59±16% in body wall, and 36±5.1% in
pyloric caeca.
150

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
A. Bodywall
PCB#153 concentration (ng g -1 lipids)
1200
1000
800
600
400
200
B. Pyloric caeca
PCB#153 concentration (ng g -1 lipids)
700
600
500
400
300
200
100
0
0
151
R 2 = 0,93
p≤0.0001
0 200 400 600 800 1000 1200 1400 1600
∑ 5PCB concentration (ng g -1 lipids)
R 2 = 0,83
p≤0.0001
0 100 200 300 400 500 600 700 800 900
∑ 5PCB concentration (ng g -1 lipids)
Figure 40. Asterias rubens. Linear regressions between PCB 153 and the
other considered PCB congeners (∑ 5PCB) concentrations (ng g -1 total
lipids) measured in the body wall (A) or pyloric caeca (B) of sea stars.
Reactive oxygen species (ROS) production
ROS production by amoebocytes is presented in Fig. 41. No significant difference between
the different sampling sites (p ANOVA=0.19) was found for ROS production of non-stimulated
amoebocytes . On the contrary, significant differences were found for ROS production by
bacteria-stimulated amoebocytes (p ANOVA=0.011). The highest values were measured in S01.
Sea stars from Scharendijke displayed the lowest response. Sea stars collected in all other
stations produced intermediate amounts of ROS. No significant correlation was found
between this biological effect and contaminant concentrations measured in sea stars.

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
Figure 41 Asterias rubens. ROS production (stimulated and non-stimulated; total chemiluminescence, Relative
Light Units (RLU) 10 -6 cells; mean+SD; n=5) measured in sea stars sampled in the southern North Sea.
Histogram bars sharing the same superscript do not differ significantly.
CYP1A immunopositive protein (CYP1A IPP) induction
CYP1A IPP induction was measured by competitive ELISA in sea stars collected in the
different stations (Fig. 42).
Figure 42. Asterias rubens. CYP1A IPP induction (induction fold; mean+SD; n=5) measured in sea stars
sampled in the North Sea using competitive ELISA. Histogram bars sharing the same superscript do not differ
significantly.
152

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
Significant differences (p ANOVA≤0.0001) were found between the different stations: sea stars
from 230, S01 and Knokke produced more CYP1A IPP. Intermediate values were measured
in sea stars from Breskens and Scharendijke, while the remaining stations displayed lower
CYP1A IPP induction values. Highly significant regressions (p≤0.0001 and p=0.002
respectively) were calculated between CYP1A induction and PCB 153 or ∑ 6PCB
concentrations measured in the pyloric caeca (Fig. 43). Maximal induction of CYP1A IPP
production reached 43 folds in sea stars sampled in station 230.
DISCUSSION
CYP1A IPP induction (Time fold)
60
50
40
30
20
10
R 2 = 0,44
p≤0.0001
∑PCB PCB#153
153
R 2 = 0,32
p=0.002
0
0 200 400 600 800 1000 1200 1400 1600
PCB concentration (ng g -1 lipids)
Figure 43. Asterias rubens. Linear regressions between CYP1A induction
(induction fold) and PCB 153 or ∑ 6PCB concentrations (ng g -1 total
lipids) measured in sea stars pyloric caeca.
Sea stars from Breskens and Knokke appeared to be the most contaminated by the four
considered metals. Sea stars from Oostende and 230 also showed significant levels of Pb and
Cd. Interestingly, all these stations are located in the plume of the Scheldt river which is
known to flow to the SW along the Belgian coast and then to turn back up to the NE in front
of Oostende (hydrodynamics models are available at http://www.mumm.ac.be/EN/Models/
Operational/Currents/index.php). The Scheldt river is known to be heavily contaminated by
heavy metals (Bayens 1998). In this context, it is surprising that sea stars from station S01, in
the very mouth of the Scheldt river, and those from stations 710 and GC1, located in the
plume, did not appear as much contaminated. Scharendijke was of particular interest, as sea
stars from this station displayed relatively high metallic levels in their body wall. Being
located in a closed estuary system, this station is not affected by tidal currents, allowing the

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
sedimentation of very fine particles which are enriched in Cu (Danis et al. Chap. IV.1). The
origin of Cu in this station is unclear.
PCB levels measured in sea stars also displayed significant differences between the sampling
stations. Again, the sea stars from Knokke, Breskens and station 230 showed the most
contaminated pyloric caeca. In S01, sea stars were collected in too low numbers to allow PCB
analysis.
Strong and highly significant regressions were found between the concentration of PCB 153
and the sum of other analysed congeners, and PCB 153 accounts for a significant part of the
total PCB burden. This suggests that the analysis of this sole congener could be sufficient to
screen PCB levels in biota, a proposal previously made by Atuma and collaborators (1996).
PCB 153 contribution to the total PCB burden was found to be dependent on the considered
body compartment, which is in agreement with previous experimental results (Danis et al.
Chap. III.2).
Contaminant levels measured in the present study were in the same order of magnitude as
those determined in previous field studies (Table 30). Previous studies were conducted at
broader scales, including wide areas of the North Sea and/or sampling over important
distances (Everaarts et al. 1998, den Besten et al. 2001, Coteur et al. 2003a).
Table 30. Comparisons among contaminants concentrations (ranges in sea stars) measured in the present study
and those reported in previous studies in the North Sea.
nm=not measured; PrS=Pre-spawning period PoS=Post-spawning period
Body
compartment
Zn
(µg g -1
DW)
Cd
(µg g -1
DW)
Cu
(µg g -1
DW)
Pb
(µg g -1
DW)
PCB
153
(ng g -1
lipids)
154
∑ 6PCB
(ng g -1
lipids)
Area of study PeriodReference
Pyloric
PrS
caeca 110-410 0.28-2.13 10.5-186 0.61-2.51 105-615 330-1700 Along Belgian coast
1 †
45.1-282 0.56-2.36 5.12-51.1 0.63-2.03 104-463 396-1700 Along Belgian coast PrS 2 ‡
Along Dutch and PoS
111-198 0.41-0.95 5.40-28.9 0.57-1.90 nm nm German coasts
3 §
Southern and central PrS
nm nm nm nm 40-1050 103-785 areas
4 **
Southern and central PoS
nm nm nm nm 25-290 73-2175 areas
4
nm 0.2-3 7.9-52.1 0.7-3.6 41-1054 105-2804 Central region PrS 5 ††
nm 0.2-2.1 6.8-24.2 0.3-1.3 13-314 101-1000 Central region PoS 5
† Present study
‡ Danis et al. Chap. IV.1
§ Coteur et al. 2003a
** Everaarts et al. 1998
†† den Besten et al. 2001

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
Body wall 131-525 0.16-2.30 1.36-2.26 0.63-2.35 116-642 395-1550 Along Belgian coast PrS 1
Along Belgian coast PrS
91-210 0.22-0.51 0.04-1.45 0.26-1.43 118-445 430-1480 and Scheldt river
2
Along Dutch and PoS
115-190 0.26-0.41 1.22-2.09 0.71-1.81 155-534 525-1470 German coasts
3
It is noteworthy that a similar range of concentrations was found in the present small scale
study. This suggests that small scale patterns of contamination occur and can blur general
trends due to main contamination sources like discharges from a contaminated river.
Examples from stations 230, 710 and GC1 in the present study illustrated much small scale
differences. Large scale studiues should take such aspects into account.
In order to assess the health status of sea stars, the main immune response of these organisms
(ROS production by amoebocytes) was measured in collected individuals. Significant
differences were found between stations regarding ROS production by bacteria-stimulated
amoebocytes. Immunomodulation by xenobiotics has already been shown in echinoderms
(Coteur et al. 2002a,c, 2003a,c), and is generally thought to lead to altered defence against
infections (Livingstone et al. 2000). No significant correlation was measured between this
biological effect and the various contaminant levels measured in sea stars. Coteur and
collaborators (2003a) assessed contaminants effects on ROS production by sea stars immune
cells in a large scale study, carried out along the Dutch and German coasts up to the mouth of
the Elbe river, and along a transect from the Elbe’s mouth towards the centre of the North
Sea. This study showed an immunodepression of sea stars from stations located in the mouth
of the Elbe river and in front of the North Sea Canal. No immunodepression was observed in
the present study, in stations similarly located in the mouth of a laarge river (viz. Scheldt).
This discripancy can be attributed to the different scale at which these studies were carried out
and to probable differences in contaminants patterns, leading to differing immune response in
sea stars.
Measures of CYP1A IPP induction in the pyloric caeca displayed significant variations in the
sampling area. Highly significant regressions were found between CYP1A IPP induction and
∑ 6PCB or PCB 153 concentrations measured in sea stars pyloric caeca. CYP1A is known to
be inductible by a limited range of highly toxic contaminants, such as dioxins, furans, or
coplanar PCBs, which are also ubiquitously present in the environment (Stegeman et al.
2001). In the present work, coplanar PCB concentrations were not measured, but their
presence is generally considered to be linked to that of other, less toxic congeners (Metcalfe
1994). This probably explains the relatively low (though not negligible) values of the
155

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
determination coefficients calculated for these regressions. Previous large-scale studies by
den Besten et al. (2001) have considered total cytochrome P450 levels measurements to
assess the effects of environmental contamination on the production of this enzyme by sea
stars. Among cytochrome P450 isoforms, CYP1A is the only one that is known to be related
to dioxin-like compounds exposition. Den Besten and collaborators (2001) measured total
cytochrome P450 induction reaching 1.89 in a large scale situation. In the present study,
maximal induction reached 43 time folds between station 230, and station 340, showing a
high sensitivity of the ELISA over the commonly used CO-difference spectrum of dithionite
reduced samples approach. It is noteworthy that the highest induction value of CYP1A IPP
was found between two closely located stations, emphasizing the importance of variations
occuring at relatively short distance (22.6 km). High induction of CYP1A IPP can potentially
lead to adverse effects in sea stars, as this enzyme system is responsible for the convertion of
insoluble organic compounds to more soluble metabolites, which are often more toxic than
the parent compounds (Walker & Peterson 1994). The products resulting from CYP1A
enzymatic activities also undertake the oxidation of endogenous substrates, eg fatty acids and
steroid hydroxylation (Nelson et al. 1996). High levels of CYP1A can therefore disrupt the
endocrine system of sea stars, leading to dysfunctionning in the reproduction and
development, which can in turn affect populations perennity and benthic ecosystem
structuration (Bucheli & Fent 1995).
In conclusion, the general contamination status of the Belgian coast and Scheldt mouth was
very much comparable to that measured in other parts of the North Sea, reaching similar, and
sometimes higher levels in sea stars. The present study highlights the existence of complex
patterns of contamination occuring at small scale which can blur obvious sources, such as
river Scheldt contribution. Analyzing the contaminants in sea stars allowed to take into
account all the potential contamination sources (viz. sediments, seawater and food) and their
combined effects on two essential aspects of sea star biology (ROS production and CYP1A
induction), in field conditions. These aspects were found to be affected in the considered
region, which can potentially lead to a serious threat to North Sea benthic ecosystems.
AKNOWLEDGEMENTS
Research supported by a Belgian Federal Research Program (SSTC, Contract MN/11/30).
Grateful thanks are due to S. Dutrieux, D. Gillan, G. Radenac and R. Morgan for helpful
assistance in offshore and sea shore samplings. M. Warnau was and Ph. Dubois is Research
Associate, and G. Coteur Senior Research Assistant of the National Fund for Scientific
156

Bioaccumulation and effects of PCBs and heavy metals in the sea star Asterias rubens from the North Sea
Research (NFSR, Belgium). We thank the captain and crew of RV Belgica for their very kind
and efficient assistance and the Modelling Unit of the Mathematical Model of the North Sea
(MUMM, A. Pollentier) for granting ship time. Contribution of the "Centre Interuniversitaire
de Biologie Marine" (CIBIM).
157

158

Echinoderms as bioindicators of sediment-associated metals and PCBs
IV.3 Echinoderms as bioindicators, bioassays and impact assessment tools of
sediment-associated metals and PCBs in the North Sea.
Archives of Environmental Contamination and Toxicology 45:190-202
Coteur G a , Gosselin P b , Wantier P c , Chambost-Manciet Y a , Danis B a , Pernet Ph a Warnau M a
&Dubois Ph a
a. Laboratoire de Biologie Marine (CP 160/15), Université Libre de Bruxelles, 50, Av. F. D.
Roosevelt, B-1050 Bruxelles, Belgium.
b. Laboratoire de Biologie Marine, Université de Mons-Hainaut, Av. du Champs de Mars,
Pentagone, 7000 Mons.
c. Laboratoire de Chimie Organique, Université de Mons hainaut, Av. Maistriau 19, 7000
Mons.
159

ABSTRACT
Echinoderms as bioindicators of sediment-associated metals and PCBs
The study assessed the occurrence, possible toxicity and impact of sediment-associated metals
and PCBs in the coastal zone of the southern North Sea using echinoderms as representatives
of the macrobenthos. Metals and PCBs were analysed in the sediments and in the body
compartments of the starfish Asterias rubens from 11 stations. The general toxicity of
sediment-associated contaminants was assessed by bioassays using embryonic and larval
developments of both A. rubens and the sea urchin Psammechinus miliaris. The impact of
contamination was assessed by measuring cellular immune responses of A. rubens collected
in the same stations.
Contamination of the starfish by metals and PCBs closely reflected that of the sediments.
However, bioaccumulation was element-specific for metals and depended on the chlorination
pattern for PCBs. The sediment-associated contaminants appeared to be toxic in both the A.
rubens and P. miliaris developmental assays. Moreover, metals were shown to affect the
immune responses of starfishes living in contaminated stations. The most significant effects
on biological responses were recorded in the plumes of the Scheldt/Rhine/North Sea Canal
and the Elbe/Weser rivers.
KEYWORDS
North Sea, contamination, heavy metals, PCBs, echinoderms, immunity, early development.
160

INTRODUCTION
Echinoderms as bioindicators of sediment-associated metals and PCBs
The North Sea is one of the most productive bodies of water in the world. The benthos is an
important part of this ecosystem and can be considered as a reservoir of biodiversity and of
economic resources (NSTF 1993). Moreover, a part of the demersal species depends on the
benthos for their food source. The North Sea is also the catchment basin of fluvial waters
from several important and highly polluted rivers. The dumping of dredge material and the
atmospheric deposition bring additional inputs of contaminants (NSTF 1993). When entering
the North Sea, contaminants of major concern such as heavy metals and polychlorinated
biphenyls (PCBs) tend to adsorb on particulate matter and to precipitate due to their low
solubility in seawater (NSTF 1993a). As a consequence, the benthic component represents a
main target of anthropogenic contaminants.
Echinoderms comprise numerous species that dominate quantitatively and/or qualitatively the
benthos and thus play a structuring role in this habitat (Saier 2001). Alterations of their
populations will thus affect the whole community and threaten the equilibrium of the
ecosystem where they live. Some echinoderms were shown to be valuable indicators of
contamination since they accumulate metals or PCBs as a function of the contamination level
of the environment (Knickmeyer et al. 1992, Temara et al. 1998b, Warnau et al. 1998,
Schweitzer et al. 2000). The embryonic and larval developments of sea urchins are regular
toxicity assays in monitoring and risk assessment programmes (Environmental Canada 1992,
Warnau et al. 1996a). The effects of contamination on echinoderms have been addressed
mainly by the use of molecular responses as biomarkers of contamination (e.g., Everaarts et
al. 1998, Den Besten et al. 2001). For instance, the acetylcholinesterase activity and the DNA
integrity in starfishes were lower in stations located near important estuaries such as that of
the Elbe or the Rhine river (Everaarts et al. 1998, Den Besten et al. 2001). However, little is
known on the impact of contamination in the field at the cellular and individual level. Thus,
the risk of contamination for the health of these essential species is poorly investigated.
The present study is an integrated approach of sediment-associated contaminants in the
southern North Sea. Echinoderms were used as bioindicators of contamination by metals and
PCBs and as bioassays for their possible toxicity. Moreover, the effect of metals and PCBs on
the health of a benthos-structuring echinoderm was assessed.
Environmental contamination by metals and PCBs was determined by analyzing these
contaminants in the sediments and in the starfish Asterias rubens. This species lives in close
contact with the sediments (especially in North Sea subtidal habitats) and is a key-species in
the North Sea benthic ecosystem (Saier 2001). These animals were shown to accumulate
161

Echinoderms as bioindicators of sediment-associated metals and PCBs
contaminants via seawater, food and sediments (Temara et al. 1996, Warnau et al. 1999,
Danis et al. Chap. III.2). Moreover, this starfish was shown to be an efficient bioindicator of
these contaminants (Knickmeyer et al. 1992, Temara et al. 1998b).
Toxicity of the sediment-associated contaminants was assessed using embryos and larvae of
both the starfish A. rubens and the sea urchin Psammechinus miliaris as bioassays. Indeed,
both sea urchin and starfish embryonic stages (from fertilization to gastrula) are particularly
sensitive to the presence of metals and PCBs (Pagano et al. 1985, Den Besten 1989,
Kobayashi 1995).
The health of A. rubens collected in the stations was assessed through the activity of its
immune system. This was chosen considering the importance of this system for the survival
of individuals (and thus the permanence of echinoderm populations) in a pathogen-rich
environment such as the North Sea (Billen et al. 1990). One of the contaminant toxicity
pathways could occur via the alteration of the immune system and the subsequent onset of
infections. The main immune responses of echinoderms are the phagocytosis (ingestion of
non-self material by effector cells) and the production of reactive oxygen species (ROS,
which represent a cytotoxic mechanism participating in the destruction of micro-organisms)
(Chia and Xing 1996, Gross et al. 1999). These immune responses are carried out by the only
type of circulating immune cells in A. rubens, the amoebocyte, and have been well
characterized in previous studies (Vanden Bossche and Jangoux 1976, Coteur et al. 2002a,b).
MATERIALS AND METHODS
Sampling stations
Sampling activities were carried out during cruise 9910 of Oceanographic R.V. Belgica from
April 19 to 23, 1999. The sampling sites were distributed regularly along the Dutch and
German coasts (LN stations) at similar distances from the coast in order to obtain more or less
similar environmental conditions (coastal areas with fine sediments) (Fig. 44).
162

Echinoderms as bioindicators of sediment-associated metals and PCBs
Figure 44. Map of the southern North Sea showing the location of the sampling
stations. Arrows indicate the river mouths.
Two stations (BW5 and BW6) are located more offshore. The latter ones, as well as BW3, are
part of a transect describing a gradient of contamination starting from the mouth of the rivers
Elbe and Weser and extending offshore (Stebbing & Dethlefsen 1992). These BW stations are
actually located along the plume of the Elbe/Weser which extends at right angle to the coast
(Becker et al. 1992).
Collection and preparation of samples
At each station, a number of environmental variables were measured using a SCTD device
(Seacat SBE19). These variables include depth, bottom water temperature, oxygen, and
salinity (Table 31). At each station, six sediment samples were collected using a Reineck box
corer. The upper layer (top 5cm) from each core was collected and two sub-samples placed,
respectively, in 400ml and 50ml hermetically-sealed polyethylene (PET) containers. These
containers were stored at –20°C. Sediment samples from the 50ml containers were used for
the early development assays and those from the 400ml containers were used for metal and
PCB analyses.
163

Table 31. Environmental parameters measured at each station.
Echinoderms as bioindicators of sediment-associated metals and PCBs
Parameters LN1 LN2 LN3 LN4 LN5 LN6 LN7 LN8 BW3 BW5 BW6
Depth (m) 25 25 27 26 26 24 22 25 30 39 40
Water
- temperature (°C) 9.1 8.8 8.8 8.6 8.6 8.5 8.4 7.8 7.3 7.3 7.1
- oxygen (mg l -
1 )
13.0 11.9 11.5 10.5 10.6 10.5 10.9 10.8 9.7 10.2 9.5
- salinity (psu) 34.4 32.4 31.6 31.6 31.4 32.5 34.1 34.4 34.2 33.9 32.8
Fraction

Echinoderms as bioindicators of sediment-associated metals and PCBs
water (Millipore), and filtered on glass microfiber filters (Whatman GF/A). Metal
concentrations were determined by atomic emission spectrometry using a Jobin-Yvon 38+ Ion
Charged Plasma Spectrometer (ICPS). Accuracy of the method was tested using certified
reference material (Mytilus edulis tissues, CRM n°278, Community Bureau of Reference).
Analysed CRM were always within 12% of the mean certified values. Detection limits for Zn,
Cu, Cd, and Pb were respectively 0.002, 0.002, 0.001, 0.014µg of metal per ml of digested
sample. The emission intensities were converted to concentrations (µg g -1 ) using a calibration
curve obtained by standard-addition over an internal reference matrix prepared for each
sample type (sediments, pyloric caeca and body wall).
PCB analyses
Seven polychlorinated biphenyls recommended by the ICES (International Council for
Exploration of the Sea) (viz., IUPAC congeners #28, 52, 101, 118, 138, 153 and 180) were
analysed in sediments and in the starfish body wall. Extraction of the samples was carried out
by the combination of three solvents: dichloromethane, methanol and water according to the
method of Booij & Van den Berg (1994). This technique is known to efficiently extract PCBs
and total lipids. Purification of the extracts was carried out on a sodium sulphate, aluminium
oxide and acid silica gel column. For sediments, a purification step with mercury was carried
out in order to remove sulphurs (Goerlitz and Law 1971). Samples were analysed with a
GC/MS GCQ (Finnigan) equipped with an auto-sampler AS9000 and a capillary column CP-
Sil 8 (50m length, 0.25mm id, 0.25µm film thickness). Initial temperature of the column was
90°C, then increased to 180°C at 15°C/min and held for 6min, then to 220°C at 4°C/min and
held for 2min and finally increased to 275°C at 5°C/min. The carrier gas used was helium at a
flow rate of 30cm/sec. Injection mode was “splitless”. Mass spectra were acquired in electron
impact mode “multiple reaction monitoring”. Accuracy of the method was tested using
certified material reference from the BCR (sediments from the harbour “Nova Scotian” in
East of Canada and mackerel oil -CRM n°350-). Analysed CRM were always within 15% of
the mean certified values (except for congener 101 analyses that fell within 30% of the mean
certified value). The detection limits were between 0.01 and 0.1ng g -1 DW, depending on the
PCB congener.
Early development assays
The sediment samples were dried at 70°C for 72h. The seawater used for these assays was
collected in the intertidal zone at Ambleteuse (France), stored at 12-14°C and filtered on
0.22µm mesh-size before use.
165

Echinoderms as bioindicators of sediment-associated metals and PCBs
Genitor specimens of Asterias rubens and Psammechinus miliaris were collected in the
intertidal zone at Ambleteuse (Pas-de-Calais, France) and Luc-sur-Mer (Normandie, France),
respectively. They were maintained in separate aquaria at the laboratory in Mons (33‰
salinity, 14°C temperature) and fed ad libitum. Spawning was induced by injection into the
general body cavity of 10µl/g body weight of a 100µM 1-methyladenine solution (for A.
rubens, Strathmann 1987) or 20µl/g body weight of a 0.5M KCl solution (for P. miliaris,
Gosselin & Jangoux 1998). Sperm from three males were pooled before their use for
fertilization. Eggs from each female were distributed in –at least- one plastic Petri dish (15cm
diameter) filled with seawater. Fertilization started by adding 30µl of sperm from the same
male pool to the Petri dishes. The fertilization quality was assessed by (1) the presence of the
fertilization membrane in at least 70% of the eggs after 1h at 14°C and (2) the success of the
first embryonic cleavage after 3h at 14°C. Zygotes from 3 to 5 females were mixed before use
in the early development assay.
All assays were conducted simultaneously using the same mixed pooled of zygotes. The
assays were performed in Falcon six-well plates. Two plates were used for each sampling
station, one per studied species. For every station, 0.1g of the six sediment samples was added
to a corresponding well. Then, 10ml of seawater and 250-300 embryos were added in each
well (final sediment concentration: 1% w:v). Control experiments were carried out by
omitting the addition of sediments. Control experiments resulted in 67.4 ± 9.6% (mean ± sd,
n=6) and 89.4 ± 3.7% (mean ± sd, n=6) of normal larvae of A. rubens and P. miliaris,
respectively. Plates with developing embryos were incubated at 14°C for 72h. Offspring
quality, expressed as percentage of normal larvae, was assessed by examination of 100 larvae
per well using an Olympus TO41 inverted light microscope. The morphological criteria used
to distinguish normal larvae from abnormal or retarded larvae are adapted from Warnau and
Pagano (1994). Normal larvae are kidney-shaped (A. rubens) or cone-shaped (P. miliaris). A
complete archenteron and pairs of coelomic pouches are present. P. miliaris larvae also
present two pairs of larval arms supported by skeletal rods. Abnormal larvae were affected by
anomalies in shape, skeleton and/or internal cavities; delayed larvae showed unaltered larval
morphology but were smaller (i.e. with a delayed development) than normal larvae.
Immune responses of Asterias rubens
Two essential parameters of the starfish immune system were considered, the coelomic
amoebocyte concentration (CAC) and the production of reactive oxygen species (ROS) by
amoebocytes. Amoebocytes ROS production was measured by the peroxidase,
luminol–enhanced chemiluminescence method developed by Coteur et al. (2002a). Briefly,
166

Echinoderms as bioindicators of sediment-associated metals and PCBs
3ml of coelomic fluid were collected in an equal volume of anticoagulant buffer. The cell
concentration was measured (by cell counting on a Thoma haemocytometer), allowing
calculation of the CAC. This suspension was then centrifuged and eventually resuspended in
Ca ++ , Mg ++ -free artificial seawater (CMFASW) to adjust amoebocyte concentration to 10 6
cells ml -1 . A stock solution of luminol and horseradish peroxidase (HRP) in DMSO was
freshly diluted 100 fold in artificial seawater (ASW, Sigma) (final concentrations of HRP and
luminol were, respectively, 5x10 -1 mg ml -1 and 2.5x10 -1 mg ml -1 ). The reaction started by
adding 400µl of amoebocyte suspension in 500µl of luminol / HRP solution and 100µl of a
Micrococcus luteus suspension (2.5x10 9 bacteria ml -1 ) in round bottom test tubes that were
subsequently placed at 13°C. The chemiluminescence was measured (integration time per
tube: 10sec) every 10min over a two-hour period on a Lumat LB 9507 Luminometer
(Berthold) equipped with a 100 fold attenuation filter. Results are expressed as the sum of all
10min interval measurement for 10 6 cells ml -1 (total chemiluminescence) for stimulated
amoebocytes.
Data access
Raw data used in this study are available in the IDOD database of the Management Unit of
the Mathematical Model of the North Sea (Royal Belgian Institute of Natural Sciences)
(http://www.mumm.ac.be/datacentre/) .
Statistical analyses
Differences in biological responses and contamination levels in the sediments and starfish
body compartments among sampling stations were tested by analysis of variance (ANOVA),
followed by the multiple comparison test of Tukey (Zar 1996). The level of significance was
set at a = 0.05. When using the ANOVA test on the percentage of normal larvae, an arcsin-
transformation of the data (x’=Arcsin[√x]) was carried out to ensure normality (Zar 1996).
The relationship between contamination, environmental and biological variables was studied
by multivariate statistics. In order to avoid highly intercorrelated variables (which would
introduce a bias in the factor analysis), hierarchical clustering (single linkage method using
Pearson distances) was realised on contamination and environmental variables (Manly 1986).
Variables showing distances below 0.2 (arbitrary limit) were grouped by summing their
standardised values for each station. These variables (grouped or not) together with biological
ones were then studied by factor analysis using the principal components method with an
extraction matrix based on Pearson correlation coefficients and the "varimax" method of
factor rotation (Manly 1986, Wilkinson 1988). Additionally, K-means clustering analysis was
used to show the grouping of the sampling stations according to the same variables (Manly
167

Echinoderms as bioindicators of sediment-associated metals and PCBs
1986, Wilkinson 1988). K-means clustering splits the set of stations into a selected number of
groups by maximizing between-cluster variation relative to within-cluster variation (it is
similar to doing a one-way analysis of variance where the groups are unknown and the largest
F value is sought by reassigning members to each group). All statistical analysis were carried
out using the Systat ® software (Wilkinson 1988).
RESULTS
Contamination level of the sampling sites
As shown in Table 1, the fraction BW5≈BW6) and in LN1, while the lower levels
were found in the stations LN3, LN5 and LN7 (Table 33). The discrimination of the stations
was less obvious with the metal concentrations in the

Echinoderms as bioindicators of sediment-associated metals and PCBs
SD 1.05 5.16 2.24 4.28 1.99 0.78 1.19 1.30 3.14 4.13 2.11
Pb Means 13.7 36.1 33.0 40.3 18.8 29.9 20.5 24.1 47.2 22.2 39.4
SD 12.6 9.0 15.7 32.4 6.5 6.2 10.2 10.6 3.0 7.5 4.6
Zn Means 29.3 71.7 40.2 66.9 25.5 70.6 38.8 49.2 104.5 62.8 73.7
SD 13.5 14.8 25.6 50.7 7.8 3.8 23.1 10.5 23.3 13.5 6.1
-: missing samples (lost during processing)
Table 33. Comparison of the metal concentrations, in the total and the 0.05, Tukey’s test).
Metals in sediments p
ANOVA
Total fraction
Cd

Echinoderms as bioindicators of sediment-associated metals and PCBs
Table 34. Heavy metal concentrations (means and standard deviations, n=10) in the body wall and the pyloric
caeca of the starfish Asterias rubens collected in Dutch and German stations in April 1999 (µg g -1 dry weight).
Metals in A. rubens LN1 LN2 LN3 LN4 LN5 LN6 LN7 LN8 BW3 BW5 BW6
Body wall
Cd Means 0.26 0.29 0.38 0.29 0.31 0.34 0.43 0.35 0.32 0.41 0.33
SD 0.11 0.11 0.17 0.13 0.12 0.10 0.13 0.14 0.13 0.06 0.12
Cu Means 2.00 1.59 2.03 1.45 1.78 1.69 1.29 1.53 1.45 1.22 2.09
SD 0.54 0.33 0.50 0.29 0.48 0.32 0.17 0.54 0.33 0.13 0.48
Pb Means 0.71 0.72 1.18 0.77 1.48 0.89 1.26 1.81 0.99 1.22 1.13
SD 0.27 0.21 0.44 0.20 0.54 0.26 0.66 1.68 0.39 0.27 0.32
Zn Means 115.3 129.1 124.2 118.6 124.7 146.8 168.7 129.4 167.6 190.2 168.2
SD 47.2 26.5 23.7 35.0 32.3 31.5 19.4 41.6 53.4 37.5 56.0
Pyloric caeca
Cd Means 0.59 0.53 0.33 0.60 0.43 0.62 0.56 0.81 0.95 0.41 0.77
SD 0.22 0.20 0.19 0.36 0.28 0.39 0.21 0.35 0.81 0.34 0.66
Cu Means 5.7 10.5 5.4 7.8 13.5 17.1 28.6 21.5 30.1 15.9 20.9
SD 2.1 6.3 2.8 4.1 6.5 7.1 23.7 5.9 15.8 5.3 14.1
Pb Means 0.57 0.78 1.45 0.97 1.90 1.20 1.53 1.65 1.33 1.83 1.38
SD 0.19 0.36 1.14 0.42 0.99 0.48 0.76 0.44 0.69 0.69 0.77
Zn Means 198.0 143.5 133.5 110.8 164.2 152.6 168.5 172.5 181.0 148.8 117.4
SD 39.0 34.0 22.3 33.5 62.7 37.6 25.9 25.6 63.5 72.0 37.7
The concentrations of metals in the starfish body wall were not significantly different among
the stations (except for the higher concentration of Pb in the body wall of starfishes collected
in LN8 compared to LN1, LN2 and LN4) (Table 35). The classification of the stations in
terms of metal concentrations in pyloric caeca varied from one metal to another. For Cd and
Cu, BW3 was characterized by the highest concentrations in contrast to the LN3 station. For
Pb, the contamination level was significantly higher in LN5 compared to LN1, LN2 and LN4.
For Zn, LN1 and BW3 were significantly more contaminated than LN4 (Table 35).
Table 35. Comparison of the metal concentrations, in the starfish body wall and pyloric caeca, measured in the
different stations. Stations which do not differ in terms of metal concentration are underlined (p>0.05, Tukey’s
test). NS, not significant.
Metals in A. rubens p
ANOVA
Body wall
Cd NS LN
7
Cu NS BW
6
BW
5
LN
3
170
LN
3
LN
1
LN
8
LN
5
Increasing contamination
BW
6
LN
6
LN
6
LN
2
BW
3
LN
8
LN
5
LN
4
LN
2
BW
3
LN
4
LN
7
LN
1
BW
5

Pb 0.006 LN
8
Zn 0.005 BW
5
Pyloric caeca
Cd 0.007 BW
3
Cu

Echinoderms as bioindicators of sediment-associated metals and PCBs
PCB 153 Means 4.86 - 1.71 6.56 0.88 2.86 6.07 1.52 6.78 3.45 2.19
SD - - 1.53 5.42 - 2.64 0.81 1.22 3.85 3.54 1.00
PCB 138 Means 6.61 - 1.97 9.99 0.92 3.69 7.46 1.69 7.75 3.97 2.92
SD - - 2.04 9.56 - 3.57 2.62 1.38 4.81 4.14 1.32
PCB 180 Means 2.55 - 0.69 6.73 0.47 1.59 3.33 0.49 1.80 0.98 0.67
SD - - 0.69 6.33 - 2.42 0.58 0.30 0.73 0.97 0.26
Body wall
PCB 28 Mean
s
87.6 68.6 50.5 51.7 68.4 62.1 - 50.7 66.5 73.6 53.5
SD 32.7 36.2 31.1 30.2 20.8 33.4 - 24.9 32.8 34.8 26.6
PCB 52 Mean
s
47.3 37.2 48.1 42.5 43.8 38.1 - 36.9 39.3 45.0 36.5
SD 19.7 5.2 39.4 10.7 11.1 20.8 - 22.2 9.4 13.8 17.4
PCB 101 Mean
s
88.9 92.2 74.6 122.5 75.3 98.6 - 64.1 73.7 62.0 50.9
SD 23.3 15.5 20.2 49.2 16.6 26.1 - 14.7 12.9 13.9 21.3
PCB 118 Mean
s
94.2 102.7 82.1 162.5 101.6 135.0 - 88.7 82.3 67.3 54.2
SD 39.7 32.9 14.6 79.0 26.5 54.2 - 7.4 25.7 19.2 22.8
PCB 153 Mean
s
307.0 371.8 288.0 533.6 275.9 368.4 - 275.2 331.1 198.5 155.0
SD 162.1 123.1 88.5 196.7 74.2 145.8 - 55.5 100.2 30.2 103.4
PCB 138 Mean
s
266.4 334.1 265.2 536.2 280.8 369.6 - 290.1 307.6 217.4 153.8
SD 138.5 113.6 74.8 225.8 80.4 155.6 - 59.1 84.9 52.4 79.6
PCB 180 Mean
s
11.2 14.3 20.4 23.1 11.3 17.3 - 10.3 15.3 10.9 21.0
SD 9.3 11.2 13.0 11.3 6.0 9.6 - 3.9 2.3 1.5 29.9
-: missing samples (not collected)
Table 37. Comparison of the PCB congener concentrations, in the total sediments and starfish body wall,
measured in the different stations. Stations which do not differ in terms of PCB congener concentrations are
underlined (p>0.05, Tukey’s test). NS, not significant. Stations with insufficient replicate samples were omitted.
PCB congener p
ANOVA
Total sediments
PCB 28 0.001 LN
7
PCB 52

PCB 153 NS BW
3
PCB 138 NS BW
3
PCB 180 NS LN
7
Body wall
PCB 28 NS
PCB 52 NS
PCB 101 0.010 LN
4
PCB 118 0.018 LN
4
PCB 153 0.015 LN
4
PCB 138 0.018 LN
4
PCB 180 NS LN
4
Echinoderms as bioindicators of sediment-associated metals and PCBs
LN
7
LN
7
BW
3
LN
6
LN
6
LN
2
LN
6
BW
6
Early development of A. rubens and P. miliaris
Embryos of both the starfish A. rubens and the sea urchin P. miliaris were exposed, during
their development, to sediments collected in the studied stations. For both species, a
significant reduction in the percentages of normal larvae was observed when embryos were
exposed to sediments of some stations (Fig. 2a and Table 8). This reduction was mainly due
to increased proportions of delayed larvae while the frequency of abnormal larvae remained
constant (data not shown). The development assays performed with P. miliaris larvae clearly
discriminated the LN stations, whose sediments induced no developmental alterations, from
the BW stations (in the plume of the Elbe and Weser rivers), whose sediments induced the
lowest proportions of normal larvae of all assays (Fig. 2a and Table 8). In contrast, the
development assay using A. rubens larvae indicated that the percentages of normal larvae
were significantly reduced by sediments of the LN2 station compared to three other LN
173
BW
5
BW
5
stations (LN1, LN7 and LN8) (Fig. 45a and Table 38).
LN
6
LN
2
LN
2
LN
6
LN
2
LN
3
LN
6
LN
6
BW
5
LN
1
LN
5
BW
3
BW
3
LN
6
BW
6
BW
6
LN
3
LN
5
LN
1
LN
1
LN
8
BW
3
LN
3
LN
3
BW
6
LN
3
LN
8
LN
3
LN
5
LN
2
LN
8
LN
8
LN
8
BW
3
BW
3
LN
5
LN
1
LN
5
LN
8
LN
3
LN
8
LN
3
LN
1
BW
5
BW
5
BW
5
BW
5
BW
5
BW
6
BW
6
BW
6
BW
6
LN
8

Echinoderms as bioindicators of sediment-associated metals and PCBs
Figure 45. Biological responses measured in Dutch and German stations. (a) Effects of sediments on the early
development of Asterias rubens and Psammechinus miliaris (means + sd, n=6). (b) Immune responses of
Asterias rubens: coelomic amoebocyte concentration (CAC) and bacteria-stimulated amoebocyte ROS
production (RLU: Relative Light Units) (means + sd, n=10).
Table 38. Comparison of the biological responses measured in the different stations. Stations which do not differ
in terms of biological response are underlined (p>0.05, Tukey’s test).
Test p
ANOVA
Embryotoxicity tests
A. rubens (% normal
larvae)
P. miliaris (% normal
larvae)
Immunotoxicity test
ROS production
(10 3 RLU / 10 6 cells ml -1 )
0.004 LN2 LN4 BW
3

Immune responses of Asterias rubens
Echinoderms as bioindicators of sediment-associated metals and PCBs
The starfishes collected in the different stations were assayed for two parameters of their
immune system: the concentration of amoebocytes in the coelomic fluid (CAC) and the
production of reactive oxygen species (ROS) by bacteria-stimulated amoebocytes (Fig. 45b).
The CAC did not differ significantly among starfishes from different stations (p ANOVA=0.437).
This seems to be due to the high variability commonly associated to this physiological
response. In contrast, the ROS production varied significantly among starfishes from the
studied stations (Table 38). The ROS production by starfishes collected in the BW stations (in
the Elbe/Weser zone) was the lowest and differed significantly from that measured in LN5
and LN6 (Table 38). Among the LN stations, the lowest ROS response was measured in
starfishes from LN3 which differed significantly from the ROS production in starfishes from
LN5 (Table 38).
Integrated data analysis
In total, 34 contamination and environmental variables were available for subsequent
analyses. However, such large data sets are difficult to interpret. Moreover, some of these
variables might be highly intercorrelated (R pearson > 0.8), thus introducing a bias in the
multivariate analyses. Consequently, a first approach was to determine highly correlated
variables by hierarchical clustering (Fig. 46). Variables were grouped when their Pearson's
distances were below 0.2. Six groups were apparent in this analysis: (1) metal levels in the

PB_A
R
MET_TOT_
MO
MET_INF6
3
TEMP_O
X
HIGH_P
CB
LOW_PC
B
PB_TEG
PB_CP
CD_TEG
ZN_TEG
CU_CP
CD_CP
CU_TOT
PB_TOT
ZN_TOT
CD_TOT
GRAN63
PB_INF63
ZN_INF63
CU_INF63
CD_INF63
P180_TEG
WAT_SAL
CU_TEG
WAT_OX
WAT_TEMP
P118_TEG
P138_TEG
P153_TEG
P101_TEG
P180_SED
P118_SED
P138_SED
P153_SED
P101_SED
P52_SED
P28_SED
P28_TEG
ZN_CP
P52_TEG
Echinoderms as bioindicators of sediment-associated metals and PCBs
0.0 0.1 0.2 0.3 0.4 0.5
Distances
Figure 46. Cluster tree of environmental and contamination variables. Variables showing Pearson’s distance
below 0.2 were grouped. Six groups were determined: metals levels in the

Echinoderms as bioindicators of sediment-associated metals and PCBs
Figure 47. Relationships between biological, environmental and contamination variables. All variables were
used in a factor analysis using the principal component method. For abbreviations, see legend of Fig. 3.
The first two factors accounted for 51.0% of the total variance of the data. A link between the
metal contamination levels of the total sediments and of the starfishes was observed. Indeed
the MET_TOT_MO variable and some of the variables concerning metal levels in A. rubens
organs were located along the X axis (representing the first factor; Fig. 47). However, this
relationship was highly dependent on the metal and the body compartment considered. The
copper concentration in the pyloric caeca (CU CP) was linked with the MET_TOT_MO
group whilst in the body wall (CU TEG) it was found in the opposite quadrant (Fig. 4). On
the contrary, the concentration of zinc in the body wall (ZN TEG) was located along the
MET_TOT_MO vector on the X axis whilst in the pyloric caeca (ZN CP) it was located
almost on the Y axis (Fig. 47). The PB_AR group was located close to the origin of both axes
indicating that other factors than those analysed in this study influence the accumulation of Pb
in A. rubens from the environment. The ROS production and the P. miliaris embryotoxicity
parameters were linked with the metal contamination cluster (viz. they are located on the
other end of the X axis compared to the metal contamination cluster; Fig. 47). The coelomic
amoebocyte concentration (CAC) was found along the metal contamination vectors, although
less clearly.
TEMP_OX
EMB_Pm
LOW_PCB
P52_TEG
PLCL_stim
P28_TEG
1
-1 1
CU_TEG
PB_AR CD_CP
HIGH_PCB
WAT_SAL
P180_TEG
-1
FAC
TOR
1
The Y axis (representing the second factor) comprised, at the negative end, the level of PCB
congener #180 (the most highly chlorinated congener studied) in the body wall of A. rubens
and, at the positive end, the level of PCB congener #28 (the most weakly chlorinated
177
ZN_CP
EMB_Ar
CAC
CD_TEG
MET_TOT_MO
MET_INF63
CU_CP
ZN_TEG

Echinoderms as bioindicators of sediment-associated metals and PCBs
congener studied) in the body wall of the starfish (Fig. 47). The LOW_PCB variable tended to
place itself on the positive part of this axis. The Y axis (factor 2) thus appears to describe
mostly the duality in the chlorination pattern of PCB congeners with respect to the
contamination of the stations. This interpretation is, however, less obvious than that of the X
axis (factor 1). As described above, the concentrations of highly chlorinated PCB congeners
in the sediments and in the starfish body wall were strongly intercorrelated (and gathered in
the HIGH_PCB group). Similarly, the concentrations of weakly chlorinated PCB congeners
(#28 and 52) in the starfish body wall were found in the same quadrant as the LOW_PCB
group (#28, 52 and 101 in the sediments) indicating a relationship between the levels of
weakly chlorinated PCB contamination of the sediment and of the starfishes (Fig. 4). The A.
rubens embryotoxicity parameter is found on the positive part of the Y axis, suggesting a link
with the contamination by PCB congener #180. The position of other variables such as the
bottom water characteristics, the zinc level in the pyloric caeca of A. rubens or the levels of
metals in the fine fraction of the sediments are uneasy to interpret.
An attempt was made to classify the different sampling stations according to the set of
variables used in the factor analysis (biological, environmental and contamination variables).
The K-means clustering method allows an objective grouping of these stations in a user-
defined number of clusters. This technique, using three groups, indicated that the three BW
stations separated from a cluster made up of the LN1 to LN5 stations while the last group
consisted in the LN6 to LN8 stations.
DISCUSSION
All four metals studied (Cd, Cu, Pb and Zn) showed a similar behaviour with respect to the
contamination of the sediments in the sampling stations. This may suggest that the differences
in physico-chemical properties of the four metals are not a determining factor for the
contamination pattern of the stations. The concentrations of metals in the total sediments are
also highly correlated with the percentage of the fine fraction in the sediments (used to
estimate the organic matter content). Indeed, this fraction is often the most contaminated due
to the large surface to volume ratio of fine sediment grains and to its increased organic matter
content. Thus, the load of metals in the total sediments will increase along with the percentage
of the fine fraction. The BW and LN1 stations showed the highest concentrations of metals in
the total sediments, probably reflecting the influence of, respectively, the Elbe/Weser and the
Scheldt rivers. The metal concentration values in the sediments collected in the BW stations
178

Echinoderms as bioindicators of sediment-associated metals and PCBs
match closely those reported previously in the same stations (Cofino et al. 1992) except for
Cd that was 10 fold more concentrated in the sediments collected in the present study.
The fact that metal concentrations in the sediments and in some of the starfish organs are
linked confirms the indicator value of A. rubens, the contamination of the starfishes reflecting
that of the environment. However, it remains unclear whether the link between sediment and
starfish contaminations reflects an accumulation of metals directly from the sediments or an
indirect relationship. The transfer of metals from the sediments to the starfish might occur via
the sediment interstitial waters that are often more contaminated than the water column
(Burgess et al. 1992). However, in our hands, this relationship varied from one metal to
another suggesting that the bioavailability of each metal is different. Moreover, each metal
appeared to have an organ-specific accumulation pattern, except Pb, which was uniformly
distributed in the pyloric caeca and body wall of A. rubens. Temara et al. (1997b) already
showed that the body compartment is the factor explaining the largest part of the variability
associated to metal concentrations in the starfish A. rubens.
The different polychlorinated biphenyl (PCB) congeners separated into two distinct groups
with respect to the contamination of the sediments. The level of chlorination was found to be
the distinctive feature of these two groups. It has been recently shown that the organic carbon
normalised partition coefficient (Kd) between sediment and water of PCB congeners
increased with the number of chlorine atoms in such a way that weakly chlorinated PCBs
have a lower affinity for sediments compared to highly chlorinated PCBs (Booij et al. 1997).
Moreover, according to the chlorine substitution pattern, the degradation and the transport of
PCB congeners operate differently, weakly chlorinated congeners being transported on a
larger geographical scale (Boon et al. 1985, Brown et al. 1987, Bedard and May 1996). These
differences in physico-chemical properties probably account for the grouping of the
congeners observed in the present study. The station located off the Ems river (i.e. LN7) was
heavily contaminated in weakly chlorinated PCB congeners while highly chlorinated PCB
congeners appeared uniformly distributed. This suggests that the sources of PCB
contamination which influence LN7 contain a higher proportion of weakly chlorinated
congeners. The concentrations of individual congeners measured in the sediments are
comprised in the concentration ranges previously reported for the North Sea (Skei 1994,
Laane et al. 1999).
The concentrations of PCB congeners in the starfish body wall could not be directly
compared to those found in the literature since the latter addressed the contamination of the
pyloric caeca or the whole body. A strong relationship was found between the concentrations
179

Echinoderms as bioindicators of sediment-associated metals and PCBs
of PCBs in the sediments and those in the body wall of the starfishes. Thus, A. rubens is able
to reflect accurately the PCB contamination level of the environment. Since PCBs are highly
hydrophobic and their concentration in seawater is very low (sediments constitute the main
reservoir of PCBs), the relationship between sediment and starfish contaminations suggests
that PCBs are directly accumulated from the sediments. This was previously shown in
laboratory experiments for the sediment-grazing sea urchin Lytechinus pictus (Schweitzer et
al. 2000). The accumulation of PCBs in the starfish body appears to be congener-dependent;
the correlation between sediments and starfish contaminations was strong with respect to the
highly chlorinated congeners, somewhat less for the weakly chlorinated congeners and almost
nonexistent for congener #180. The particular behaviour of congener #180 (the most highly
chlorinated congener studied) could be related to its substitution pattern (a 2,3,4,5-substitution
on one biphenyl ring) that was suggested to hinder its accumulation by starfishes
(Knickmeyer et al. 1992). The relationship between concentrations of weakly chlorinated
PCBs in the sediments and in the starfish body wall was less obvious compared to the highly
chlorinated congeners, indicating that the two groups of congeners are differentially
bioaccumulated (i.e. differentially taken up and/or differentially eliminated) by the starfish.
Similarly, Knickmeyer et al. (1992) compared the PCB congener patterns in the
"zooplankton" and in the starfish A. rubens in order to study the bioaccumulation of specific
congeners along the food chain. They found that highly chlorinated congeners (with the
exception of congener #180) were more efficiently accumulated in the starfish compared to
weakly chlorinated congeners. In eel (Anguilla anguilla), weakly chlorinated congeners were
more efficiently eliminated than highly chlorinated congeners (De Boer et al. 1994). These
results suggest either (or both) a more efficient uptake of highly chlorinated congeners or a
higher elimination (i.e. biotransformation) of weakly chlorinated congeners.
Biotransformation enzymes of the cytochrome P450 family have been demonstrated in A.
rubens (Den Besten 1998). Hence, both mechanisms could be responsible for the weaker
relationship between the levels of weakly chlorinated congeners in the sediments and in the
body wall of A. rubens.
The concentration of weakly chlorinated PCB congeners in the body wall did not differ
significantly among the stations. In contrast, the concentrations of highly chlorinated PCB
congeners in the starfish body wall were elevated in the LN4 station and lowest in the BW
stations. This suggests that the Rhine or the North Sea Canal are important sources of
bioavailable, highly chlorinated, congeners.
180

Echinoderms as bioindicators of sediment-associated metals and PCBs
The toxicity of sediments was studied using echinoderm developmental bioassays. The results
of these assays, performed with P. miliaris, are negatively related to metal concentrations in
the total fraction of the sediments (i.e. the fraction used in the assays). Heavy metals are
known to interact with fertilization process and embryonic development of echinoderms (for a
review, see Kobayashi 1995). In the present study, the highest metal concentrations in the
sediments were measured at the station off the Scheldt (LN1 station) and at the Elbe/Weser
stations (BW stations). Hence, these sediments form a reservoir of metals that appear to be
toxic towards echinoderm embryos and larvae. Contaminating concentrations were probably
much lower in the present study than those found to be embryotoxic in laboratory
experiments. However, the observed toxicity could result from an additive effect and/or a
synergy between the possible influences of different metals. It is indeed well documented that
combinations of metals are generally more toxic than the individual components (Kobayashi
and Fujinaga 1976, Pagano et al. 1996). In this study, embryonic development was never
arrested in the presence of sediments but only delayed. Kobayashi (1995) reported the same
effect when embryos were exposed to dissolved metals at very low concentrations.
The higher sensitivity of sea urchin embryos (compared to starfish embryos) to metals could
be partly explained by the differentiation of a larval skeleton (Warnau & Pagano 1994).
Unlike starfish larvae, sea urchin plutei develop a larval skeleton during their embryonic
development (Burke 1987). During this period, plutei are endotrophic and specifically take up
calcium from seawater for skeletogenesis. Heavy metals could be taken up along with
calcium. This is supported by the fact that some heavy metals compete with calcium fixation
in the skeleton (Pagano et al. 1982, Temara et al. 1996) resulting in skeletal abnormalities
(Pagano et al. 1982, 1986, Warnau & Pagano 1994). The results of developmental assays
performed with A. rubens seem to be linked to the level of PCB congener #180. As many
chemicals, PCBs have an inhibitory effect on echinoderm development at high concentrations
(Pagano et al. 1986, Den Besten et al. 1989). The developmental defects are generally similar
to those observed with heavy metals (Kobayashi 1995). Tests using A. rubens embryos do not
discriminate the stations as clearly as those using P. miliaris. However, the main effects were
observed in the stations off the Rhine/North Sea Canal (LN4 and, to a lesser extent, LN2)
where maximum PCB concentrations in the starfishes were measured. Further studies are
needed in order to confirm the impact of PCBs on the development of A. rubens in field
conditions.
The impact of the contamination on adult specimens of A. rubens were studied by measuring
their immune responses, considering the importance of these responses for the survival of
181

Echinoderms as bioindicators of sediment-associated metals and PCBs
these animals. The production of ROS by the amoebocytes of A. rubens was inhibited both in
the stations off the Elbe/Weser rivers (BW stations) and off the North Sea Canal (LN3).
Moreover, this immune response was linked with the levels of metals in the sediments and in
the starfish organs. The impact of marine contaminants on ROS production by invertebrate
immunocytes is highly variable from one study to another, ranging from almost total
suppression to significant stimulation or absence of effect (see Baier-Anderson & Anderson
2000 for a review). In flatfishes, the ROS produced by hepatocytes increased along an
increasing gradient of pollution in the German Bight (i.e. along BW stations) described during
the Bremerhaven Workshop (Moore 1992). In starfishes, the ROS production was enhanced
by contamination with dietary cadmium while it was inhibited in starfishes transferred from a
reference site to a heavily metal-contaminated site (Coteur 2002, Coteur et al. in press). The
effects of metals on this immune response might thus occur by complex, possibly interacting,
mechanisms instead of by a direct, target-specific action on the immune cells that would
hardly explain such variable effects.
The amoebocyte ROS production is enhanced by PCB contamination of sea urchins or
starfishes (Coteur et al. 2001; Danis et al. Chap. III.1, III.3, III.4). However, among the
different congeners tested by these authors, only coplanar (non-ortho-substituted) congeners
were effective. The absence of a correlation between A. rubens ROS production and PCB
contamination, as observed in this study, might thus reflect this congener specific action since
non-ortho-substituted PCBs were not measured.
The concentration of immunocytes in the body fluids has long been used for ecotoxicological
studies in invertebrates. In contrast to other cellular responses, the impact of xenobiotics on
the immunocyte concentration is generally constant and consists of an increase of this
parameter upon in vivo metal exposure (Coles et al. 1995) or upon complex in situ exposure
(Pipe et al. 1995). In echinoderms, the amoebocyte concentration was increased by in vivo
contamination with dissolved cadmium or by in situ exposure along a metal contamination
gradient (Coteur 2002, Coteur et al. 2003b). In contrast, in the present study, the CAC did not
differ significantly among starfishes from the different stations; neither did it correlate with
the metal or PCB contamination levels. However, the large variability associated to this
immune parameter as well as its sensitivity to other types of stress (Coteur et al. 2004)
probably prevents an obvious relationship with the contamination levels of the stations.
Previous work in the Southern North Sea by den Besten et al. (2001) showed that molecular
biomarkers in A. rubens discriminated the Elbe and Scheldt plumes in comparison with off-
shore stations in the central North Sea. However, these biomarkers did not differ significantly
182

Echinoderms as bioindicators of sediment-associated metals and PCBs
between coastal stations (either in the river plumes or not). On the contrary, the
developmental assays and immune responses, investigated in the present study, allowed the
distinction of sediment toxicity among the different coastal stations. Altogether, the biological
responses measured in this study characterized three areas: the area off the
Scheldt/Rhine/North Sea Canal and the area off the Elbe/Weser rivers, where biological
responses were affected, and the area located between the North Sea Canal and the Weser
plume, where no biological impact was found. However, each biological response was
differentially modulated by the contamination in these areas. The ROS production was clearly
affected in the stations located off the Elbe/Weser estuaries (BW stations) and, to a lesser
extent, in the station off the North Sea Canal (LN3 station). The early development of P.
miliaris was affected only in the Elbe/Weser zone. The A. rubens development was impaired
in the station located off the Rhine (LN2). It thus appears that the high contaminations in the
area off the Scheldt/Rhine/North Sea Canal and in the area off the Elbe/Weser rivers are
different in term of biological impact. An exclusive study of the contaminant concentrations
in these stations is therefore not sufficient to reflect efficiently the potential risk for
echinoderm populations. In contrast, when various biological responses are included in a
global analysis of the stations, it is possible to describe biologically deleterious
contaminations.
It is concluded that both metals and PCBs are bioavailable to the starfish A. rubens in the
considered area and that the contamination of the starfish closely reflects that of the
sediments. However, bioaccumulation is element-specific for metals and depends on the
chlorination pattern for PCBs. The sediment-associated contaminants appeared to be toxic in
both the A. rubens and P. miliaris developmental bioassays which showed a high sensitivity
as compared to molecular biomarkers in A. rubens. Moreover, metals were shown to affect
the immune responses of starfishes living in contaminated stations.
ACKNOWLEDGEMENTS
The authors wish to thank Prof. Pierre Rasmont for his helpful advice concerning multivariate
analysis and Mrs Mitra Razmjoo for reviewing the English writing. We also wish to thank the
captain and the crew of the A962 R.V. Belgica. G. Coteur and B. Danis are holders of a FRIA
doctoral grant. P. Gosselin and P. Wantier were researchers financed by the “Service des
affaires Scientifiques, Techniques et Culturelles” (SSTC) of the Belgian government. Ph.
Dubois and M. Warnau are, respectively, Research Associate and Honorary Research
183

Echinoderms as bioindicators of sediment-associated metals and PCBs
associate of the National Fund for Scientific Research (NSFR, Belgium). Research supported
by a Belgian federal research program (SSTC, contract MN/11/30). Contribution of the
“Centre Interuniversitaire de Biologie Marine” (CIBIM).
184

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
IV.4 Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and
sea stars of the intertidal zone in the southern North Sea and the Channel
Danis B a , Debacker V b , Trujilo Miranda C b & Dubois Ph a
a : Marine Biology Laboratory (CP 160/15), Free University of Brussels, 50 avenue F.D.
Roosevelt, B-1050 Bruxelles, Belgium
b : Laboratoire de Spectroscopie de Masse, Centre for Analysis of Residues in Traces
(CART), University of Liège B6c, B-4000 Liège, Belgium
185

ABSTRACT
Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
There is considerable concern regarding dioxin-like compounds (DLCs) in the marine
environment: these ubiquitous contaminants are highly resistant to degradation, highly
accumulated by marine organisms and extremely toxic. Concentrations of dioxin-like
compounds (DLCs) including 7 polychlorodibenzo-p-dioxins, 10 polychlorodibenzofurans
and 4 coplanar polychlorinated biphenyls were determined in sediments, mussels (Mytilus
edulis) and sea stars (Asterias rubens) from 5 intertidal stations distributed along the Belgian
coast and Channel. The induction of a biomarker, cytochrome P450 immunopositive protein
(CYP1A IPP), was also measured in sea star pyloric caeca. Although no significant
differences were found between the considered stations, DLC levels were found to be
relatively high in biota, especially when considering the toxicity of these compounds.
Particular concern arises from TEQ values determined in mussels from all locations. Sea stars
were found to be more discriminant between the stations. CYP1A IPP induction was found to
be significantly related to DLC levels measured in sea stars, and allowed a significant
discrimination between the considered stations.
KEYWORDS
PCDDs, PCDFs, PCBs, Asterias rubens, Mytilus edulis, sediments, North Sea
186

INTRODUCTION
Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
A wide variety of toxic compounds are known to be present in the marine environment, at the
global scale (Clark 1997). Some of these compounds have a natural origin while others result
exclusively from anthropogenic activity. These contaminants can be dissolved in water,
bound to sediments, or accumulated in marine organisms (Lohse 1991, Elijarrat et al. 1991).
The highest concentrations are usually found in the vicinity of harbours or estuaries, where
rivers drain toxicants to the sea. Some contaminants, because of their physico-chemical
properties, have become truly ubiquitous, and have been detected even in the most remote
locations, such as the polar regions, or the deep sea (Ballschmiter et al. 1997, Stegeman et al.
2001). Polychlorinated dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs) and biphenyls
(PCBs) are three families of such global contaminants. The most toxic member of these
families is the 2,3,7,8-tetrachloro-dibenzo-p-dioxin (2,3,7,8-TCDD). This component, and its
chemical relatives, are known to cause pleiotropic responses, at very low doses, in a range of
organisms (Payne et al. 1987, Ashley et al. 1996, Behnish et al. 2001). As a consequence,
there is now considerable public, politic and scientific concern regarding exposure to dioxin-
like compounds (DLCs), which include 7 PCDD, 10 PCDF and 12 PCB congeners (WHO
1999, US EPA 2000a,b). These compounds combine a high resistance to degradation
(biological, chemical, or physical), efficient accumulation by marine organisms and severe
toxicity. This results in serious health hazards for organisms and ecosystems, including man,
through consumption of contaminated seafood. Cytochrome P450 (CYP1A) induction is
usually considered as an early biomarker of contamination by DLCs (Bucheli & Fent 1995,
Stegeman 1995, Hahn 2002a). The main function of the CYP1A-dependent monooxygenase
system is to convert relatively insoluble organic compounds to soluble metabolites that are
eventually eliminated. However, the resulting products can be more toxic than the parent
compounds (Walker & Peterson 1994).
Mussels and sea stars are interesting test organisms because they are rather ubiquitous and
sedentary and play an important role in structuring several benthic ecosystems (Menge 1982,
Saier 2001). The blue mussel (Mytilus edulis) has been extensively used as a bioindicator
species in monitoring programmes (e.g. the Mussel Watch). This program began in 1986 to
monitor spatial and temporal trends of chemical contamination by chemically analyzing
mussels (and oysters) collected at fixed sites throughout the coastal United States (Goldberg
et al. 1978). Monitoring programmes using bivalves have also been adopted in Europe
(French “RNO” programme). The common sea star (Asterias rubens, L.) has been used as a
candidate species for monitoring a range of contaminants in the North Sea (Everaarts et al.
187

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
1998, den Besten et al. 2001, Coteur et al. 2003a, Stronkhorst et al. 2003). The effects of
contaminants on its reproductive success (den Besten et al. 1989), development (Kobayashi
1995, Coteur et al. 2003a), immune system (Coteur et al. 2003a,b, Danis et al. Chap. III.4),
and DNA integrity (Sarkar & Everaarts 1995) have been well studied. Available data on
mussels indicate that DLC compounds are accumulated by these organisms, although
apparently not at levels directly dangerous for human health (Karl et al. 2002, Knutzen et al.
2003). Data on DLC levels in sea stars are completely lacking. However, recent experimental
studies showed that dioxin-responsiveness of a CYP1A immunopositive protein (CYP1A
IPP) can be measured in sea stars using a specially developed ELISA (Danis et al. Chap.
III.4).
The goal of the present study is to compare the information provided by a chemical approach
(measuring DLCs in sediments and biota) and by a biomarker (the measurement of CYP1A
IPP induction in sea stars) for assessing the impact of DLCs in intertidal stations of the
southern North Sea and the Channel. The intertidal zone in this region is often structured by
mussels and sea stars, is touristically important, and amateur mussels gathering represents a
regular hobby.
MATERIALS AND METHODS
Field trips and sample preparation
All samples were collected between October and November 2001. Two replicates of the
upper 5 cm layer of the sediments were collected in 5 intertidal stations, along the French and
Belgian coasts (Table 39), and immediately frozen (-20°C) in 400 ml polypropylene
containers. Sea stars (A. rubens) and mussels (M. edulis) were collected by hand (seashore
fishing) in the same stations.
Table 39. Positions and characteristics of the sampling stations
Station code
Coordinates
(N)(E)
188
Salinity
(‰)
Ambleteuse 50°44.50 1°35.00 33.0
Nieuwpoort 51°08.80 2°42.80 32.5
Oostende 51°13.80 2°54.40 32.0
Wenduine
Knokke
51° 17.80
51°20.80
3°04.40
3°17.80
32.7
32.0

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
Only specimens belonging to the same size-class (5-7 cm from the tip of the arm to the centre
of the mouth for sea stars and 3-4 cm shell length for mussels) were considered. Sea stars
collected for contaminants analysis were dissected and pyloric caeca were pooled by 5 (3
pools per station) and frozen at -20°C until analysis. Pyloric caeca of 3 individuals per pool
were sub-sampled (approx. 1 ml) for CYP1A immunopositive protein (CYP1A IPP)
measurements. These sub-samples (n=9) were frozen in liquid nitrogen and transferred to a
deep freezer (-80°C). Collected mussels (3 pools per station; ca. 1kg fresh wt with shells per
pool) were washed with seawater, and frozen with the shell (-20°C) before analysis.
PCDD/Fs analysis
All analyses were carried out in the Laboratory of Mass Spectrometry, Centre for the Analysis
of Trace Residues (CART), University of Liège (ISO-17025 accreditation, certificat n°212-
T), using the method described by Focant et al. (2001).
Briefly, samples were extracted using an Accelerated Solvent Extractor (ASE 200, Dionex),
before gravimetricallly determining their lipid content. After dissolving the extracts in
hexane/dichloromethane (1:1, 50 ml) fat extracts were spiked with 10 µl of the internal
standard (17 13 C-labelled PCDD/F congeners, EDF-4144, LGC Promotion and 4 13 C-labelled
c-PCB congeners, Campro Scientific WP-LCS).
An HCDS column (disposable silica; 28 g acidic, 16 g basic, 6 g neutral) was added to the
classical Power-Prep set of columns (multi-layer silica columns (4 g acid, 2 g base and 1.5 g
neutral), basic alumina (8 g) and PX-21 (2 g) carbon column), all manufactured by Fluid
Management System (USA). Seventeen PCDD/Fs congeners, and 4 coplanar PCBs (IUPAC
#77, 81, 126, 169) were quantified in the samples using a gas chromatography column
coupled to a high resolution mass spectrometer (GC-HRMS, Autospec, Ultima). The mass
spectrometer was operated in the electron impact ionization mode using selected ion
monitoring (SIM).
Results were expressed either as pg g -1 of lipid weight or in terms of toxicity, using WHO fish
TEF (Van den Berg et al. 1998) as pg WHO TEQ g -1 of lipids weight. Quality controls (QCs)
were performed following the EN1948-3 norms, for both samples and blanks (Focant et al.
2001).
Adaptation of the procedure to analyze sediment samples
Sediment samples were analyzed as a whole without previous sieving. However, prior to
analysis, large shell fragments were removed. Samples (ca. 30g) were weighed and dried
overnight in an oven (75°C). Samples were extracted using the ASE system, using toluene
189

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
instead of hexane/dichloromethane as the extracting solvent. The following steps were
conducted as above.
Cytochrome P450 immunopositive protein (CYP1A IPP) induction
CYP1A IPP induction was quantified using competitive ELISA (Danis et al Chap. III.4). The
ELISA was carried out using competition for anti-CYP1A antibodies between the CYP1A
immunopositive protein from sea star samples and biotinylated ß-naphtoflavone (BNF)-
injected trout (Oncorhyncus mykiss) microsomes displaying high concentrations of CYP1A
(Biosense, Norway). Multiwell plates were coated with anti-CYP1A (rabbit anti-fish CYP1A
peptide, polyclonal antibody purchased from Biosense, Norway). Wells were washed with
phosphate-buffered saline (PBS, Sigma, Europe), and nonspecific binding sites were blocked
with PBS containing bovine serum albumin (BSA, Sigma, Europe). Wells were washed again,
and added with biotinylated trout microsomes (except for the blank wells). Samples or
standards (with adjusted protein concentration) were then added to the wells. Competition
was allowed to take place and after five washing steps, extravidin-HRP (Sigma, Europe) was
added to all the wells. The plate was then incubated and the wells were washed again using
PBS. Chromogen TMB (Pierce, United Kingdom) was added to all the wells and the plate
was incubated in obscurity. Sulfuric acid (Sigma, Europe) was added to the wells to stop the
reaction. Optical density was measured in the 96-well plates at 450 nm using a Packard
Spectracount microplate reader.
Data analyses
Differences between measured DLC concentrations or CYP1A IPP induction in the different
stations were tested using the non-parametric Kruskal-Wallis test followed by a multiple
comparison test of the means (Dunn test, Conover 1980). Regressions between the
concentrations measured in sea star pyloric caeca and mean induction of CYP1A IPP
measured in corresponding individuals were tested using simple linear regression procedures
(Zar 1996), applying a Bonferroni criterion to the significance level. The level of significance
for statistical analyses was always set at a = 0.05.
RESULTS
cPCB and PCDD/F concentrations in sediments
Contaminant concentrations were measured in the bulk sediment in five intertidal stations
(Table 40A and Fig. 48). No significant differences were found between DLC levels
determined in sediments from the different sampling stations (p Kruskal-Wallis>0.1), probably due
190

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
to the low number of replicates. However, some trends are apparent. Levels of ∑ 7PCDDs
were higher in sediments sampled in Ambleteuse and Nieuwpoort, intermediate in Knokke
and were below the quantification limit in Oostende and Wenduine. Sediments from
Nieuwpoort displayed relatively high levels of ∑ 10PCDFs, while the concentrations of these
compounds were intermediate in Oostende and lower in the remaining stations. ∑ 4c-PCB
were slightly higher in sediments from Nieuwpoort and were below the quantification limits
in the all the other stations.
Figure 48. Contaminant levels (mean pg g -1 dry wt.; n=2) measured in sediments from the different sampling
stations.
Table 40. Concentrations of DLCs measured in (A) sediments (Mean; n=2; pg g -1 dry wt.), (B) mussels (Mytilus
edulis; Mean±SD; pg g -1 lipids) or (C) sea stars (Asterias rubens; Mean±SD; pg g –1 lipids) from the southern
North Sea and the Channel. ND=not detected; values

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
OCDD ND 0.007 ND ND ND
∑ 7PCDDs 1.023 0.929 0.071 0.071 0.354
TCDF 0.008 0.541 0.009 0.008 0.008
PeCDF 1 0.001 0.023 0.171 0.009 0.011
PeCDF 2 0.014 1.074 0.014 0.014 0.014
HxCDF 1 0.005 0.173 0.005 0.005 0.005
HxCDF 2 0.003 0.180 0.005 0.006 0.003
HxCDF 3 0.003 0.150 0.003 0.003 0.003
HxCDF 4 0.003 0.003 0.003 0.003 0.003
HpCDF 1 0.005 0.077 0.005 0.005 0.005
HpCDF 2 ND 0.007 ND ND ND
OCDF ND ND ND ND ND
∑ 10PCDFs 0.042 2.227 0.215 0.053 0.051
∑ 17PCDD/Fs 1.065 3.156 0.286 0.124 0.405
PCB 77 0.001 0.001 0.001 0.001 0.001
PCB 81 ND ND ND ND ND
PCB 126 0.100 0.118 0.100 0.100 0.100
PCB 169 0.001 0.001 0.001 0.001 0.001
∑ 4cPCB 0.102 0.120 0.102 0.102 0.102
B. Mussels
Station
Ambleteuse Nieuwpoort Oostende Wenduine Knokke
(n=3) (n=3) (n=3) (n=1) (n=2)
TCDD ND 4.59 5.74 ± 0.38 7.10 13.1
PeCDD 3.25 ± 0.25 6.26 ± 0.96 5.79 ± 0.75 6.38 7.53
HxCDD 1 4.72 ± 1.64 8.98 ± 5.89 3.85 ± 0.13 4.49 5.16
HxCDD 2 7.41 ± 3.22 9.88 ± 4.65 10.1 ± 1.76 13.3 12.5
HxCDD 3 6.07 ± 1.68 6.46 ± 4.03 5.57 ± 1.20 5.96 6.22
HpCDD 57.3 ± 15.3 87.3 ± 68.8 75.1 ± 4.67 97.3 86.0
OCDD 165 ± 62.0 314 ± 294 293 ± 52.7 489 278
∑ 7PCDDs 240 ± 73.9 431 ± 370 399 ± 60.0 624 408
TCDF 162 ± 43.9 176 ± 25.8 168 ± 30.9 130 178
PeCDF 1 13.3 12.2 13.6 ± 1.18 15.3 21.0
PeCDF 2 36.9 ± 8.07 46.7 ± 14.0 42.9 ± 3.07 39.5 60.8
HxCDF 1 6.27 ± 3.32 19.1 ± 15.6 10.1 ± 3.85 11.7 19.1
HxCDF 2 4.92 ± 2.09 7.12 ± 4.55 4.51 ± 0.89 4.80 8.22
HxCDF 3 0.34 ND 0.24 ND ND
HxCDF 4 6.89 ± 4.42 11.9 ± 8.86 7.70 ± 0.09 8.02 12.5
HpCDF 1 13.8 ± 4.40 32.6 ± 25.9 29.9 ± 9.69 34.4 46.6
HpCDF 2 0.77 ± 0.49 4.77 ± 5.70 1.87 ± 0.43 1.84 3.16
OCDF 12.5 ± 4.34 33.4 ± 22.7 42.3 ± 22.1 62.6 65.6
∑ 10PCDFs 248.4 ± 63.0 332 ± 115 321 ± 62.1 309 415
∑ 17PCDD/Fs 488 ± 123 763 ± 484 721 ± 120 932 823
PCB 77 4120 ± 611 3760 ± 1810 3290 ± 2530 3150 3270
PCB 81 534 ± 460 902 ± 719 195 ± 16.6 179 121
192

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
PCB 126 833 ± 77.0 705 ± 63.8 861 ± 113 843 927
PCB 169 84.9 ± 8.38 82.6 ± 5.91 86.3 ± 13.4 97.0 117
∑ 4cPCB 5570 ± 970 5450 ± 2428 4430 ± 2670 4270 4430
C. Sea stars
Station
Ambleteuse Oostende Wenduine Knokke
(n=3) (n=3) (n=2) (n=2)
TCDD 0.82 ± 0.05 7.73 ± 0.47 7.87 11.8
PeCDD 2.06 7.79 ± 0.23 7.96 7.66
HxCDD 1 1.03 2.72 ± 0.58 3.41 2.97
HxCDD 2 3.21 11.7 ± 0.84 12.3 10.9
HxCDD 3 1.51 ± 0.02 4.62 ± 0.48 5.16 4.32
HpCDD 10.8 ± 0.25 32.9 ± 2.55 31.9 30.6
OCDD 30.3 ± 0.57 78.2 ± 4.95 75.1 69.9
∑ 7PCDDs 47.3 ± 5.68 146 ± 5.28 143.7 138
TCDF 70.3 ± 6.37 124 ± 14.9 134.9 156
PeCDF 1 ND 0.93 ± 0.30 ND 1.50
PeCDF 2 6.75 22.0 ± 3.83 21.4 44.4
HxCDF 1 0.58 0.43 ± 0.19 0.74 0.66
HxCDF 2 ND ND ND 0.47
HxCDF 3 ND ND ND ND
HxCDF 4 0.57 3.50 ± 0.31 2.83 4.27
HpCDF 1 ND 3.61 ND ND
HpCDF 2 ND ND ND ND
OCDF ND ND ND ND
∑ 10PCDFs 75.5 ± 0.84 152 ± 17.6 159 207
∑ 17PCDD/Fs 123 ± 0.27 298 ± 16.7 303 345
PCB 77 7280 ± 1590 7290 ± 1460 6360 2430
PCB 81 355 ± 20.9 96.1 ± 20.4 229 190
PCB 126 552 ± 59.5 1030 ± 141 988 1360
PCB 169 ND 92.7 ± 2.40 95.4 131
∑ 4cPCB 8070 ± 1673 8440 ± 1465 7670 4110
The ratios between the most concentrated congener and the sum of all congeners of a given
class of contaminants were calculated (Fig. 49). In sediments, PeCDD was the most abundant
among PCDDs (except in Knokke, where TCDD was the most abundant), PeCDF2 was the
most abundant of PCDFs (except in Oostende, where PeCDF1 was the most abundant) and
PCB 126 displayed the highest concentrations among cPCBs. When considering all the
contaminants, the predominant congener varies from one station to another: Ambleteuse
contaminant pattern was dominated by PeCDD (79% of ∑ 21DLCs), Nieuwpoort by PeCDF2
(33% of ∑ 21DLCs), Oostende by PeCDF1 (44% of ∑ 21DLCs), Knokke by TCDD (61% of
∑ 21DLCs), and Wenduine by PCB 126 (44% of ∑ 21DLCs).
193

% major congener/sum
100
90
80
70
60
50
40
30
20
10
0
PeCDD
Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
PeCDF2
PCB126
OCDD
Figure 49. Calculated ratios between the major congener of a given class
of contaminant and the sum of all the congeners of that given class, in
sediments (white), mussel (light grey) and sea stars (dark grey).
The contribution of the different contaminants to the total toxicity is presented in Fig. 50A.
Depending on the considered station, the different contaminant families contribute very
differently to the total toxicity: in Ambleteuse and Knokke, the major part of the total TEQ is
attributable to PCDDs, while PCDFs contribute the most to toxicity in Nieuwpoort and
Oostende.
A.
B.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
TCDF
194
PCB77
∑PCDDs ∑PCDFs ∑c-PCBs
OCDD
TCDF
PCB77
Ambleteuse Nieuwpoort Oostende Wenduine Knokke
∑PCDDs ∑PCDFs ∑cPCBs
Ambleteuse Nieuwpoort Oostende Wenduine Knokke

C.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
∑PCDDs ∑PCDFs ∑cPCBs
Ambleteuse Oostende Wenduine Knokke
Figure 50. Contribution to toxicity (%TEQ) of the different classes of
contaminants (PCDDs, PCDFs, cPCBs) in sediments (A), mussels (B) and
sea stars (C) from the different sampling stations.
PCDD/F and c-PCB concentrations in mussels
Mussels (Mytilus edulis) were analyzed for their content in PCDDs, PCDFs and c-PCBs
(Table 40B, Fig. 51). No significant differences were found between DLC levels measured in
mussels from the different sampling stations (p Kruskal-Wallis>0.1), but some trends could be
highlighted.
Figure 51. Contaminant levels (mean±SD, pg g -1 lipids) measured in mussels (Mytilusedulis) from the different
sampling stations.
195

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
Table 41 presents the concentrations (pg TEQ g -1 lipids) measured in mussels. Total TEQ
were remarkably constant, with values ranging from 124 pg TEQ g -1 lipids in Nieuwpoort to
159 pg TEQ g -1 lipids in Knokke. The situation was very different from that found in
sediments analyses: regardless of the considered station, most of the TEQ was attributable to
c-PCBs (Fig. 50B).
Table 41. Contaminant TEQs (ng WHO TEQ g -1 lipids, mean±SD, n = 3, except in Knokke (n=2) and Wenduine
(n=1)) measured in mussels from the different sampling stations. Values in bold represent the major congener,
for each sampling station, and for each contaminant group.
Station
Contaminant Ambleteuse Nieuwpoort Oostende Wenduine Knokke
TCDD 0.18 1.65 ± 2.54 5.74 ± 0.38 7.10 13.1
PeCDD 2.27 ± 1.71 4.28 ± 3.51 5.79 ± 0.75 6.38 7.53
HxCDD 1 0.47 ± 0.16 0.90 ± 0.59 0.39 ± 0.01 0.45 0.52
HxCDD 2 0.50 ± 0.47 0.99 ± 0.46 1.01 ± 0.18 1.33 1.25
HxCDD 3 0.61 ± 0.17 0.44 ± 0.46 0.56 ± 0.12 0.60 0.62
HpCDD 0.57 ± 0.15 0.87 ± 0.69 0.75 ± 0.05 0.97 0.86
OCDD 0.02 ± 0.01 0.03 ± 0.03 0.03 ± 0.01 0.05 0.03
∑ 7PCDDs 4.44 ± 1.68 7.50 ± 4.17 8.52 ± 1.07 9.78 10.8
TCDF 16.2 ± 4.39 17.7 ± 2.58 16.9 ± 3.09 13.1 17.8
PeCDF 1 0.23 ± 0.38 0.21 ± 0.35 0.68 ± 0.06 0.77 1.05
PeCDF 2 18.4 ± 4.03 23.3 ± 7.00 21.4 ± 1.54 19.8 30.4
HxCDF 1 0.63 ± 0.33 1.91 ± 1.56 1.01 ± 0.39 1.17 1.91
HxCDF 2 0.49 ± 0.21 0.48 ± 0.51 0.45 ± 0.09 0.48 0.82
HxCDF 3 0.02 ± 0.01 0.02 0.02 0.02 0.02
HxCDF 4 0.69 ± 0.44 1.19 ± 0.89 0.77 ± 0.01 0.80 1.25
HpCDF 1 0.14 ± 0.04 0.33 ± 0.26 0.30 ± 0.10 0.34 0.47
HpCDF 2 0.01 0.03 ± 0.05 0.02 ± 0.00 0.02 0.03
OCDF 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 0.01 0.01
∑ 10PCDFs 36.8 ± 8.75 45.1 ± 12.3 41.5 ± 3.59 36.4 53.8
∑ 17PCDD/Fs 41.3 ± 8.13 52.6 ± 15.5 50.1 ± 4.33 46.2 64.6
PCB 77 0.41 ± 0.06 0.38 ± 0.18 0.33 ± 0.25 0.31 0.33
PCB 81 0.05 ± 0.05 0.09 ± 0.07 0.02 0.02 0.01
PCB 126 83.3 ± 7.70 70.5 ± 6.38 86.1 ± 11.3 84.3 92.7
PCB 169 0.85 ± 0.08 0.83 ± 0.06 0.86 ± 0.13 0.97 1.17
∑ 4cPCB 84.7 ± 7.79 71.8 ± 6.12 87.3 ± 11.7 85.6 94.2
Total TEQ 126 ± 15.9 124 ± 21.6 137 ± 16.00 132 159
The ratios between the most concentrated congener and the sum of all congeners (for each
class of contaminants) measured in mussels is presented in Fig. 49. OCDD was the most
abundant among dioxins, TCDF was the most abundant of furans and PCB 77 displayed the
196

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
highest concentration among c-PCBs. When considering all contaminants together, PCB 77
was always the most concentrated congener, whichever station was considered (proportion
ranged from 60 to 67%, with a low variability) (Table 40B).
PCDD/F and c-PCB concentrations in sea stars
Results of the analyses are shown in Table 40C and Fig. 52 (sea star samples from
Nieuwpoort were lost during DLC analysis). The situation was more contrasted than in the
case of sediments and mussels analyses, as significant differences were found between
∑ 10PCDFs levels measured in sea stars from the different stations, (0.01

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
among dioxins, TCDF was the most concentrated among furans and PCB 77 was the major
congener among c-PCBs. When considering all contaminants together, PCB 77 was always
predominant, but its degree of predominance was lower in Knokke (55%) than in the other
sampling stations (values ranging from 80 to 88%) (Table 40C).
The contribution of the different classes of contaminants to the total toxicity in sea stars
displayed the same scheme as in mussels (Fig. 50C). The major part of toxicity was due to c-
PCBs, this trend being even more manifest than in the case of mussels.
In order to assess the possible biomagnification of the different congeners, bioconcentration
factors (BCF; ratio between mean concentration in sea stars and mean concentration in
mussels) were calculated, for each sampling station (Fig. 53). Coplanar PCBs were found to
be biomagnified in almost all cases (BCFs ranging from 0.67 to 3.01), whereas the other
contaminant classes usually displayed BCF values ≤1.
Bioconcentration Factor
2,5
2
1,5
1
0,5
0
TCDD
PeCDD
HxCDD 1
HxCDD 2
HxCDD 3
HpCDD
OCDD
Ambleteuse Oostende Wenduine Knokke
∑7PCDDs
TCDF
198
PeCDF 2
HxCDF 1
DLC congener
HxCDF 4
∑10PCDFs
∑17PCDD/Fs
Figure 53. Bioconcentration factors (ratio between the mean concentration measured in sea stars and the mean
concentration measured in mussels) calculated for the different congeners, in each sampling station.
CYP1A IPP induction in sea stars
The induction of a CYP1A IPP was measured using competitive ELISA in sea stars collected
in the different stations (Fig. 52). Significant differences (p Kruskal-Wallis≤0.0001) were found
between the different stations: sea stars from Knokke displayed the highest CYP1A IPP
PCB 77
PCB 81
PCB 126
PCB 169
∑4cPCBs

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
levels. The lowest induction values were recorded in sea stars from Ambleteuse. Values
measured in sea stars from Oostende, Nieuwpoort and Wenduine were intermediate.
Regressions were calculated between mean CYP1A IPP induction and DLC concentrations or
TEQ in corresponding sea star pools. When considering concentrations, significant
regressions were found between CYP1A IPP induction and ∑ 10PCDFs or ∑ 17PCDD/Fs
concentrations measured in sea stars. Determination coefficients respectively reached 0.68
(p=0.002) and 0.63 (p=0.004). Highly significant regressions were also found between
CYP1A IPP induction and TEQs for all contaminant classes (Fig. 54). Determination
coefficients ranged between 0.56 (CYP1A IPP induction as a function of ∑ 7PCDDs TEQ in
sea stars) and 0.79 (CYP1A IPP induction as a function of ∑ 10PCDFs TEQ in sea stars).
CYP1A IPP induction (Time fold)
5
4
3
2
1
y = 0,11x + 1,12
R 2 = 0.56
p=0.008
y = 0,10x + 0,52
R 2 = 0.79
p≤0.0001
∑7PCDD ∑10PCDF ∑17PCDD/Fs ∑4cPCB
y = 0,06x + 0,69
R 2 = 0.73
p=0.001
199
y = 0,03x + 0,12
R 2 = 0.73
p=0.001
0
0 20 40 60 80 100 120 140 160 180 200
DLC concentration (pg TEQ g -1 lipids)
Figure 54. Regressions between mean CYP1A IPP induction (Time fold) and TEQs values of the different DLC
classes (pg TEQ g -1 lipids) determined in sea star pyloric caeca. R 2 : corrected determination coefficient.
DISCUSSION
No significant differences in DLC levels measured in sediments, mussels or sea stars were
found between the different sampling stations, except for ∑ 10PCDFs in sea stars. This lack of
significant differences is partly due to the relatively low number of replicates. Most DLC
concentrations in sediments were below the limits of quantification. Sea stars were found to

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
be the most discriminant compartment, displaying significant differences for ∑ 10PCDF levels
between the stations. Differences in the discriminating power of sediments, mussels or
asteroids were probably attributable to a generally low-contrasted contamination situation,
and to the importance of bioavailability: patterns and concentrations of hydrophobic
contaminants in aquatic organisms are generally considered to be determined not only by
concentrations found in their surrounding environment, but also by physiological processes
(Barron 1990). In this context, it is noteworthy that the major congeners in the three
considered DLC classes are the same in mussels and sea stars, and that they strongly differ
from those found in sediments.
In sea stars, the relative contribution of c-PCBs to the total DLC burden tended to be different
in Knokke than in the other stations, while in mussels, there were no differences among the
stations. This observation highlights bioindicating differences between the two organisms,
probably due to their contrasted biology (food source, metabolism rates,…), resulting in
bioaccumulative differences (Pruell et al. 2000). Another striking difference between the
relative patterns of congeners is found among cPCBs: PCB 77 displays very low levels in
sediments, whereas in mussels and sea stars, it is the most abundant PCB congener (1 to 2
orders of magnitude more concentrated than the other PCB congeners). This can be due to
bioaccumulative capacity by organisms, and/or contamination occurring mainly via seawater
(Danis et al. Chap. III.2, III.3). DLC concentrations measured in sea star pyloric caeca
displayed lower variability (measured as standard deviation) than those determined in
mussels. PCDD/F levels in sea stars were lower than those measured in mussels, while PCB
77 was found to be more concentrated in asteroids. This probably results from differences in
uptake efficiency and metabolization processes (Walker & Peterson, 1994), which seem to be
more efficient in sea stars than in mussels.
Biomagnification was found to occur along the mussels=>sea stars food chain almost
exclusively for c-PCBs. The other contaminant groups (highly chlorinated PCDDs and
PCDFs) almost always displayed higher concentrations in mussels than in sea stars. It is
noteworthy that experimental contamination of sea stars by PCB 77 via food resulted in very
similar BCF values (1.84±0.43 in experimental conditions compared to 1.68±0.65 in the
present study) (Danis et al. Chap. III.3).
One of the main sources of human exposure to chemicals such as PCBs (with the exception of
accidental and occupational exposure) is intake via seafood (Friberg 1988, Svensson et al.
1991). Direct evidence of health hazard attributable to contaminated seafood is poorly
documented. Existing studies are often subject to considerable debate (Knutzen et al. 2003)
200

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
and causal relationships were rarely established. General proposals that PCBs taken up from
diet are associated with breast cancer are controversial (Key & Reeves, 1994; Smith, 2000).
In various studies carried out in the Netherlands, environmental exposure to dioxins and
PCBs has been associated with impaired cognitive development and greater susceptibility to
infections and allergy (Patandin et al. 1999, Weisglas-Kuperus et al. 2000). TEQs were
calculated for mussels sampled in this study, in order to compare them to existing regulations
and to results from other studies. Most countries use the Tolerable Daily Intake (TDI) for
assessing the health risks linked to exposure to dioxin-like compounds. The TDI is derived
from the assumption that there is a threshold dose level below which no toxic effect is
produced by the compound in an animal model. The World Health Organization (WHO)
Expert Committee recommends a TDI for TCDD at 10 pg kg -1 body weight day -1
(WHO/EURO 1991). The WHO European Centre for Environment and Health (ECEH) and
the International Programme on Chemical Safety (IPCS) readjusted the threshold level,
considering that occasional short-term exposition beyond the TDI probably does not represent
a serious health threat, as far as the average intake was not exceeded. ECEH and IPCS
recommended a TDI of 4 pg WHO TEQ kg -1 body weight day -1 (CoT 2001). The levels
measured in mussels in this study ranged between 124-159 pg TEQ g -1 lipids, which roughly
represents values of 1 pg TEQ g -1 fresh wt. A normal adult, weighing around 70 kg, should
not exceed 280 pg TEQ day -1 , that is 280 g fresh wt per day (soft tissues only) of mussels
from the area considered in the present study. A meal of mussels usually ranges between
500g and 1 kg of mussels (with shells), which corresponds to ca. 125-250 g fresh wt.
The levels measured in mussels in the present study were one order of magnitude higher than
those found by Karl et al. (2002) in blue mussels sampled in German supermarkets, Miyata et
al. (1987, 1994) in mussels from the Osaka bay and in coastal areas from Japan, Broman et al.
(1992) in the northern Baltic sea, Haynes and Toohey (1995) in Port Philip Bay (Australia)
and Moon et al. (2000) along the Korean coast. DLC levels and congener patterns in mussels
determined in the present study were very similar to those found by Abad et al. (2002) in
mussels from Catalonia.
Measures of CYP1A IPP induction in sea star pyloric caeca displayed highly significant
differences between the sampling stations. Sea stars collected in Knokke displayed the highest
induction factor, while lower induction factors were measured in Ambleteuse. The situation
was much more contrasted than in the case of DLC analyses in the different compartments.
On the other hand, strong and highly significant regressions were found between DLC levels
(determined as concentrations or TEQs) in sea stars, and CYP1A IPP induction, supporting a
201

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
good responsiveness of sea stars to DLCs. The unexplained variations of measured CYP1A
IPP induction could be due either to the presence of other CYP1A-inducing contaminants in
the environment (such as PAHs), or to endogenous factors, such as endocrine regulation. If
DLC metabolization occurs in sea stars as in vertebrates, an assumption supported by
observations reported elsewhere (Danis et al. Chap. III.1, III.4), high levels of CYP1A IPP in
sea stars can lead to adverse effects, as this enzyme system has an important role in
endogenous metabolization of steroids and fatty acids (Bucheli & Fent 1995). Disruption of
the endocrine system of sea stars can in turn affect the reproduction and development of these
ecosystem-structuring species.
Mean levels of induction measured in the present study ranged from 1.51 to 4.11 time folds,
which contrasts with higher values reported for the same area by Danis et al. (Chap. IV.2),
which reached 43 time folds. In the present study, the sampling of sea stars was carried out
during the fall, which corresponds to the beginning of the gametogenesis period, while, in the
former study, organisms were sampled in February, which corresponds to the prespawning
period in A. rubens. During both periods, sea stars undergo important changes in their steroid
metabolism (Voogt et al. 1991). The important differences in CYP1A IPP induction found
between the sampling periods underlines the importance of the choice of the sampling period
for CYP1A IPP measurements in sea stars.
CONCLUSIONS
Measuring DLC concentrations in sediments and biota (mussels and sea stars) from the
Belgian coast and the Channel allowed us to integrate the differential bioaccumulation
efficiencies of these two bioindicator species. Although the levels found in sediments were
often below quantification limits, measured concentrations were found to be relatively high in
biota, given the toxicity of these compounds. Special concern arises from TEQ levels
determined in mussels. The induction of CYP1A IPP was found to be related to DLCs levels
measured in sea stars. Measurement of this biomarker allowed to highlight significant inter-
station differences, which only appeared as trends when considering DLC analyses alone.
ACKNOWKLEDGEMENTS
Grateful thanks are due to Ir Gauthier Eppe for technical advice on PCDD analysis. Research
supported by a Belgian Federal Research Programmes (SSTC, Contract MN/12/95 and
EV/ENZ13). BD was holder of a FRIA doctoral grant. PhD is Research Associate of the
202

Levels and effects of PCDD/Fs and c-PCBs in sediments, mussels and sea stars
National Fund for Scientific Research (NFSR, Belgium). Contribution of the "Centre
Interuniversitaire de Biologie Marine" (CIBIM).
203

204

V. GENERAL DISCUSSION
205
GENERAL DISCUSSION
The present study focused on the bioaccumulation and effects of PCBs in the common NE
Atlantic sea star Asterias rubens. The first part of the study was achieved in laboratory
conditions in order to examine the biokinetics of these compounds and their effects at
immune and subcellular levels in sea stars. The second part focused on the southern North
Sea contamination status, using the sea star as an indicator of contamination, and examining
the effects of PCBs on various aspects of this organism biology in field situations. As each
chapter has already been discussed individually, the present chapter will address some more
integrated aspects, including the bioaccumulation behaviour of PCBs in A. rubens in field and
experimental conditions, and on PCB effects on immune and subcellular responses in the sea
star.
Finally, overall recommendations for the design of future monitoring programmes will be
addressed.
The bioaccumulation of PCBs in sea stars
The experimental exposures carried out in this study have shown that A. rubens efficiently
accumulates PCBs. Uptake kinetics displayed a strong congener specificity, and were more
specifically related to the planar conformation of the various congeners. In most body
compartments, non-coplanar PCBs accumulation was described using saturation models.
Saturation concentrations were generally found (1) to be reached quite fast in target sites
(whithin a few days to a few weeks), whichever contamination source was considered, and (2)
to remain constant until the end of the exposure period, showing that sea stars can be used to
pinpoint a PCB contamination quite rapidly after occurrence. Previous studies on PCB
accumulation in sediments-exposed marine invertebrates have shown that PCBs (considered
as Aroclor mixtures) were also taken up by these organisms following saturation kinetics
(Courtney & Langston 1978, McLeese et al. 1980).
In contrast, coplanar PCBs were not accumulated following saturation kinetics, but rather
appeared to reached a concentration peak, followed by a sudden drop after approximately 9
days of exposure. The fact that this drop only occurred for c-PCBs is a first indication of the
occurrence of a targeted metabolization of these congeners in sea stars. However, this
behaviour was not observed in our study on 14 C-PCB 77 bioaccumulation in A.rubens. This
observation is most probably due to the fact that the latter experiment was carried out using a

206
GENERAL DISCUSSION
radiolabelled congener, which implies that if a c-PCB-targeted metabolization takes place in
sea stars, the method does not allow making a difference between the congener itself and its
degradation products if the latter one bears the 14 C-labelling. Among coplanar PCBs,
congener 126 is known to be metabolized in a range of organisms by cytochrome P450
isozymes (Hong et al. 1998); existing evaluations of PCB 77 toxicity and metabolization are
blurred by important variations in effects and between species (Safe 1994).
The efficiency and rate at which PCBs are accumulated in sea stars varied among the
considered body compartments and exposure routes. Among the considered body
compartments, body wall often displayed the highest bioaccumulative potency, reaching high
PCB levels whichever exposure route was considered. This feature, along with the fact that
body wall is easily dissected and important in terms of weight, highlights the particular
interest for its use in field studies. It would interstingly complement information gathered via
analysis of the pyloric caeca, which are generally the only organs considered separately in
monitoring studies (Everaarts et al. 1998, den Besten et al. 2001, Stronkhorst et al. 2003). No
differences were found in field or experimental conditions between PCB levels in aboral or
oral body wall although the latter is in direct contact with sediments. Since only oral body
wall is in direct contact with sediments, this finding indicates that PCB translocation
processes readily occur in the body wall of the sea star. Among the body compartments
considered in the present study, rectal caeca have rised many interesting questions regarding
their possible involvment in regulating PCB levels in sea stars. In this perspective, the use of
radiotracer techniques has brought very valuable information. 14 C-PCB 153 activities reached
in these organs were between 1 and 2 orders of magnitude lower than in the other
compartments, while activities measured during the 14 C-PCB 77 experimental exposure
reached similar values in rectal caeca than in the other compartments. This discrepancy could
be the result of the presence of 14 C-labelled PCB 77 degradation products, more soluble than
the parent compound, and which would be eliminated via this route (the lumen of rectal caeca
are in relation with the outside environment via the very short rectum). Rectal caeca are
known to play a pivotal role in sea star digestion and excretion processes (Jangoux 1982,
Warnau & Jangoux 1999). As no activity increase was recorded in seawater at the end of the
exposure period, potential metabolites were not yet eliminated although the coplanar PCB
drop was observed around the first 9 days of exposure. PCB metabolites are often thought to
be more toxic than the parent compounds (Walker & Peterson 1994). Therefore the possibly
long residence time in rectal caeca could induce secondary toxicity on the organism, leading
to adverse effects on various aspects of the sea star biology. In the framework of the present

207
GENERAL DISCUSSION
study, the occurrence of PCB metabolites could not be verified using classical detection
methods, due to the extreme minuteness of this organ (15-50 mg wet wt).
Among non-coplanar PCBs, the rate at which accumulation occurred in body wall of
sediment-exposed sea stars depended on the chlorination degree of the considered congener:
the ‘lighter’ congeners were accumulated faster than the other congeners. In pyloric caeca,
however, the accumulation rate displayed a different pattern among the congeners, probably
due to variable inter-compartment fluxes.
In both experimental and field conditions, all considered PCB congeners were
bioconcentrated by A. rubens.
All the experimental studies demonstrated that seawater was the most efficient route for PCB
uptake in sea stars, reaching relative bioaccumulation efficiency up to 3 orders of magnitude
higher than those measured in sediment- or food-exposed organisms. The same observation
was made for sea urchins and cuttlefish (see Danis et al. Chap VII.1, VII.2), and in previous
studies on marine polychaetes (Courtney & Langston 1978). Therefore, although PCB levels
in seawater generally reported in field studies are extremely low compared to those measured
in sediments (Schulz-Bull et al. 1995), seawater constitutes a non-negligible route for PCB
contamination in marine invertebrates. This observation can explain certain discrepancies
found between PCB levels and patterns determined in sediments and biota in field conditions,
which are probably the result of the seawater contribution to sea star PCB contamination.
In the experimental work carried out in the present study, the exposure concentrations were
chosen to match those reported in the field for the corresponding compartment (seawater,
sediments or food). The final concentrations reached in exposed sea stars (background +
incorporated) generally matched those measured in the field as well as those determined in
previous studies (Everaarts et al. 1998, den Besten 2001, Stronkhorst et al. 2003). This
observation showed that our experimental protocols correctly simulated actual field situations,
and also that working with experimentally-designed PCB mixtures did not interfere with
uptake efficiencies measured in sea stars. The parity beween experimental observations and
levels determined in field situations also shows that the interference with other environmental
contaminants (i.e. heavy metals, PAHs, pesticides) only alters PCB accumulation to a limited
extent.
The use of an organism as a bioindicator implies that contaminant levels measured in its
various body compartments reflect those measured in the surrounding environment. Hence, an
organism that is able to regulate its internal PCB levels cannot be used as a bioindicator. The
sea star A. rubens differentially accumulates PCBs, depending on the congeners planarity (viz.

208
GENERAL DISCUSSION
dioxin-like conformation). Therefore, the sea star ability to accurately reflect environmental
levels appears congener-dependent. It can be considered as a suitable indicator species for
medium-chlorinated PCB congeners: for certain non-coplanar congeners (e.g. 118 and 138),
strong relationships were found in the field between concentrations measured in sediments
and those determined in the sea star body wall. On the other hand, the sea star appears to be
able to regulate to a certain extent its content in coplanar PCBs. The latter observation has
two main implications: (i) A. rubens cannot be strictly considered as an indicator organism
for c-PCBs and (ii) being accumulated by sea stars, c-PCBs probably affect essential aspects
of sea star biology in field conditions, potentially leading to deleterious effects.
The effects of PCBs in sea stars
The present study addressed effects of PCB exposure on A. rubens biology in experimental as
well as in field conditions. Two main biological aspects were considered: ROS production by
sea stars amoebocytes and the induction of a CYP1A immunopositive protein (CYP1A IPP).
In experimental conditions, PCBs were found to significantly alter ROS production by sea
stars amoebocytes in a congener-dependent manner. The immune system of sea stars injected
with structurally-contrasting PCBs (PCB 153 and PCB 77) was found to respond only to the
coplanar congener. This response occurred at all considered doses, except the highest, at
which ROS production dropped to control levels. ROS production in sea stars experimentally
exposed to PCB 77 via seawater, sediments or food was also found to drop after a varying
lapse of time, ranging from 4 to 10 days. In the multiple exposure experiment, a strict
relationship was found between ROS production and coplanar PCB concentrations measured
in sea star tissues. In the latter experiment, ROS production was also found to drop after a few
days of exposure. All these observations showed the congener-specificity of ROS production
in sea stars: among the 10 different congeners tested, only the coplanar PCBs were found to
significantly affect, and probably impair sea star immune system. The immunomodulation by
coplanar PCB has already been described in sea urchins (Coteur et al. 2001), amphipods
(Borgmann et al. 1990), fish (Duffy et al. 2001), marine mammals (Ross et al. 1996) and man
(Vos & Van Loveren 1998).
In our field studies, ROS production by sea stars amoebocytes was found to be affected and
sometimes strongly impaired by contaminant levels. No direct relationships were found
between immunomodulation and non-coplanar PCB levels measured in sea stars, which is in
agreement with the c-PCB specific effect on ROS observed in our experimental approach.
Therefore, field observed immunomodulation was most probably due to the presence of other

209
GENERAL DISCUSSION
immunotoxic contaminants (viz. c-PCBs, heavy metals). Unfortunately the c-PCB
contribution to immunotoxicity could not be determined because none of our studies
considered ROS production along with coplanar PCB concentrations measurement.
ROS production by amoebocytes constitutes the main defence line against infectious agents.
Therefore levels of ROS production by sea stars amoebocytes measured in the field and the
levels reached after experimental exposure of sea stars to coplanar PCBs can lead to adverse
effects in sea stars. Impairment of ROS production can thus induce a higher susceptibility of
sea stars populations to disease (Winston & di Giulio 1991).
ROS measurements carried out in field studies displayed quite high variability. Part of this
variability can be explained by exogenous factors other than xenobiotics occurring in the
environment (e.g. temperature, microorganism abundance). As is the case for most
physiological processes, temperature for instance has been shown to influence ROS
production in sea stars (Coteur et al. 2004). In the latter study, the coldest months of the year
were characterized by an increase in ROS production, acting in a treshold-dependent way. It
was shown that ROS production measured in January displayed higher variability than during
the other months of the year; authors concluded that this period should be avoided for
sampling in the framework of monitoring studies.
During the present study a novel ELISA was setup and applied to measure the induction of a
cytochrome P450 immunopositive protein (CYP1A IPP) in experimental and field conditions.
Experimental work during which sea stars were exposed to a mixture of 10 PCB congeners
via sediements has shown that the induction of this protein was related to PCB exposure, in a
congener-specific fashion: coplanar PCBs alone were found to strongly induce the
production of CYP1A IPP according to a dose-dependent relationship. When sea stars were
exposed to PCB 77 via three different exposure pathways (seawater, food and sediments),
CYP1A IPP was also found to be induced, whichever uptake route was considered. In the
field, CYP1A IPP induction was shown to be a valuable biomarker: it was found to be
strongly related to PCB levels measured in sea stars. CYP1A is one of the most studied
biomarker in marine ecotoxicology (e.g. Bucheli & Fent 1995, Stegeman 1995, Hahn 2002a).
Its occurrence has been shown in marine molluscs (Stegeman 1985, Michel et al. 1993,
Livingstone & Goldfarb 1998, Peters et al. 1998), crustaceans (James & Little 1980, Singer et
al. 1980) and annelids (Lee 1981, Lee et al. 1981). The existence of CYP1A has also been
demonstrated in sea stars (den Besten et al. 1990b, 1993) and, more recently, the first
echinoderm CYP genes were identified by in digestive tissues of a sea urchin (Snyder 1998).

210
GENERAL DISCUSSION
However, the mechanisms underlying CYP1A induction in invertebrates are still much
debated (see e.g. Hahn 2002a,b).
Kinetic and biological effect studies carried out in the present work have highlighted many
similarities between the dioxin responsiveness of CYP1A IPP induction in sea stars and that
described in vertebrates. This strongly suggests the existence of similarities in the mechanism
of toxicity of these compounds. In vertebrates, the toxicity of dioxin-like compounds implies
a cytosolic receptor, the Aryl hydrocarbon Receptor (AhR), which occurrence has been
shown in certain invertebrate animals (Hahn 2002b). Invertebrates exhibit substantial
differences in the structural and regulatory characteristics of their biotransformation systems
and dioxin-susceptibility as compared to vertebrates (Hahn 1998). In vertebrates, CYP1A
expression is regulated by the AhR, and is known to play a pivotal role in metabolizing
endogenous substances such as steroid hormones (Bucheli & Fent 1995). If similarities exist
between CYP1A IPP induction in sea stars and in vertebrates, several factors are susceptible
to influence the CYP1A enzyme system, having potential consequences on its use as a
biomarker. These factors can have an exogenous or endogenous origin. The dominant abiotic
factor which influences CYP1A induction in vertebrates is temperature, which can affect
measurements by changing the availability of contaminants in the organism’s surrounding
environment, or by affecting the induction and activity of the enzymatic system (Stegeman &
Hahn 1994). CYP1A IPP induction in sea stars may also be affected by factors such as
nutrition, although in fish CYP1A is still inducible in moderately starved specimens (Jimenez
& Stegeman 1990, Vigano et al. 1993). Finally, as it is the case in fish (Elskus et al. 1989,
Lange et al. 1993, Stegeman & Hahn 1994), CYP1A IPP induction can be affected by sex or
by the stage in reproductive cycle. However, a previous study found no significant differences
between total cytochrome P450 contents in male and female sea stars (den Besten et al.
1990b). Our field studies showed significant differences in CYP1A IPP induction ranges
according to the period of the year, with lower induction factors measured in sea stars
collected in automn than in those sampled during the spring. This difference could be
explained by either or both factors mentionned before (temperature and reproductive status).
Although further research is needed to better characterize the influence of surrounding
seawater temperature and sea star reproductive status on CYP1A IPP induction, this
biomarker was found to be significantly related to contaminant levels measured in the field. In
the considered area (viz. the southern North Bay), PCB contamination may thus represent a
significant threat for sea star populations. Although being a prerequisite for detoxification,
CYP1A can also be involved in several toxic mechanisms. This enzyme system can convert

211
GENERAL DISCUSSION
protoxicants to more reactive products, a mechanism which is well described for a range of
PAH compounds (Stegeman & Lech 1991, Stegeman et al. 1992, Stegeman & Hahn 1994).
An alteration in enzyme activity by xenobiotics may also have adverse effects on endogenous
substrate metabolism and normal physiological processes, and more specifically on the
steroid metabolism (Elskus et al. 1989, 1992). In female fish, the alteration of sexual steroids
by xenobiotic-induced CYP1A has been suggested to be a cause for reduced reproductive
success of fish populations living in contaminated areas (Spies & Rice 1988). Disturbance of
the catalytic cycle of CYP can also lead to the production of ROS which are potentially toxic
and mutagenic (Stegeman et al. 1992, Stegeman & Hahn 1994). As it has been reported in
fish, experimental exposure of sea stars to PCBs have shown that ROS production can be
partly due to CYP1A IPP activity, as ROS production was found to be directly proportional to
CYP1A IPP induction. More research is needed to characterize the consequences of induced
CYP1A IPP levels in sea stars, but this does not question the clear ecotoxicological relevance
of this biomarker.
In sea stars, CYP1A IPP content was found to increase with increasing exposure
concentrations of coplanar PCBs. Ethoxyresorufin-o-deethylase (EROD) is one of the
enzymatic activities of the CYP system; it is thought to be the most sensitive catalytic probe
to determine the inductive response of CYP in fish (Goksøyr & Förlin 1992). EROD activity
has been measured in sea stars and plaice (Pleuronectes platessa) (data not shown) using
sensitive fluorimetric methods. Although lower than in flatfish, EROD activity was detected
in sea stars, further supporting the possible occurrence of vertebrate-like dioxin-
responsiveness in sea stars. CYP1A isozymes are known to carry out a large range of
activities (phase I AHH, P450RED, and phase II GST, UDPGT) (Van der Oost et al. 2003),
none of which have been addressed in the present study. In further research, it would thus be
interesting to characterize these activities in sea stars as it would complement information
provided by CYP1A IPP content measurements.
Conclusions-recommendations
The overall results obtained in this study have lead to question the existing international
regulations applying to monitoring PCBs and related compounds in the marine environment,
but also questionning the design of monitoring programmes. The selection of congeners, for
instance, should be revisited because of the pivotal influence of the congener planarity on
many aspect of PCBs ecotoxicology. The fact that coplanar PCB are generally not considered
for monitoring programmes is probably the result of historical analytical difficulties

212
GENERAL DISCUSSION
(insufficient detection limits, coelution,…) which have now been technically overcome. As
significant changes on essential aspects of sea stars biology were almost only attributable to
these compounds, they should imperatively be included in monitoring programmes. Results
obtained in the present study have also shown that designing programmes should be done
with caution regarding the sampling season, the sampling scale, the body compartments and
the congeners to be monitored.
The sea star Asterias rubens appeared to be quite resistant to a wide range of PCB levels.
However, contaminants levels measured in some areas of the southern North Sea subtly affect
its immune and endocrine systems, with relatively low short-term risk for this species. This
does not mean that other species in this region undergo the same low risk level, or that
ecosystems structured by sea stars may not become affected in the long term. More data is
needed to apprehend fine mechanisms and long-term variations, by continuously monitoring
this xenobiotic-stressed region, including the widest possible range of contaminants and
biological warning systems.

VI. REFERENCES
213
REFERENCES
Abad E, Llerena JJ, Saul J, Caixach J, Rivera J, 2002. Study on PCDDs/PCDFs and co-PCBs
content in food samples from Catalonia (Spain). Chemosphere 46:1435-1441
Adams WJ, Kimerle BA, Barnett JW jr, 1992. sediment quality and aquatic life assessment.
Environ Sci Technol 26 (10):1865-1875
Addison RF, Edwards AJ, 1988. Hepatic microsomal mono-oxygenase activity in flounder
Platichthys flesus from polluted sites in Langesundfjord and from mesocosms experimentally
dosed with diesel oil and copper. Mar Ecol Prog Ser 46: 51-54.
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 F, Yrjanheikki E,
1994. Toxic equivalency factors for dioxin-like PCBs. Chemosphere 28:1049-1067
Ahlborg UG, Brouwer A, Fingerhut MA, Jacobson JL, Jacobson SW, Kennedy SW, Kettrup
AAF, Koeman JH, Poiger H, Rappe C, Safe SH, Seegal RF, Tuomisto J, van den Berg M,
1992. Impact of polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls on human
and environmental health, with special emphasis on application of the toxic equivalency
factor concept. Eur J Pharmacol Environ Toxicol, Pharmacol Sect 228:179 199.
Alzieu C, Michel P, 1998. L’étain et les organoétains en milieu marin. Biogéochimie et
écotoxicologie, Repère océan (vol. 15). (Rapport IFREMER)
AMAP, 1998. AMAP Assessment Report: Arctic Pollution Issues. Arctic Monitoring and
Assessment Program (AMAP), Oslo, Norway, xiiq 859 pp. ISBN 82-7655-061-4.
AMAP, 2000. PCB in the Russian Federation: Inventory and Proposals for Priority Remedial
Actions. Executive Summary. AMAP Report 2000: 3, 2000. ISBN 82-7971-008-6. 27 pp.
Anderson RS, 1993. Modulation of nonspecific immunity by environmental stressors. In:
Couch JA, Fournie JW (eds): Advances in fisheries science, pathobiology of marine and
estuarine organisms, CRC press, London, pp 483-510
Anderson RS, Brubacher LL, Calvo LMR, Burreson EM, Unger MA, 1997. Effect of in vitro
exposure to tributyltin on generation of oxygen metabolites by oyster hemocytes. Environ Res
74:84-90.
Anderson RS, Mora LM, Thomson SA, 1994. Modulation of oyster (Crassostrea virginica)
hemocyte immune function by copper, as measured by luminol-enhanced chemiluminescence.
Comp Biochem Physiol 108(C):215-220.
Arukwe A, 2001. Cellular and molecular responses to endocrine-modulators and the impact
on fish reproduction Mar Pollut Bull 42(8):643-655
Aryoshi N, Yoshimura H, Oguri K, 1993. Identification of in vitro metabolites of
2,4,6,2’,4’,6’-hexachlorobiphenyl from phenobarbital-treated dog liver microsomes. Biol
Pharmacol Bull 16:852-857
Ashley CM, Simpson MG, Holdich DM, Bell DR, 1996. 2,3,7,8-tetrachloro-p-dioxin is a
potent toxin and induces cytochrome P450 in the crayfish Pacifastacus leniusculus. Aquat
Toxicol 35:157-169

214
REFERENCES
Atuma SS, Linder CE, Andersson Ö, Bergh A, Hansson L, Wicklund-Glynn A, 1996. CB153
as indicator for congener specific determination of PCBs in diverse fish species from Swedish
waters. Chemosphere 33:1459-1464
Babior M, 1984. The respiratory burst of phagocytes. J Clin Invest 73:599-601
Baier-Anderson C, Anderson RS, 2000. Immunotoxicity of environmental pollutants in
marine invertebrates. Recent Adv Mar Biotechnol 5:189-225.
Baier-Anderson C, Anderson RS, 2000. Immunotoxicity of environmental pollutants in
marine invertebrates. In: Fingerman M, Nagabhushanam R (eds), Recent advances in marine
biotechnology, Vol. 5 Immunobiology and pathology, Sceince Publishers Inc, Enfield (NH,
USA), pp.189-225
Balls PW, 1988. The control of trace metal concentrations in coastal seawater through
partition onto suspended particulate matter. Neth J Sea Res 22:213-218.
Ballschmiter K, Zell M, 1980. Analysis of polychlorinated biphenyls (PCBs) by glass
capillary chromatography. Fresnius Z Anal Chem 302: 20-31
Ballschmiter KH, Froescheis O, Jarman WM Caillet G, 1997. Contamination of the deep-sea.
Mar Pollut Bull 34(5):288-289
Barnes H, Blackstock J, 1973. Estimation of lipids in marine animals and tissues: detailed
investigation of the sulfophosphovanillin for «total lipids». J Exp Mar Biol Ecol 12, 103-118
Barron MG, 1990. Bioconcentration. Environ Sci Technol 24: 1612-1618
Barron MG, Yurk JJ, Crothers DB, 1994. Assessment of potential cancer risk from
consumption of PCBs bioaccumulated in fish and shellfish. Environ Health Perspect 102: 562
567.
Baukloh V, Bohnet HG, Trapp M, Heeschen W, Feichtinger W, Kemeter P. Biocides in
human follicular fluid. Ann NY Acad Sci. 1985;442:240-50.
Bayens WFJ, 1998. Trace metals in the Westerschelde estuary : a case study of a polluted,
partially anoxic estuary. Hydrobiologia (vol. 366) :167pp.
Bayne BL, Clarke KR, Cray JS, 1988. Biological effects of pollutants. Mar Ecol Prog Ser
(special edition), 278pp. (whole issue).
Becker G, 1990. Review of the physical oceanography of the North Sea. Neth J Sea Res
26:161-238.
Becker GA, Dick S, Dippner JW, 1992. Hydrography of the German Bight. Mar Ecol Prog
Ser (Special) 91:9-18.
Bedard DL, May RJ, 1996. Characterization of the polychlorinated biphenyls in the sediments
of the Woods Pond: evidence for microbial dechlorination of Aroclor 1260 in situ. Environ
Sci Technol 30:237-245.
Bello SM, Franks DG, Stegeman JJ, Hahn ME, 2001. Acquired resistance to aryl hydrocarbon
receptor agonists in a population of Fundulus heteroclitus from a marine Superfund site: in
vivo and in vitro studies on the induction of xenobiotic- metabolizing enzymes. Toxicol Sci
60:77-91.
Benedict JC, Lin TM, Loeffler IK, Peterson RE, Flaws JA, 2000. Physiological role of the
aryl hydrocarbon receptor in mouse ovary development. Toxicol Sci 56:382-388.

215
REFERENCES
Benisch PA, Hosoe K, Sakai S, 2001. Bioanalytical screening methods for dioxin and dioxinlike
compounds - a review of bioassay/biomarker technology. Environ Int 27:413-439
Bergen BJ, Nelson WG, Pruell RJ (1996) Comparison of nonplanar and coplanar PCB
congener partitioning in seawater and bioaccumulation in blue mussels (Mytilus edulis).
Environ Toxicol Chem 15:1517-1523
Bergqvist PA, Bergek S, Hallback H, Rappe C, Slorach SA, 1989. Dioxins in cod and herring
from the seas around Sweden. Chemosphere 19:513-516.
Bertheussen, K. (1981). Endocytosis by echinoid phagocytes in vitro I. Recognition of foreign
matter. Dev. Comp. Immunol. 5, 241-250.
Bidleman TF, Olney CE, 1973. Chlorinated hydrocarbons in the Sargasso Sea atmosphere and
surface water. Science 183 , pp. 516–51
Billen G, Joiris C, Meyer-Reil L, Lindeboom H, 1990. Role of bacteria in the North Sea
ecosystem. Neth J Sea Res 26:265-293.
Binyon J, 1978. Some observations upon the chemical composition of the starfish Asterias
rubens L with particular reference to strontium uptake. J Mar Biol Assoc UK 58:441-449.
Birkeland C, 1989. The influence of echinoderms on coral reef communities. In: Jangoux M,
Lawrence JM (eds), Echinoderm Studies, Vol. 3, Balkema, Rotterdam, pp. l-80.
Birnbaum LS, 1995. Workshop on perinatal exposure to dioxin-like compounds. V.
Immunologic effects. Environmental Health Perspectives 103 (Suppl.2):157–160.
Bjerregaard P, 1988. Effect of selenium on cadmium uptake in selected benthic invertebrates.
Mar Ecol Prog Ser 48: 17-28
Boese BL, Lee H, Echols S, 1997. Evaluation of a first order model for the prediction of the
bioaccumulation of PCBs and DDT from sediment into the marine deposit-feeding clam
Macoma nasuta, Environ Toxicol Chem 16(7):1545-1553
Boese BL, Specht DT, Pelletier J, Randall R, 1996. Evaluation of PCB and
hexachlorobenzene biota-sediment accumulation factors based on ingested sediment in a
deposit-feeding clam. Environ Toxicol Chem 15(9):1584-1589
Booij K, van den Berg C, 1994. Comparison of techniques for the extraction of lipids and
PCBs from benthic invertebrates. Bull Environ Contam Toxicol 53:71-76.
Booij K, Van der Meer J, Kwast D, De Boer JL, 1997. Sorption of polychlorinated biphenyls
by North Sea sediments. J Sea Res 37:49-66.
Boon JP, Van Zantvoort MB, Govaert MJMA, Duinker JC, 1985. Organochlorine in benthic
polychaetes (Nepthys spp.) and sediments from the southern North Sea. Identification of
individual PCB compounds. Neth J Sea Res 19:93-109.
Boon JP, Everaarts JM, Ten Hallers CC, Sleiderink HM, Beyer J, Goksøyr A, Klungsøyr J,
Wilhelmsen S, Halbrook RS, Shugart LR, Ariese F, Nieuwenhuize J, Van Liere JM, Vethaak
AD, Eggens ML, 1995. The integrated North Sea Programme 1991-92: theme
microcontaminants. Mar Environ Res: 39, 345-346
Borgmann J, Norwood WP, Ralph KM, 1990. Chronic toxicity and bioaccumulation of
2,5,2’,5’- and 3,4,3’,4’-tetrachlorobiphenyl and Aroclor 1242 in the amphipod Hyalella
azteca. Arch Environ Contam Toxicol 19:558-564

216
REFERENCES
Borlakoglu JT, Wilking JPG, 1993. Correlations between the molecular structures of
polyhalogenated biphenyls and their metabolism by hepatic microsomal monooxygenases.
Comp Biochem Physiol 105(C):113-117
Boucaud-Camou E, 1973. Etude de l’appareil digestif de Sepia officinalis L. (Mollusque
Céphalopode). Essai d’analyse expérimentale des phénomènes digestifs. PhD Thesis,
University of Caen, France, 208 pp.
Boucaud-Camou E, Boucher-Rodoni R, 1983. Feeding and digestion in Cephalopods. In:
Saleuddin ASM, Wilbur KM (eds) The Mollusca, vol. 5: Physiology, Part 2, New York and
London, Academic Press, pp. 149-187
Boucher-Rodoni R, Boucaud-Camou E, Mangold K, 1987. Feeding and digestion. In: Boyle
PR (ed) Cephalopod life cycles. Vol II. Comparative Reviews. Academic Press, New York,
London, p 85-108
Bradford MM, 1976. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Analyt Biochem 72:242-
254
Brattsen LB, 1979. Ecological significance of mixed-function oxidations. Drug. Metabol.
Rev. 10:35–58.
Bright DA, Dushenko WT, Grundy SL, Reimer KJ, 1995. Effect of local and distant
contaminant sources: polychorinated biphenyls and other organochlorines in bottom-dwelling
animals from the Arctic estuary. Sci Tot Environ 160/161:265-283
Broman D, Näf C, Rolff C, Zebühr Y, Fry B, Hobbie J, 1992. Using ratios of stable nitrogen
isotopes to estimate bioaccumulation and flux of polychlorinated dibenzo-p-dioxins (PCDDs)
and dibenzofurans (PCDFs) in two food chains from the northern Baltic. Environ Toxicol
Chem 11:331-345
Brouwer A, Van Den Berg KJ, 1986. Binding of a metabolite of 3,4,3’,4’-tetrachlorobiphenyl
to transthyretin reduced vitamin A transport by inhibiting the formation of the protein
complex carrying both retinol and thyroxin. Toxicol Appl Pharmacol 88:301–312.
Brown BE, 1982. The form and function of metal-containing ‘granules’ in invertebrate
tissues. Biol Rev Cam Philos Soc 57:621-667
Brown DG, Lanno RP, van den Heuvel MR, Dixon DG, 1995. HPLC determination of plasma
thiocyanate concentrations in fish blood: application to laboratory pharmacokinetic and fieldmonitoring
studies. Ecotox Environ Saf 30: 302-308
Brown ER, Sinclair T, Keith L, Beamer P, Hazdra JJ, Nair V, Callaghan O, 1977. Chemical
pollutants in relation to diseases in fish. Ann New York Acad Sci 298:535-546.
Brown JFJr, Bedard DL, Brennan MJ, Carnahan JC, 1987. Polychlorinated biphenyl
dechlorination in aquatic sediments. Science 236:709–712.
Brumley CM, Haritos VS, Ahokas J.T, Holdway DA, 1995. Validation of biomarkers of
marine pollution exposure in sand flathead using Aroclor 1254. Aquat Toxicol 31: 249-262.
Bruner KS, Fisher S, Landrum P, 1994. The role of the zebra mussel, Dreissena polymorpha,
in contaminant cycling: II. Zebra mussel contaminant accumulation from algae and
suspended particles, and transfer to the benthic invertebrate, Gammarus fasciatus. J Great
Lakes Res 20:735-750

217
REFERENCES
Brunner S, Hornung E, Santi H, Wolff E, Piringer OG, Altschuh J, Bruggeman R, 1990.
Henry’s law constants for polychlorinated biphenyls: Experimental determination and
structure–property relationships. Environ Sci Technol 24:1751-1754.
Bruno JF, Bertness MD, 2001. Habitat modification and facilitation in benthic marine
communities. In: Bertness MD, Gaines SD, Hay ME, editors. Marine community ecology.
Sunderland, MA: Sinauer Associates, Inc, pp. 201-218.
Bruschweiler BJ, Würgler FE, Fent K. 1996. An ELISA assay for cytochrome P4501A in fish
liver cells. Environ Toxicol Chem 15(4):592-596
Bryan GW, 1984. Pollution due to heavy metals and their compounds. In: Kinne O (ed):
Marine Ecology, vol5, Wiley J &sons, Chichester pp. 1289-1432
Bucheli TD, Fent K, 1995. Induction of cytochrome P450 as a biomarker for environmental
contamination in aquatic ecosystems. Crit Rev Environ Sci Technol 25:201-268
Buhler DR, Wang-Buhler JL, 1998. Rainbow trout cytochrome P450s: purification, molecular
aspects, metabolic activity, induction and role in environmental monitoring. Comp Biochem
Physiol 121(C):107-137
Burbach KM, Poland A, Bradfield CA, 1992. Cloning of the Ah receptor cDNA reveals a
distinctive ligand-activated transcription factor, Proc Natl Acad Sci USA 89:8185-8189.
Burch SW, Fitzpatrick LC, Goven AJ, Venables BJ, Giggleman MA, 1999. In vitro
earthworm Lumbricus terrestris coelomocyte assay for use in terrestrial toxicity identification
evaluation. Bull Environ Contam Toxicol 62:547-554.
Burckmeier BFA, Juttner I, Schramm KW, Winkler R, Steinberg CEW, Kettrup A, 1997.
PCBs and PCDD/Fs in lake sediments of Grosser Arbersee, Bavarian Forest, South Germany.
Environ Pollut 95:19-25
Burgeot T, Bocquéné G, Truquet P, Le Dean L, Galgani F, 1994. Induction of EROD activity
in red mullet (Mullus barbatus) along the French Mediterranean coasts. Sci Total Environ
142: 213-220.
Burgess RM, Schweitzer KA, McKinney RA, Phelps DK, 1992. Contaminated marine
sediments: water column and interstitial toxic effects. Environ Toxicol Chem 12:127-138.
Burke M, Mayer R, 1974. Ethoxyresorufin: direct fluometric assay of a microsomal Odealkylation
wich is preferentially inducible by 3- methylchloranthrene. Drug Metab Dispos
2: 583-588.
Burke RD, 1987. Echinoderm metamorphosis: comparative aspects of the change in form. In:
Jangoux M, Lawrence JM (eds) Echinoderm Studies. Vol 3, Balkema, Rotterdam pp 81-108.
Bustamante P, 1998. Etude des processus de bioaccumulation et de détoxication d'éléments
traces (métaux lourds et terres rares) chez les mollusques céphalopodes et bivalves pectinidés.
Implication de leur biodisponibilité pour le transfert vers les prédateurs. PhD Thesis,
University of La Rochelle, France, 290 pp.
Bustamante P, Caurant F, Fowler SW, Miramand P, 1998a. Cephalopods as a vector for the
transfer of cadmium to top marine predators in the north-east Atlantic Ocean. Sci Tot Environ
220:71-80
Bustamante P, Cherel Y, Caurant F, Miramand P, 1998b. Cadmium, copper and zinc in
octopuses from Kerguelen Islands, Southern Indian Ocean. Polar Biol 19:264-271

218
REFERENCES
Bustamante P, Cosson RP, Gallien I, Caurant F, Miramand P, 2002. Cadmium detoxification
processes in the digestive gland of cephalopods in relation to accumulated cadmium
concentrations. Mar Environ Res 53:227-241
Bustamante P, Grigioni S, Boucher-Rodoni R, Caurant F, Miramand P, 2000.
Bioaccumulation of 12 trace elements in the tissues of the nautilus Nautilus macromphalus
from New-Caledonia. Mar Pollut Bull 40(8):688-696
Butcher JB, Gauthier TD, Garvey EA, 1997, Use of historical PCB aroclors measurement:
hudson river fish data. Environ Toxicol Chem 16(8):1618-1623
Butler RB, Kelley ML, Powell WH, Hahn ME, Van Beneden RJ, 2001. An aryl hydrocarbon
receptor homologue from the soft-shell clam, Mya arenaria: evidence that invertebrate AHR
homologues lack TCDD and BNF binding. Gene 278:223-234.
Butty JS, Holdway DA, 1997. Assessment of the octopus Octopus pallidus, as a potential
bioindicator of xenobiotic challenge: baseline studies. Mar Pollut Bull 34(7): 564-570
Cairns J, 1983. Are single species toxicity tests alone adequate for estimating environmental
hazard? Hydrobiologia 100:47-57.
Cairns T, Doose GM, Froberg JE, Jacobson RA, Siegmund EG, 1989. Analytical chemistry of
PCBs. In: Waid JS (ed): PCBs and the environment. CRC Press, Boca Ration, FL.
Canicatti C, 1990. Lysosomal enzyme pattern in Holothuria polii coelomocytes. J Invert
Pathol 56:70-74.
Carr RS, Long ER, Windom HL, Chapman DC, Thursby G, 1996. Sediment quality
assessment studies of Tampa Bay, Florida. Environ Toxicol Chem 15, 1218-1231
Castro BG, Guerra A, 1990. The diet of Sepia officinalis (Linnaeus 1758) and Sepia elegans
(D’Orbigny 1835) (Cephalopoda, Sepioidea) from the Ria de Vigo (NW Spain). Scientia
Marina 54, 375-388
Celander M, Förlin L. 1991. Catalytic activity and immunochemical quantification of hepatic
cytochrome P450 in beta-naphtoflavone and isosafrol treated rainbow trout (Oncorhyncus
mykiss). Fish Physiol Biochem 9:189-197
Chapman PM, 1995. Ecotoxicology and Pollution: key issues. Mar Pollut Bull 31:167-177
Chapman PM, 1996. A test of sediment effects concentrations: DDT and PCB in the southern
California bight. Environ Toxicol Chem 15(7):1197-1198
Cheah DMY, Wright PFA, Holdway DA, Ahokas JT, 1995. Octopus pallidus cytochrome P-
450: characterization and induction studies with ß-naphthoflavone and Aroclor 1254. Aquat
Toxicol 33, 201-214
Chia FS, Xing J, 1996. Echinoderm coelomocytes. Zool Studies 35:231-254
Chiu A, Chiu N, Beaubier J, Nalsnik R, Singh D, Hill WR, Lau C, Riebow J, 2000. Effects
and mechanisms of PCB ecotoxicity in food chains: algae=>fish=>seal=>polar bear. J
Environ Sci Health C18(2):127-152
Christensen ER, Lo CK, 1986. Polychlorinated biphenyls in dated sediments of Milwaukee
Harbour, Wisconsin, USA. Environ Pollut (Series B) 12:217-232.
Clark RB, 1997. Marine Pollution. Oxford University Press, 161 pp.
Clarke MR, 1996. Cephalopods as prey. III. Cetaceans. Philosophical Transactions of the
Royal Society of London Ser B 351, 1053-1065

219
REFERENCES
Clements WH, Kiffney PM. Assessing contaminant effects at higher levels of biological
organization. Environ Toxicol Chem 1994;13:357-359.
Cofino WP, Smedes F, de Jong SA, Abarnou A, Boon JP, Oostingh I, Davies IM, Klungsøyr
J, Wilhelmsen S, Law RJ, Whinnett JA, Schmidt D, Wilson S (1992) The chemistry
programme. Mar Ecol Prog Ser (Special) 91:47-56.
Coles JA, Farley SR, Pipe RK, 1995. Alteration of the immune response of the common
marine mussel Mytilus edulis resulting from exposure to cadmium. Dis Aquat Org 22:59-65.
Conover WJ, 1980. Practical nonparametric statistics. Wiley & Sons, NY, 493 pp.
Coombs TL, George SG (1978) Mechanisms of immobilisation and detoxification of metals
in marine organisms. In: Mclusky DS & Berry AJ (eds) Physiology and behaviour of marine
organisms. Pergamon Press, New York, p 179-187
Cooper EL, 1992. The capacities of earthworms to heal wounds and to destroy allografts are
modified by polychlorinated biphenyls (PCB). J Invert Pathol 60:59-63.
Cooper KR, 1989. Efects of polychlorinated dibenzo-p-dioxins and polychlorinated
dibenzofurans on aquatic organisms. Rev Aquat Sci 1:227-242
CoT, 2001. Committee on Toxicity of Chemicals in Food, Consumer Products and the
Environment. Statement on the tolerable daily intake for dioxins and dioxin-like
polychlorinated biphenyls. Available at:
http://www.foodstandards.gov.uk/committees/cot/cot_dioxin_stat.pdf.
Coteur G, Danis B, Fowler SW, Teyssié JL, Dubois Ph, Warnau M, 2001. Effects of PCBs on
reactive oxygen species (ROS) production by the immune cells of Paracentrotus lividus
(Echinodermata). Mar Pollut Bull 42(8):667-672
Coteur G, 2002a. The cellular immunity of Asterias rubens L. (Echinodermata):
characterization and impact of metal contaminations. PhD thesis (in French). Université Libre
de Bruxelles, Brussels. 179 pp.
Coteur G, DeBecker G, Warnau M, Jangoux M, Dubois Ph, 2002b. Differentiation of immune
cells challenged by bacteria in the common European starfish, Asterias rubens
(Echinodermata). Eur J Cell Biol 81:413-418.
Coteur G, Warnau M, Jangoux M, Dubois Ph, 2002c. Reactive oxygen species (ROS)
production by amoebocytes of Asterias rubens (Echinodermata). Fish & Shellfish Immunol.
12(3):187-200
Coteur G, Gosselin P, Wantier P, Chambost-Manciet Y, Warnau M, Danis B, Pernet Ph,
Dubois Ph, 2003a. Echinoderms as bioindicators, bioasssays and impact assessment tools of
sediment associated metals and PCBs in the North Sea, Arch. Environ Toxicol Chem 45
(2):190-202
Coteur G, Pernet Ph, Gillan D, Joly G, Maage A, Dubois Ph, 2003b. Field contamination of
the starfish Asterias rubens by metals. Part 1: Short- and long-term accumulation along a
pollution gradient. Environ Toxicol Chem 22(9):2136-2144
Coteur G, Gillan D, Joly G, Pernet Ph, Dubois, 2003c. Field contamination of the starfish
Asterias rubens by metals. Part 2: Effects on the cellular immunity. Environ Toxicol Chem
22(9):2145-2151
Coteur G, Corriere N, Dubois Ph, 2004. Environmental factors influencing the immune
responses of the common European starfish (Asterias rubens), Fish & Shellf Immunol 16:51-
63

220
REFERENCES
Courtney WAM, Langston WJ, 1978. Uptake of polychlorinated biphenyls (Aroclor 1254)
from sediment and from seawater in two intertidal polychaetes. Environ Pollut 15: 303-309
Creutzberg F, Wapenaar P, Duineveld G, Lopez Lopez N, 1984. Distribution and density of
the benthic fauna in the Southern North Sea in relation to bottom characteristics and
hydrographic conditions. Rapp P-v Réun Cons Perm Int Explor Mer 183:226-233.
Croxall JP, Prince PA, 1996. Cephalopods as prey. I. Seabirds. Phil Trans Roy Soc London
Ser B 351, 1023-1043
Cummins JE, 1988. Extinction: the PCB threat to marine mammals. Ecologist 18(6):193-195.
Daan R, Lewis WE, Mulder M, 1990. Biological effects of discharged oil-contaminated drill
cuttings in the North Sea (Report NIOZ 1990-5). Netherlands Institute for Sea
Research/NIOZ. Den Burg, Texel, The Netherlands.
Dahlgaard H, 1981. Bioindicators for monitoring radioactive pollution of the marine
environment-experiments on the feasibility of Mytilus as a bioindicator in estuarine
environments with some comparisons to Fucus. Risø Reports 15. Risø, Roskilde, Denmark
Daston GP, Gooch JW, Breslin WJ, Shuey DL, Nikiforov AI, Fico TA, Gorsuch JW, 1997.
Environmental estrogens and reproductive health: A discussion of the human and
environmental data. Reproductive Toxicol 11:465-481.
Daumas RA, Laborde PL, Marty JC, Saliot A, 1976. Influence of sampling method on the
chemical composition of water surface film. Limnol Oceanogr 21:319-326
de Boer J, Stronck CJN,Traag WA,van der Meer J, 1993. Non-ortho and mono-ortho
substituted chlorobiphenyls and chlorinated dibenzo-p-dioxins and dibenzofurans in marine
and freshwater fish and shellfish from the Netherlands. Chemosphere 26,1823-18 42.
de Boer J, Van der Valk F, Kerkhoff MAT, Hagel P (1994) 8-Year study on the elimination
of PCBs and other organochlorine compounds from eel (Anguilla anguilla) under natural
conditions. Environ Sci Technol 28:2242-2248.
Dean M, Hamon Y, Chimini G, 2001. The human ATP-binding cassette (ABC) transporter
superfamily, J. Lipid Res. 42:1007–1017.
Deiana L, Congiu AM, Arena N, 1983. Effetti embrio-tossici del nickel solfato sul
Paracentrotus lividus: osservazioni preliminari. Bolletino della Societa Italiana di Biologia
Sperimentale 11:1718-1724
den Besten PJ, 1998a. Cytochrome P450 monooxygenase system in echinoderms. Comp
Biochem Physiol 121(C):139-146
den Besten PJ, 1998b. Concepts for the implementation of biomarkers in environmental
monitoring. Mar Environ Res 46(1-5): 253-256
den Besten PJ, Herwig HJ, Zandee DI, Voogt PA, 1989. Effect of cadmium and PCBs on
reproduction of the sea star Asterias rubens: aberration in the early development. Ecotoxicol
Environ Saf 18:173-180
den Besten PJ, Hervig HJ, Zandee DI, Voogt PA, 1990a. Cadmium accumulation and
metallothionein-like proteins in the sea star Asterias rubens. Arch Env Contam Toxicol
19:856-862.
den Besten PJ, Herwig HJ, Van Donselaar EG, Livingstone DR, 1990b. Cytochrome P450
monooxygenase system and benzo(a)pyrene metabolism in echinoderms. Mar Biol 107:171-
177

221
REFERENCES
den Besten PJ, Elenbaas JML, Maas JR, Dieleman SJ, Herwig HJ Voogt PA, 1991. Effects of
cadmium and polychlorinated biphenyls (Clophen A50) on steroid metabolism and
cytochrome P-450 monooxygenase system in the sea star Asterias rubens L, Aquat Toxicol
20 (1-2):95-110
den Besten PJ, Lemaire P, Livingstone DR, Woodin B, Stegeman JJ, Herwig HJ, Seinen W.
1993. Time-course and dose-response of the apparent induction of the cytochrome p450
monooxygenase system of pyloric caeca microsomes of the female sea star Asterias rubens L.
by benzo(a)pyrene and polychlorinated biphenyls. Aquat Toxicol 26:23-40
den Besten PJ, Valk S, van Weerlee E, Nolting RF, Postma JF, Everaarts JM, 2001.
Bioaccumulation and biomarkers in the sea star Asterias rubens (Echinodermata: Asteroidea):
a North Sea field study. Mar Environ Res 51:365-387
Denison MS, Seidel SD, Rogers WJ, Ziccardi M, Winter GM, Heath-Pagliuso S, 1998.
Natural and synthetic ligands for the Ah receptor, in: A. Puga, K. Wallace (Eds.), Molecular
Biology of the Toxic Response, Taylor & Francis, Philadelphia, pp. 393-410
Dickhut RM, Andren AW, Armstrong DE, 1986. Aqueous solubilities of six polychlorinated
biphenyl congeners at four temperatures. Environ Sci Technol 20:807-810.
Dikkeboom R, Tijnagel J, Mulder E, Van Der Knaap W, 1987. Hemocytes of the pond snail
Lymnea stagnalis generate reactive forms of oxygen. J Invert Pathol 49:321-331.
Dinnel PA, Pagano G, Oshida PS, 1988. A sea urchin test system for marine environmental
monitoring. In: Burks RD, Mladenov PV, Lambert PV, Parsley RL (eds): echinoderm
biology, Balkema, Rotterdam,
Duce RA, Quinn JG, Olney CE, Piotrowicz SR, Ray BJ, Wade TL, 1972. Enrichment of
heavy metals and organic compounds in the surface microlayer of Narraganset Bay, Rhode
Island. Science 176:161-163
Duffy JE, Carlson E, Li Y, Prophete C, Zelikoff JT, 2002. Impact of polychlorobiphenyls
(PCBs) on the immune function of fish: age as a variable determining adverse outcome. Mar
Environ Res 54:559-563
Duinker JC, Schultz DE, Petrick G, 1988. Selection of polychlorinated biphenyl congeners
for analysis in environmental samples. Mar Pollut Bull 19:19-25
Duinker JC, Hillebrandt MTJ, Zeinstra T, Boon JP, 1989. Individual chlorinated biphenyls
and pesticides in tissues of some cetacean species from the North Sea and the Atlantic Ocean;
tissue distribution and biotransformation. Aquat Mamm15:95-124.
Duinker JC, Schulz DE, Petrick G, 1991. Analysis and interpretation of chlorobiphenyls:
possibilities and problems. Chemosphere 23:1009-1028.
Duncan DM, Burgess EA, Duncan I, 1998. Control of distal antennal identity and tarsal
development in Drosophila by spineless-aristapedia, a homolog of the mammalian dioxin
receptor, Genes Dev 12:1290-1303
Dyrynda EA, Law RJ, Dyrynda PEJ, Kelly CA, Pipe RK, Ratcliffe NA, 2000. Changes in
immune parameters of natural mussel Mytilus edulis populations following a major oil spill
(‘Sea Empress’, Wales, UK). Mar Ecol Prog Ser 206:155-170
Eganhouse RP, Gossett RW, 1991. Sources and magnitude of bias associated with
determination of polychlorinated biphenyls in environmental samples. Anal Chem
63:2130–2137.

222
REFERENCES
Eggens M, Bergman A, Vethaak D, 1995. Seasonal variation of hepatic EROD activity in
flounder (Platichtys flesus) in the dutch Wadden Sea. Mar Environ Res 39:231-234
Eggens M, Galgani F, 1992. Ethoxyresorufin-O-deethylase (EROD) activity in flatfish: fast
determination with a fluorescence plate reader. Mar Environ Res 33: 213-221.
Eisma D, 1973. Sediment distribution in the North Sea in relation to marine pollution. In:
Goldberg, D. (ed.), North Sea Science. The MIT Press, Cambridge, pp. 131–152.
Eisma D, Kalf J, 1987. Dispersal, concentration and deposition of suspended matter in the
North Sea. J Geol Soc London 144, 161–178.
Eisma D, Irion G, 1988. Suspended matter and sediment transport. In: Salomons, W., Bayne,
B.L., Duursma, E.K., Fo¨rstner, U. (eds.), Pollution of the North Sea: an Assessment.
Springer-Verlag, Berlin, pp. 20–35.
Elder D, 1976. PCBs in NW Mediterranean coastal waters. Mar Pollut Bull 7:63-64
Elijarrat E, Caiyach J, Rivera J, De Torres M, Cinebrada A, 2001. Toxic potency assessment
of non- and mono-ortho PCBs, PCDDs, PCDFs and PAHs in the Northwestern Mediterranean
sediments (Catalonia, Spain). Environ Sci Technol 35:3589-3594
Elskus AA, Stegeman JJ, Susani LC, Black D, Pruell RJ, Fluck SJ, 1989. Polychlorinated
biphenyls concentration and cytochrome P450E expression in winter flounder from
contaminated environments. Mar Environ Res 28: 25-30
Elskus AA, Pruell RJ, Stegeman JJ, 1992. Endogenously-mediated, pretranslational
suppression of cytochrome P4501A in PCB-contaminated flounders. Mar Environ Res 34:97-
101
Ema M, Sogawa K, Watanabe N, Chujoh Y, Matsushita N, Gotoh O, Funae Y, Fujii-
Kuriyama Y, 1992. cDNA cloning and structure of mouse putative Ah receptor, Biochem.
Biophys Res Commun 184:246–253.
Environment Canada, 1992. Biological test method: fertilization assay using echinoids (sea
urchins and sand dollars). Report EPS 1/RM/27. Environment Canada, Ottawa.
Eppe G, 1996. Mise au point d’une méthode d’analyse des dioxines dans des émissions
industrielles. Mémoire présenté en vue de l’obtention du grade d’Ingénieur Civil Chimiste,
University of Liège, 65 pp.
Everaarts JM, Fischer CV, 1989. Micro contaminants in surface sediments and macrobenthic
invertebrates of the North Sea. NIOZ-Rapport (6). Nederlands Instituut voor Onderzoek der
Zee. North Sea Benthos Survey (ICES).
Everaarts JM, Otter E, Fischer CV, 1990. Cadmium and polychlorinated biphenyls: different
distribution patterns in North Sea benthic biota. Neth J Sea Res 26:75-82
Everaarts JM, Sarkar A, 1996. DNA damage as a biomarker of marine pollution: strand
breaks in seastars (Asterias rubens) from the North Sea. Wat Sci Tech 34(7-8):157-162
Everaarts JM, den Besten PJ, Hillebrand MThJ, Halbrook RS, Shugart LR, 1998. DNA strand
breaks, cytochrome P-450-dependant monooxygenase system activity and levels of
chlorinated biphenyl congeners in the pyloric caeca of the sea star (Asterias rubens) from the
North Sea. Ecotoxicology 7:69-79
Everaarts JM, Booij K, Fischer CV, Maas YEM, Nieuwenhuize J, 1999. Ecology of the
Chagos Archipelago. CRC Sheppard, MRD Seaward.
Falandysz J, Kannan K, Kawano M, Rappe C, 2000. Relative contribution of chlorinated

223
REFERENCES
naphtalenes,-biphenyls,- dibenzofurans and -dibenzo-p-dioxins to toxic equivalents in biota
from the south coast of the Baltic Sea. Organohalogen Compd 47:9-12.
Feldman KL, Armstrong DA, Dumbauld BR, Dewitt TH, Doty DC, 2000. Oysters, crabs, and
burrowing shrimp: review of an environmental conflict over aquatic resources and pesticide
use in Washington State’s (USA) coastal estuaries. Estuaries 23:141-176.
Fernandez-Salguerro PM, Hilbert DM, Rudikoff S, Ward JM, Gonzalez FJ, 1996.
Arylhydrocarbon receptor-deficient mice are resistant to 2,3,7,8-tetrachlorodibenzo-p-dioxin
induced toxicity. Toxicol Appl Pharmacol 140:73-179.
Fernandez-Salguero PM, Ward JM, Sundberg JP, Gonzalez FJ, 1997. Lesions of arylhydrocarbon
receptor-deficient mice. Vet Pathol 34:605-614.
Fichet D, Radenac G, Miramand P, 1998. Experimental studiesof impacts of harbour
sediments resuspension to marine invertebrates larvae: bioavailability of Cd, Cu, Pb and Zn.
Mar Pollut Bull 36(7-12):509-518
Finger JM, Smith JD, 1987. Molecular association of Cu, Zn, Cd and 210Po in the digestive
gland of the squid Nototodarus gouldi. Mar Biol 95:87-91
Fischer WS, 1988. Environmental influence on bivalve hemocyte function. In: Fischer WS
(ed): disease processes in marine bivalve molluscs. American fisheries society special
publication, Vol l18:225-237
Fisher NS, Nolan CV, Fowler SW, 1991. Scavenging and retention of metals by zooplankton
fecal pellets and marine snow. Deep Sea Res 38:1261-1275
Fisher NS, Teyssié JL, Fowler SW, Wang WX, 1996. Accumulation and retention of metals
in mussels from food and water: a comparison under field and laboratory conditions. Env Sci
Technol 30(11):3232-3242
Fitzpatrick LC, Sassani R, Venables BJ, Goven AJ, 1992. Comparative toxicity of
polychlorinated biphenyls to earthworms Eisenia foetida and Lumbricus terrestris. Environ
Pollut 77:65-69
Focant JF, Eppe G, Pirard C, De Pauw E, 2001. Fast clean-up for polychlorinated dibenzo-pdioxins,
dibenzofurans and coplanar polychlorinated biphenyls analysis of high-fat content
biological samples. J Chromato 925(A):207-221
Förlin L, Balk L, Celander M, Bergek S, Hjelt M, Rappe C, de Wit C, Jansson B. 1992.
Biotransformation enzyme activities and PCDD/F levels in pike caught in a Swedish lake.
Mar Environ Res 34:169-173
Forsythe JW, Heukelem WF, 1987. Growth, In: Boyle, P.R. (ed), Cephalopod life cycles,
vol. 2, Academic Press London, pp. 351-365
Fowler SW, 1990. Critical review of selected heavy metal and chlorinated hydrocarbon
concentrations in the marine environment. Mar Environ Res 29:1-64.
Fowler SW, Oregioni B, 1976, Trace metals in mussels from the NW Mediterranean, Mar.
Pollut. Bull. 7:26-29
Fowler SW, Polikarpov GG, Elder DL, Parsi P & Villeneuve JP, 1978. Polychlorinated
biphenyls: accumulation from contaminated sediments and water by the polychaete Nereis
diversicolor. Mar Biol 48:303-309
Friberg L, 1988. The GESAMP evaluation of potentially harmful substances in fish and other
seafood with special reference to carcinogenic substances. Aquat Toxicol 11:379-393

224
REFERENCES
Furukawa K, 1982. Microbial degradation of polychlorinated biphenyls. In A.M.
Chakrabarty, ed., Biodegradation and Detoxification of Environmental Pollutants. CRC, Boca
Raton, FL, USA, pp. 32-57
Galgani F, Bocquéné G, Luçon M, Grzebyk D, Letrouit F, Claisse D, 1991. Erod
measurements in fish from the Northwest part of France. Mar Pollut Bull 22: 494-500.
Galgani F, Payne JF, 1991. Biological effects of contaminants: microplate method for
measurement of ethoxyresorufin-O-deethylase (EROD) in fish. Palægade 2-4, Copenhagen,
Denmark. ICES Tech Mar Environ Sci 13: 11 pp.
Galgani F, Bocquéné G, Truquet P, Burgeot T, Chiffoleau JF, Claisse D, 1992. Monitoring of
pollutant biochemical effects on marine organisms of the French coasts. Oceanol Acta 15:
355-364
Garabetian F, Romano JC, Paul R, Sigoillot JC, 1993. Organic matter composition and
pollutant enrichment of sea surface microlayer inside and outside slicks. Mar Environ Res
35:323-339
George SG, Olsson PE, 1994. Metallothioneins as indicators of trace metal pollution. In:
Framer KJM (ed), Biomonitoring of coastal waters and estuaries. CRC, Boca Raton, FL. Pp.
151-171
Gerpe MS, de Moreno JEA, Moreno VJ, Patat ML, 2000. Cadmium, zinc and copper
accumulation in the squid Illex argentinus from the Southwest Atlantic Ocean. Mar Biol
136:1039-1044
Gibson GC, Skett P (eds), 1994. Introduction to drug metabolism. 2 nd edition, Blackwell
academic and professional, Chapman & Hall, Glasgow, 266 pp.
Gillesby BE, Zacharewsky TR, 1998. Exoestrogens: mechanisms of action and strategies for
identification and assessment. Environ Toxicol Chem 17(1):3-14
Gnezdilova SM, Khristoforova NK, Lipina IG, 1985. Gonadotoxic and embryotoxic effects of
cadmium in sea urchins. Symposia Biologica Hungarica 2929:239-251
Goerke H, Ernst W, 1977, Fate of 14C-labelled di-, tri-, and pentachlorobiphenyl in the
marine annelid Nereis virens. Chemosphere 9:551-558
Goerlitz DF, Law LM, 1971. Note on removal of sulfur interferences from sediment extracts
for pesticide analysis. Bull Environ Contam Toxicol 6:9-10.
Goksøyr A, 1991. Semi quantitative cytochrome P450 1A1 ELISA: a simple method for
studying the monooxygenase induction response in environmental monitoring and
ecotoxicological testing of fish. Sci Tot Environ 101:253-261
Goksøyr A, 1995. Use of cytochrome P4501A (CYP1A) in fish as a biomaker of aquatic
pollution. Arch Toxicol 17:80-95
Goksøyr A, Husøy AM, Larsen HE, Klungsøyr J, Wilhelmsen S, Maage A, Brevik EM,
Andersson T, Celander M, Pesonen M, Förlin L, 1991. Environmental contaminants and
biochemical responses in flatfish from the Hvaler Archipelago in Norway. Arch Environ
Contam Toxicol 21:486-495
Goksøyr A, Förlin L, 1992. The cytochrome P450 system in fish, aquatic toxicology and
environmental monitoring. Aquat Toxicol 22:287-312
Goldberg ED, Bowen VT, Farrington JW, Harvey G, Martin JH, Parker PL, Risebrough RW,
Robertson W, Schneider E, Gamble E, 1978. The mussel watch. Environ Conservation 5:

101-125
225
REFERENCES
Golden RJ, Noller KL, Ernstoff LT, Kaufman RH, Mittendorf R, Stillman R, Reese EA,
1998. Environmental endocrine modulators and human health: an assessment of the biological
evidence. Crit Rev Toxicol 28:109-227.
Goldstein BD, 1995. The concept of biological markers in the field of risk assessment. Stem
Cells 13(suppl.):30-32
Goldstein JA, Hickman P, Bergman H, McKinney JD, Walker MP, 1977, Separation of pure
polychlorinated biphenyl isomers into two types of inducers on the basis of induction of
cytochrome P-450 or P-448, Chem-Biol Interact 17:69–87.
Gonzalez AF, Guerra A, Pascual S, Briand P, 1998. Vulcanoctopus hydrothermalis gen. et sp.
nov. (Mollusca, Cephalopoda): an octopod from a deep-sea hydrothermal vent site. Cah Biol
Mar 39:169-184
Gonzalez FJ, Nebert DW, 1990. Evolution of the P450 gene superfamily, Trends Genet
6:182-186.
Gooch JA, Hamdy MK, 1982, Depuration and biological half-life of 14 C-PCB in aquatic
organisms. Bull Environ Contam Toxicol 28:305-312
Gooch JW, Elskus AA, Kloeper-Sams PJ, Hahn ME, Stegeman JJ. 1989. Effects of ortho and
non-ortho substituted polychlorinated biphenyl congeners on the hepatic monooxygenase
system in scup (Stenomus chrysops). Toxicol Appl Pharmacol 98:422-427
Gosselin P, Jangoux M, 1998. From competent larvae to exotrophic juvenile: a
morphofunctional study of the perimetamorphic period of Paracentrotus lividus
(Echinodermata: Echinoidea). Zoomorphol 118:31-43.
Gould JC, Cooper KR, Scanes CG. 1997. Effects of polychlorinated biphenyl mixtures and
three specific congeners on growth and circulating growth-related hormones. Gen Comp
Endocrinol 106(2):221-230
Goven AJ, Eyambe GS, Fitzpatrick LC, Venables BJ, Cooper EL, 1993. Cellular biomarkers
for measuring toxicity of xenobiotics: effects of polychlorinated biphenyls on earthworm
Lumbricus terrestris coelomocytes. Environ Toxicol Chem 12:863-870.
Goven AJ, Fitzpatrick LC, Venables BJ, 1994. Chemical toxicity and host defence in
earthworms: an invertebrate model. In: Beck G, Cooper EL, Habicht GS, Marchalonis JJ
(eds), Primordial Immunity: Foundations for the Vertebrate Immune System, Annals of the
New York Academy of Sciences, volume 712, The New York Academy of Sciences, New
York, USA, pp. 280-300
Gray JS, 1989. Do bioassays adequately predict ecological effects of pollutants?
Hydrobiologia 188/189:397-402
Green JD, Mwatibo JM, Swartz WJ, 1997. The effects of methoxychlor on early sea urchin
development. Environ. Res. 72:56-64
Green NM, 1965. A spectrophotometric assay for avidin and biotin based on binding dyes by
avidin. Biochem J 94:23c-24c.
Gross PS, Al-Sharif WZ, Clow LA, Smith LC, 1999. Echinoderm immunity and the evolution
of the complement system. Develop Comp Immunol 23:429-442

226
REFERENCES
Grzebyk D, Galgani F, 1991. Measurement of the effect of organic pollution on marine
organisms: rapid determinantion of EROD induction using plate reader. Aquat Living Resour
4: 53-59.
Guary JC, 1980. Recherches sur les transferts et la fixation du plutonium, de l'américium et
du neptunium dans le milieu marin. PhD Thesis, University of Aix-Marseille II, France, 303
pp.
Guary JC, Higgo JJW, Cherry RD, Heyraud M, 1981. High concentrations of transuranic and
natural radioactive elements in the branchial hearts of the cephalopods Octopus vulgaris. Mar
Ecol Prog Ser 4:123-126
Guary JC, Fowler SW, 1982. Experimental studies on the biokinetics of plutonium and
americium in the cephalopod Octopus vulgaris. Mar Ecol Prog Ser 7:327-335
Guns M, Van Hoeyweghen P, Vyncke W, Hillewaert H, 1999. Trace metals in selected
benthic invertebrates from Belgian coast waters (1981-1996). Mar Pollut Bull 38(12):1184-
1193.
Gunther AJ, Spies RB, Stegeman JJ, Woodin B, Carney D, Oakden J, Hain L, 1997. EROD
activity in fish is an independent measure of contaminated-induced mortality of invertebrates
in sediment bioassays. Mar Environ Res 44: 41-49.
Hahn ME, 1998. The aryl hydrocarbon receptor: a comparative perspective. Comp Biochem
Physiol 121(C):23-53
Hahn ME, 2002a. Biomarker and bioassays for detecting dioxin-like compounds in the
marine environment. Sci Tot Environ 289:49-69
Hahn ME, 2002b. Aryl hydrocarbon receptors: diversity and evolution. Chem Biol Interact
141:131-160
Hahn ME, Lamb TM, Shultz ME, Smolowitz RM, Stegeman JJ, 1993. Cytoochrome P4501A
induction and inhibition by 3,3’,4,4’-tetrachlorobiphenyl in an Ah receptor-containing fish
hepatoma cell line (PHLC-1). Aquat Toxicol 310:218-228
Hahn ME, Woodin BR, Stegeman JJ, Tillitt DE. 1998. Aryl hydrocarbon receptor function in
early vertebrates: inducibility of cytochrome P450 in agnathan and elasmobranch fish. Comp.
Biochem. Physiol. 120(C):67-75
Hansen SN, Bjerregaard P, 1995. Manganese kinetics in the sea star Asterias rubens (L.)
exposed via food or water. Mar Pollut Bull 31:127-132.
Harding GC, Addison RF, 1986. Accumulation and effects of PCBs in marine invertebrates
and vertebrates. In JS Wood (Ed) PCBs and the Environment, Vol II CRC Press, 1986, pp. 9-
30
Hardy JT, Cleary J, 1992. Surface microlayer contamination and toxicity in the German
Bight. Mar Ecol Prog Ser 91:203–210.
Hardy JT, Coley JA, Antrim LD, Kiesser SL, 1988. A hydrophobic large-volume sampler for
collecting aquatic surface microlayers: characterization and comparison with the glass plate
method. Can J Fish Aquat Sci 45:822-826.
Hardy JT, Crecelius EA, Antrim LD, Kiesser SL, Broadhurst PD, Boehm PD, Steinhauer
WG, Coogan TH, 1990. Aquatic surface microlayer contamination in Chesapeake Bay. Mar
Chem 28:333-351

227
REFERENCES
Harper N, Connor K, Steinberg M, and Safe S, 1995. Immunosuppressive activity of
polychlorinated biphenyl mixtures and congeners: Nonadditive (antagonistic) interactions.
Fundam Appl Toxicol 27:131–139.
Harrold C, Pearse JS, 1987. The ecological role of echinoderms in kelp forests. In: Jangoux
M, Lawrence JM (eds), Echinoderm studies, Balkema, Rotterdam, The Netherlands, pp 137-
233
Harvey GR, Steinhauer WG, 1976. Biogeochemistry of PCB and DDT in the North Atlantic.
In: Nriagu JO (ed) Environmental Biogeochemistry, vol. 1. Ann Arbor Science Publ, Ann
Arbor, pp. 203-221
Haynes D, Müller J, Carter S, 2000. Pesticide and herbicide residues in sediments and
seagrasses from the great barrier reef world heritage area and Queensland Coast. Mar Pollut
Bull 41:279-287.
Haynes D, Toohey D, 1995. Temporal variation in polychlorinated dibenzo-pdioxins,dibenzofu
rans,extractable organohalogens (EOX) and heavy metals in commercially
cultured mussels (Mytilus edulis) from Port Philip Bay, Victoria,Austalia. Mar Pollut Bull
30:885-891
Hayward JM, Ryland JS, 1990. The marine fauna of the British Isles and north western
Europe. II: Molluscs to Chordates. Oxford Science Publications, NY, 627 pp.
Hewitt S, Fenet H, Casellas C, 1998. Induction of EROD activity in European eel (Anguilla
anguilla) by different polychlorobiphenyls. Wat Sci Technol 38:245-252.
Higginbotham GR, Huang A, Firestone D, Verrett J, Ress J, Campbell AD, 1968. Chemical
and toxicological evaluations of isolated and synthetic chloro derivatives of dibenzo-p-dioxin,
Nature 220:702-703.
Hinton DE, 1994. Cells, cellular responses, and their markers in chronic toxicity of fishes. In:
Malins DC, Ostrander GK (Eds.), Aquatic Toxicology; Molecular, Biochemical and Cellular
Perspectives. Lewis Publishers CRC press, pp. 207-240.
Hinton DE, Baumann PC, Gardner GC, Hawkins WE, Hendricks JD, Murchelano RA,
Okihiro MS, 1992. Histopathologic biomarkers. In: Huggett RJ, Kimerly RA, Mehrle PMJr,
Bergman HL (Eds), Biomarkers: Biochemical, Physiological and Histological Markers of
Anthropogenic Stress. Lewis Publishers, Chelsea, MI, USA, pp. 155-210.
Hodgson E, Levi P, 1993. Introduction to biochemical toxicity, 2nd edition, Norwalk CT:
Appleton and Lange,
Hoff JT, Mackay D, Gillham R, Shiu WY, 1993. Partitioning of organic chemicals at the
air–water interface in environmental systems. Environ Sci Technol 27:2174–2180.
Holdway DA, Brennan SE, Ahokas JT, 1994. Use of hepatic MFO and blood enzyme
biomarkers in sand flathead (Platycephalus bassensis) as indicators of pollution in Port
Phillip Bay, Australia. Mar Pollut Bull 28: 683-695.
Hong CS, Xiao J, Bush B, Shaw SD, 1998. Environmental occurrence and potential toxicity
of planar, mono-, and di-ortho polychlorinated biphenyls in the biota. Chemosphere,
36(7):1637-1651
Honkakoski P, Negishi M, 2000. Regulation of cytochrome P450 (CYP) genes by nuclear
receptors, Biochem J 347:321-337.

228
REFERENCES
Hostens K, Hammerlinck O, 1994. The mobile epifauna of the soft bottoms in the subtidal
Oosterschelde estuary: structure, function and impact of the storm-surge barrier.
Hydrobiologia 282/283:479-496
Howard P, Boethling R, Jarvis W, Meylan W, Michalenko EN, 1991. Handbook of
Environmental Degradation Rates. Lewis Publishers, Chelsea, USA,
Hubell SP, Sikora A, Paris OH, 1965. Radiotracer, gravimetric, and calorimetric studies of
ingestion and assimilation rates of an isopod. Health Physiol 11:1485-1501
Huggett RJ, 1992. Biomarkers for chemical contamination. CRC Press, Boca Raton FL,
Huhey JE, 1983. Inorganic chemistry, 3rd ed. Harper International, Cambridge,
Hundeiker C, Pineau T, Cassar G, Betensky RA, Gleichmann E, Esser C, 1999. Thymocyte
development in Ah-receptor-deficient mice is refractory to TCDD-inducible changes. Int J
Immunopharmacol 21:841-859.
Hushka LJ, Williams JS, Greenlee WF, 1998. Characterization of 2,3,7,8tetrachlorodibenzofurandependent
suppression and AH receptor pathway gene expression in
the developing mouse mammary gland. Toxicol Appl Pharmacol 152:200-210.
Hutzinger O, Safe S, Zitko V, 1974. The Chemistry of PCBs. CRC, Cleveland, OH, USA,
Hylland K, Sandvik M, Skaare JU, Beyer J, Egaas E, Goksøyr A, 1996. Biomarkers in
flounder (Platichtys flesus): an evaluation of theeir use in pollution monitoring. Mar Environ
Res 42:223-227
IPCS, 1993. Environmental Health Criteria 140. Polychlorinated Biphenyls and Terphenyls.
WHO, Geneva,
Ito T, Matsutani T, Mori K, Nomura T, 1992. Phagocytosis and hydrogen peroxide
production by phagocytes of the sea urchin Strongylocentrotus nudus. Dev Comp Immunol
16:287-294.
Iwata H, Tanabe S, Sakai N, Nishimura A, Tatsukawa R, 1994. Geographical distribution of
persistent organochlorines in air, water and sediments from Asia and Oceania, and their
implications for global redistribution from lower latitudes. Environ Pollut 85:15-33.
Jacobson JL, Jacobson SW, 1996. Intellectual impairment in children exposed to
polychlorinated biphenyls in utero. New Engl JMed 335:783-789
James MO, 1989. Cytochrome P450 monooxygenases in crustaceans. Xenobiotica 19:1063-
1076
Jangoux M, 1982. Digestive systems: Asteroidea. In: Jangoux M, Lawrence JM (eds)
Echinoderm Nutrition. Balkema, Rotterdam, pp. 235-272
Jans D, Dubois Ph, Jangoux M, 1996. Defensive mechanisms of holothuroids
(Echinodermata): formation, role and fate of intracoelomic brown bodies in the sea cucumber
Holothuria tubulosa. Cell Tissue Res 283:99-106.
Jensen S, 1966. Report of a new chemical hazard. New Sci 32:612-615
Jensen S, Johnels AG, Olsson M, Ottlerlind G, 1969. DDT and PCB in marine animals from
swedish waters. Nature 224:247–250.
Jimenez BD, Stegeman JJ, 1990. Detoxication enzymes as indicators of environmental stress
on fish. In: Adams SM (ed), Biological indicators of stress in fish, American Fisheries
Symposium 8, American fisheries society, Bethesda, MD, 67 pp.

229
REFERENCES
Jones JM & Anderson JW, 1999. Relative potencies of PAHs and PCBs based on the
response of human cells. Environ Toxicol Pharmacol 7(1):19-26
Karbe L, Aletsee CD, Diirselen K, Heyer U, Kammann M, Kraux HJ, Rick H, Steinhout H.
Bioaccumulation and effects of plankton and benthos on the fate of contaminant. In:
Sindermann, editor. Circulation and fluxes in the North Sea. Springer-Verlag, Berlin, 1994,
pp 556-597.
Karl H, Ruoff U, Blüthgen A, 2002. Levels of dioxins in fish and fishery products on the
German market. Chemosphere 49:765-773
Kawano M, Matsushita S, Inoue T, Tanaka H, Tatsukawa R, 1986. Biological accumulation
of chlordane compounds in marine organisms from the northern North Pacific and Bering
Sea. Mar Pollut Bull 17:512-516
Keklak MM, Newman MC, Mulvey M, 1994. Enhanced uranium tolerance of an exposed
population of the eastern mosquitofish (Gambusia holbrooki Girard 1859). Arch Envrion
Contam Toxicol 27:20-24
Kelce WR, Christy RS, Laws SC, Gray LE, Kemppaven JA, Wilson EM, 1995. Persistent
DDT metabolite p,p-DDE is a potent androgen receptor antagonist. Nature 375:581–585.
Kennedy DM, 1991. A practical guide to ELISA. Pergamon, Oxford, UK,
Kennedy SW, Jones SP, 1994. Simultaneous measurement of cytochrome P450IA catalytic
activity and total protein concentration with a fluorescent plate reader. Anal Biochem 222:
217-223.
Kerkvliet NI, 1995. Immunological effects of chlorinated dibenzo-p-dioxins. Environ Health
Perspect 103(Suppl. 9), 47–53.
Kersten M, Kröncke I, 1991. Bioavailability of lead in North Sea sediments. Helgölander
Meeresuntersuschungen .45:403-409
Kersten M, Balls PW, van Enk RJ, Green N, Kramer KJM, Monteny F, Zwolsman JJG, 1994.
Background concentratrions for mmetals in the North Sea: sediments, water, mussels and
atmosphere. In: Sünderman (ed), Circulation and contaminant fluxes in the North Sea,
Springer-Verlag, Berlin, pp. 290-316
Key T, Reeves G, 1994. Organochlorines in the environment and breast cancer. The data so
far produced provide reassurance rather than anxiety. Brit Med J 308, 1520–1521
Kimura T, Taniguchi N, Tomita I, Kinea N, Yoshizaki K, Ito M, Ishikawa T, 1984.
Correlation of epizootiological observation with experimental data: chemical induction of
chromatophoromas in the croaker, Nibea mitsukurii. Natl Cancer Inst Monogr 65:139-154.
Klages NTW, 1996. Cephalopods as prey. II. Seals. Phil Trans Royal Soc London Ser
351:1045-1052
Kliewer SA, Lehmann JM, Milburn MV, Willson TM, 1999a. The PPARs and PXRs: nuclear
xenobiotic receptors that define novel hormone signaling pathways, Recent Prog Horm Res
54:345-367.
Kliewer SA, Lehmann JM, Willson TM, 1999b. Orphan nuclear receptors: shifting
endocrinology into reverse. Science 284:757–760.
Knickmeyer R, Landgraff O, Steinhart H, 1992. Cyclic organochlorines in the seastar Asterias
rubens from the German Bight, December 1988-May 1989. Mar Environ Res 33:127-143
Knutzen J, Bjerkeng B, Kristoffer N, Schlabach M, 2003. Polychlorinated

230
REFERENCES
dibenzofurans/dibenzo-p-dioxins (PCDF/PCDDs) and other dioxin-like substances in marine
organisms from the Grenland fjords,S. Norway, 1975-2001: present contamination
levels,trends and species specific accumulation of PCDF/PCDD congeners. Chemosphere
52:745-760
Kobayashi N, 1984. Marine ecotoxicological testing with echinoderms. In: Persoone G,
Jaspers E, Claus C (eds). Ecotoxicological testing for the marine environment, vol 1. State
univ Ghent and Inst Mar Scient Res Bredene pp. 341-405
Kobayashi N, 1995. Bioassay data for marine pollution using echinoderms. In: Cheremisinoff
PN (ed), Encyclopedia of environmental control technology, vol 9, Gulf Publ. Co. Houston,
pp. 539-609
Kobayashi N, Fujinaga K, 1976. Synergism of inhibiting actions of heavy metal upon the
fertilization and development of sea urchin eggs. Sci Eng Rev Doshisha Univ 17:54-69.
Kohn HW, 1983. The significance of DNA damage assays and carcinogenecity assessment.
Ann NY Acad Sci 407:106-118
Koponen K, Kukkonen JVK, Lindstrom-Seppa P, 1998. Chemical accumulation and
biotransformation enzyme acitvities of rainbow trout embryos in waterborne exposure to
PCB77. Mar Environ Res 46(1-5):475-478
Koyama J, Nanamori N, Segawa S, 2000. Bioaccumulation of waterborne and dietary
cadmium by oval squid Sepioteuthis lessoniana, and its distribution among organs. Mar Pollut
Bull 40(11):961-967
Krogenaes AK, Nafstad I, Skare JU, Farstad W, Hafne AL, 1998. In vitro reproductive
toxicity of polychlorinated biphenyl congener 153 and 126. Reproductive Toxicol
12(6):575–580,
Laane RWPM, Sonneveldt HLA, Vander Weyden AJ, Loch JPG, Groenveld G, 1999. Trend
in the spatial and temporal distribution of metals (Cd, Cu, Zn and Pb) and organic compounds
(PCBs and PAHs) in the dutch coastal zone sediments from 1981 to 1996: a model case study
for Cd and PCBs. J Sea Res 41:1-17
Lahvis GP, Lindell SL, Thomas RS, McCuskey RS, Murphy C, Glover E, Bentz M, Southard
J, Bradfield CA, 2000. Portosystemic shunting and persistent fetal vascular structures in aryl
hydrocarbon receptor-deficient mice. Proc Natl Acad Sci USA 97:10442–10447.
Lambert C, Nicolas JL, 1998. Specific inhibition of chemiluminescent activity by pathogenic
vibrios in hemocytes of two marine bivalves: Pecten maximus and Crassostrea gigas. J Invert
Pathol 71:53-63.
Langston WJ, 1990. Toxic effects of metals and the incidence of metal pollution in marine
ecosystems. In: Furness RW, Rainbow PS (eds), Heavy metals and the marine environment,
CRC Press, Boca Raton, FL. Pp. 101-122
Lara R, Ernst W, 1989. Interaction between polychlorinated biphenyls and marine humic
substances: Determination of association coefficients. Chemosphere 19:1655–1664.
Larson KG, Roberson BS, Hetrick FM, 1989. Effect of environmental pollutants on the
chemiluminescence of hemocytes from the American oyster Crassostrea virginica. Dis Aquat
Org 6:131-136.
Lee RF, 1981. Mixed function oxidase (MFO) in marine invertebrates. Mar Biol Lett 2:87-
105
Lee RF, Singer SC, Page DS, 1981. Responses of cytochrome P-450 systems in marine crab

and polychaetes to organic pollutants. Aquat Toxicol 1:355–365.
231
REFERENCES
Leece B, Denomme MA, Towner R, Li SMA, Safe S, 1985. Polychlorinated biphenyls:
correlation between in vivo and in vitro quantitative structure-activity relationships (QSARs).
J Toxicol Environ Health 16:379-388.
Letcher RJ, Klasson-Wehler E, Bergman A (2000) Methyl sulfone and hydroxylated
metabolites of polychlorinated biphenyls. In: Paasivirta J (ed) The handbook of
environmental chemistry Vol 3, Part K: New types of persistent halogenated compounds.
Springer-Verlag, Berlin, Heidelberg, pp. 315-359
Lewis DFV (ed). 1996. Cytochromes P450. Structure, function, mechanism. TJ press, UK.
348 pp.
Lewis JR, 1978. The implications of community structure of benthic monitoring studies. Mar
Pollut Bull 9: 64-67
Lin TM, Ko K, Moore RW, Buchanan DL, Cooke PS, Peterson RE, 2001, Role of the aryl
hydrocarbon receptor in the development of control and 2,3,7,8-tetrachlorodibenzo-p-dioxinexposed
male mice, J. Toxicol. Environ. Health 64(A):327–342.
Livingstone DR, 1991. Organic xenobiotic metabolism in marine invertebrates, in: R. Gilles
(Ed.), Advances in Comparative and Environmental Physiology, vol. 7, Springer-Verlag,
Berlin, Heidelberg pp. 45–185.
Livingstone DR, 1998. The fate of organic xenobiotics in aquatic ecosystems: quantitative
and qualitative differences in biotransformation by invertebrates and fish. Comp Biochem
Physiol 120(A):43-49
Livingstone DR, Kirchin MA, Wiseman A, 1989. Cytochrome P450 and oxidative
metabolism in molluscs. Xenobiotica 19: 1041-1062
Livingstone DR, Donkin P, Walker CH, 1992. Pollutants in marine ecosystems: an overview.
In: Walker CH, Livingstone DR, editors. Persistent Pollutants in Marine Ecosystems. Oxford:
Pergamon, SETAC special publication, 1992:235–263
Livingstone DR, Nasci C, Solé M, Da Ros L, O’Hara SCM, Peters LD, Fossato V, Wootton
AN, Goldfarb PS. 1997. Apparent induction of a cytochrome P450 with immunochemical
similarities to CYP1A in digestive gland of the common mussel (Mytilus galloprovincialis L.)
with exposure to 2,2’,3,4,4’,5’-hexachlorobiphenyl and Arochlor 1254. Aquat Toxicol
38:205-224
Livingstone DR, Goldfarb PS, 1998. Aquatic environmental biomonitoring: use of
cytochrome P450 and other molecular biomarkers in fish and mussels. In: Lynch J, Wiseman
A (eds): biotechnology research series, environmental biomonitoring. The biotechnology
ecotoxicology interface, vol 6. Cambridge University Press, Cambridge, pp. 101-129
Livingstone DR, Chipman JK, Lowe DM, Minier C, Mitchelmore CL, Moore MN, Peters LD,
Pipe RK, 2000. Development of biomarkers to detect the effects of organic pollution on
aquatic invertebrates: recent molecular, genotoxic, cellular and immunological studies on the
common mussel (Mytilus edulis L.) and other mytilids. Int J Environ Pollut 13(1-6):56-91
Lizen V, 1990. Caractérisation Géobiochimique de la matière particulaire en baie sud de Mer
du Nord. Master thesis, University of Liège.
Lohse J, 1991. Distribution of organochlorine pollutants in North Sea sediments. Wat Sci
Technol 21:273-278

232
REFERENCES
Long ER, Macdonald D, Smith SL, Calder FD, 1995. Incidence of adverse biological effects
within ranges of chemical concentrations in marine and estuarine sediments. Environ
Managm 19:81-97.
Lourens JM, 1996. Speciemanagement; effecten van het reguleren van het storten van zoute
baggerspecie in zee. Report RIKZ-96.017, Rijkswaterstaat, The Hague, 106 pp.
Luoma SN, Carter JL, 1991. Effects of trace metals on aquatic benthos. In: Newman MC,
McIntosh AW (eds), Metal ecotoxicology: concepts and applications, Lewis Publ, Chelsea,
MI. Pp. 261-300
Luster MI, Munson AE, Thomas PT, Holsapple MP, Fenters J, White KL, Laver LD, Dean
JH, 1988. Development oof a testing battery to assess chemical-induced immunotoxicity.
Fund Appl Toxicol 10:2-19
Malins DC, Krahn MM, Brown DW, Rhodes LD, Myers MS, McCain BB, Chan SL, 1985.
Toxic chemicals in marine sediment and biota from Mukilteo, Washington: relationships with
hepatic neoplasms and other hepatic lesions in English sole (Parophrys vetulus). J Natl
Cancer Inst 74:487-494.
Mallefet J, Ajuzie CC, Baguet F, 1994. Aspects of calcium dependence of light emission in
the ophiuroid Amphipholis squamata (Echinodermata). In: David B, Guille A, Féral JP, Roux
(eds), Echinoderms through time, Balkema, Rotterdam, The Netherlands, pp. 455-460
Mangold K, Bidder AM, Boletzky SV, 1989. Appareils excréteurs et excrétion. In: Grassé PP
(ed) Traité de Zoologie, Tome V. Mangold K (ed) Céphalopodes. Masson, Paris, pp. 437-457
Manly BFJ, 1986. Multivariate statistical methods: a primer. Chapman and Hall Ltd, London.
Mann T, Martin AW, French M, 1988. Entry of pollutants from seawater into the
spermatophore and spermatozoa of the giant octopus of the North Pacific. Mar Pollut Bull
19(12):669-671
Mantovani A, Stazi AV, Macri C, Maranghi F, Ricciardi C, 1999. Problems in testing and risk
assessment of endocrine disrupting chemicals with regard to developmental toxicology.
Chemosphere, 39:1293-1300
Marchand M, Abarnou A, Marcaillou-Lebaut C, 1990. Les polychlorobyphényles (PCB) en
milieu marin: biogéochimie et écotoxicologie. Rapp Scient Techn IFREMER 18, 162 pp.
Martin JH, Flegal AR, 1975. High copper concentrations in squid livers in association with
elevated levels of silver, cadmium, and zinc. Mar Biol 30:51-55
McCarthy JF, Shugart LR (eds), 1990. Biomarkers of environmental contamination, Lewis
publishers, Boca Raton Florida,
McLeese DW, Metcalfe CD, Pezzack DS, 1980. Uptake of PCBs from sedieemnts by Nereis
virens and Crangon septemspinosa. Arch Environ Contam Toxicol 9:507-518
McManus JP, Prandle D, 1997. Development of a model to reproduce observed suspended
sediment distribution in the southern North Sea using principal component analysis and
multiple linear regression. Cont Shelf Res 17:761-778.
Meador JP, Casillas E, Sloan CA, Varanasi U, 1995, Comparative bioaccumulation of
polycyclic aromatic hydrocarbons from sediment by two infaunal invertebrates, Mar Ecol
Prog Ser, 123:107-124
Menge BA, 1982. Effects of feeding on the environment: Asteroidea. In: Jangoux M,
Lawrence JM. Echinoderm nutrition. Balkema, Rotterdam: pp. 521-551

233
REFERENCES
Menge BA, Berlow EL, Blanchette CA, Navarrete SA, Yamada SB, 1994. The keystone
species concept: variation in interaction strength in a rocky intertidal habitat. Ecol
Monographs 64:249-286
Menge BA, Berlow EL, Blanchette CA, Navarrete SA, Yamada SB, 1994. The keystone
species concept: variation in interaction strength in a rocky intertidal habitat. Ecol
Monographs 64:249-286
Metcalfe CD, 1994. Polychlorinated biphenyls, In: Kiceniuk JW & Ray S (eds), Analysis of
contaminants in edible aquatic resources, VCH pp. 305-338
Michel XR, Suteau P, Robertson LW, Narbonne JF. 1993. Effects of benzo(a)pyrene,
3,3’,4,4’-tetrachlorobiphenyl and 2,2’,4,4’,5,5’-hexachlorobiphenyl on the xenobioticmetabolizing
enzymes in the mussel (Mytilus galloprovincialis). Aquat Toxicol 27:335-344
Miller K, 1985. Review: immunotoxicology. Clin Exptl Immunol 61:219-223
Miramand P, Guary JC, 1980. High concentrations of some heavy metals in tissues of the
Mediterranean octopus. Bull Environ Contam Toxicol 24:783-788
Miramand P, Bentley D, 1992. Concentration and distribution of heavy metals in tissues of
two cephalopods, Eledone cirrhosa and Sepia officinalis, from the French coast of the English
Channel. Mar Biol 114:407-414
Mix MC, 1986. Cancerous diseases in aquatic animals and their association with
environmental pollutants: a critical literature review. Mar Environ Res 20:1-141.
Miyata H, Takayama K, Ogaki J, Kashimoto T, Fukushima S,1987. Monitoring of PCDDs in
Osaka Bay using blue mussels. Chemosphere 16:1817-1822
Miyata H, Aozasa O, Yamamori K, Fujiwara A, Yasuda Y,1994. Survey on pollution by
dioxin and related compounds monitored by blue mussels as a biological indicator at 24
coastal areas in Japan. Organohalogen Compd 20:187-190
Mohn WW, Tiejde JL, 1992. Microbial reductive dehalogenation. Microbiol Rev 56:482–507.
Mommaerts JP, Vanden Clooster A, Roose P, 1994. Contaminants in seawater and sediments
of the North Sea (Belgian continental shelf and western Scheldt estuary). Report of
September 1994,
Moon HB, Choi HG, Kim SS, Lee PY, Ok G, 2000. Monitoring of polychlorinated dibenzop-dioxins
and polychlorinated dibenzofurans in the Korean coast using mussel and oyster.
Organohalogen Compd 46:526-529
Moore MN, 1992. Molecular cell pathology of pollutant-induced liver injury in flatfish: use of
fluorescent probes. Mar Ecol Prog Ser (Special) 91:127-133.
Moore MN, Livingstone DR, Donkin P, Bayne BL, Widdows J, Lowe DM, 1980. Mixed
function oxygenases and xenobiotic detoxication/toxication systems in bivalve molluscs.
Helgol Wiss Meereunters 33:278
Moore MR, Vetter W, Gaus C, Shaw GR, Müller JF, 2002. Trace organic ccompound sin the
marine environment. Mar Pollut Bull 45:62-68
Muir CG, Norstrom RJ, Simon M, 1988. Organochlorine contaminants in Arctic marine food
chains: accumulation of specific polychlorinated biphenyls and chlordane-related compounds.
Environ Sci Technol 22:1071-1079.
Munkittrick KR, van den Heuvel MR, Metner MR, Lockhart WL, Stegeman JJ, 1993.
Interlaboratory comparison and optimisation of hepatic microsomal ethoxyresorufin O-

234
REFERENCES
deethylase in white sucker (Catostomas commersoni) exposed to bleached kraft pulp mill
effluent. Environ Toxicol Chem 12: 1273-1283.
Murdoch MH, Chapman PM, Norman DM, Quintino VM, 1997. Spiking sediment with
organochlorines for toxicity testing. Environ Toxicol Chem 16:1504-1509
Myers MS, Stehr CM, Olsen OP, Johnson LL, McCain BB, Chan SL, Varanasi U, 1994.
Relationships between toxicopathic hepatic lesions and exposure to chemical contaminants in
English sole (Pleuronectus vetulus), starry flounder (Platichthys stellatus), and white croaker
(Genyonemus lineatus) from selected marine sites on the Pacific coast, USA. Environ Health
Perspect 102:200-215.
Nacci D, Jackim E, Walsh R, 1986. Comparative evaluation of three rapid marine toxicity
tests: sea urchin early embryo growth test, sea urchin sperm cell toxicity test and microtox.
Environ Toxicol Chem 5:521-525
Nagayama J, Kurastune M, Masuda Y, 1976. Determination of dibenzofurans in Kanechlor
and Yucho oil. Bull Environ Contam Toxicol 15:9-13
Nakahara M, Koyanagi T, Ueda T, Shimizu C, 1979. Peculiar accumulation of cobalt-60 by
the branchial hearts of Octopus. Bull Jap Soc Scient Fish 45:539-545
Napolitano GE, Richmond JE, 1995. Enrichment of biogenic lipids, hydrocarbons and PCBs
in stream-surface foams. Environ Toxicol Chem 14:197-201.
Navarro JC, Villanueva R, 2000. Lipid and fatty acid composition of early stages of
cephalopods: an approach to their lipid requirements. Aquaculture 183:161-177
Nebert DW, Goujon FM, Gielen JE, 1972. Aryl hydrocarbon hydroxylase induction by
polycyclic hydrocarbons: simple autosomal trait in the mouse, Nature New Biol 236:107-
110.
Nebert DW, Gonzalez FJ, 1987. P450 genes: structure, evolution, and regulation. Ann Rev
Biochem 56:945-993.
Nebert DW, Matthew ALR, Dieter Z, Solis WA, Yang Y, Dalton TP, 2000. Role of the
Aromatic Hydrocarbon Receptor and [Ah] Gene Battery in the Oxidative Stress Response,
Cell Cycle Control, and Apoptosis. Biochem Pharmacol 59(1):65-85
Nelson DR, Koymans L, Kamataki T, Stegeman JJ, Feyreissen R, Waxman DJ, Waterman
MR, Gotoh O, Coon MJ, Estaabrook RW, Gunsalus IC, Nebert DW, 1996. P450
superfamily: update on new sequences, gene mapping, accession numbers and nomenclature.
Pharmacogenetics 6:1-42
Nieboer E, Richardson DHS, 1980. The replacement of the nondescript term ‘heavy metal’ by
a biologically and chemically significant classification of metal ions. Environ Pollut Ser B
Chem Phys 1:3-26
Noble PB, 1970. Coelomocyte aggregation in Cucumaria frondosa: effect of
ethylenediamine tetracetate, adenosine and adenosine nucleotides. Biol Bull 139:549-556
Noël D, Bachere E, Mialhe E, 1993. Phagocytosis associated chemiluminescence of
hemocytes in Mytilus edulis (Bivalvia). Dev Comp Immunol 17:483-493.
Nolan C, Dahlgaard H, 1991. Accumulation of metal radiotracers by Mytilus edulis. Mar Ecol
Prog Ser 70:165-174
Norstrom RJ, Muir DCG, 1994. Chlorinated hydrocarbon contaminants in arctic marine
mammals. Sci Total Environ 154:107-128.

235
REFERENCES
North Sea Task Force, 1993a. North Sea quality status report 1993. Oslo and Paris
Commission. International Council for the Exploration of the Sea, London, 346 pp.
North Sea Task Force, 1993b. North Sea subregion 4 assessment report. Oslo and Paris
Commission. International Council for the Exploration of the Sea. London, 132 pp.
Oh JR (ed.), 2000. POPs in the Marine Environment: Analysis, Bioaccumulation and
Toxicity. Korea Research and Development Institute,
Olsen CR, Cutshall NH, Larsen IL, 1982. Pollutant–particle association and dynamics in
coastal marine environments: a review. Mar Chem 11:501-533.
OSPAR Commission for the Protection of the Marine Environment of the North-East
Atlantic, 2000. Quality Status Report for the North-East Atlantic, OSPAR Commission,
London, 108 pp.
Otto LJ, Zimmerman TF, Furnes GK, Mork M, Saetre R, Becker G, 1990. Review of the
physical oceanography of the North Sea. Neth J Sea Res 26:161-238.
Pagano G, Esposito A, Giordano GG, 1982. Fertilization and larval development in sea
urchins following exposure of gametes and embryos to cadmium. Arch Environ Contam
Toxicol 11:47-55
Pagano G, Cipollaro M, Corsale G, Esposito A, Ragucci E, Giordano GG, Trieff NM, 1985.
Comparative toxicities of chlorinated biphenyls on sea urchin egg fertilisation and
embryogenesis. Mar Environ Res 17:240-244.
Pagano G, Cipollaro M, Corsale G, Esposito E, Ragucci E, Giordano GG, Trieff NM, 1986.
The sea urchin: bioessay for the assessment of damage from environmental contaminants. In:
Crains JJr (ed) Community Toxicity Testing. ASTM Philadelphia, PA, USA, pp. 66-92.
Pagano G, Corsale G, Esposito A, Dinnel PA, Romana LA, 1989. Use of sea urchin sperm
and embryo bioassay in testing the sublethal toxicity of realistic pollutant levels. Adv Appl
Biotechnol Ser (5):153-163
Pagano G, His E, Beiras R, De Biase A, Korkina LG, Iaccarino M, Oral R, Quiniou F,
Warnau M, Trieff NM, 1996. Cytogenetic, developmental, and biochemical effects of
aluminium, iron and their mixtures in sea urchins and mussels. Arch Environ Contam Toxicol
31:466-474
Pagano G, Bonassi S, De Biase A, Degan P, Deeva IB, Doronin YK, Iaccarino M, Oral R,
Warnau M, Korkina LG, 1997. L-methionine induces stage-dependent changes of
differentiation and oxidative activity in sea urchin embryogenesis. Pharmacol Toxicol
81:134-143.
Patandin S, Lanting CI, Mulder PGH, Boersma ER, Sauer PJJ, Weisglas-Kuperus N, 1999.
Effects of environmental exposure to polychlorinated biphenyls and dioxins on cognitive
abilities in Dutch children at 42m onths of age. J Pediatrics 134:33–41
Patnode KA, Curtis LR, 1994. 2,2’,4,4 ,5,5-Hexachlorobiphenyl alteration of uterine
progesterone and estrogen receptors coincides with embryotoxicity in mink (must & vison).
Toxicol Appl Pharmacol 127:9-18.
Payne JF, 1976. Field evaluation of benzopyrene hydroxylase induction as a monitor for
marine pollution. Science 191:945-946
Payne JF, Penrose WR, 1975. Induction of aryl hydrocarbon (beenzo(a)pyrene) hydroxylase
in fish by petroleum. Bull Environ Contam Toxicol 14:112-116

236
REFERENCES
Payne JF, May N, 1979. Further studies on the effect of petroleum hydrocarbons on mixedfunction
oxidases in marine organisms. In Kahn MAQ, Lech JJ, Menn JJ (eds): Pesticide and
xenobiotic metabolism in aquatic organisms. American Chemical Society, Washington, D. C.,
USA, pp. 339-347
Payne JF, Fancey LL, Rahimtula AD, Porter EL, 1987. Review and perspective on the use of
mixed-function oxygenase enzymes in biological monitoring. Comp Biochem Physiol
86C:233-245
Peakall D (ed), 1994. Animal biomarkers as pollution indicators Ecotoxicology series 1,
Chapman & Hall, London, 291 pp.
Peters JM, Narotsky MG, Elizondo G, Fernandez-Salguero PM, Gonzalez FJ, Abbott BD,
1999. Amelioration of TCDD-induced teratogenesis in aryl hydrocarbon receptor (AhR)-null
mice, Toxicol Sci 47:86–92.
Peters LD, Nasci C, Livingstone DR. 1998. Variation in level of cytochrome P4501A, 2B, 2E,
3A and 4A immunopositive proteins in digestive glands of indigenous and transplanted
mussel Mytilus galloprovincialis in Venice lagoon, Italy. Mar Environ Res 46(1-5):295-299
Philips DJH, 1976, The common mussel Mytilus edulis as an indicator of pollution by zinc,
cadmium, lead and copper. I. Effects of environmental variables on uptake of metals. Mar
Biol 38:59-69
Philips DJH, 1990. Use of macroalgae and invertebrates as monitors of metal levels in
estuaries and coastal waters. In: Furness RW, Rainbow PS (eds), Heavy Metals in the Marine
Environment, CRC Press, Boca Raton, pp. 81-99
Pipe RK, Coles JA, 1995. Environmental contaminants influencing immune function in
marine bivalve molluscs. Fish and Shellfish Immunol 5:581-595
Pipe RK, Coles JA, Thomas ME, Fossato VU, Pulsford AL, 1995. Evidence for
environmentally derived immunomodulation in mussels from the Venice Lagoon. Aquat
Toxicol 32:59-73.
Pipe RK, Coles JA, Carissan FMM, Ramanathan K, 1999. Copper induced
immunomodulation in the marine mussel, Mytilus edulis. Aquat Toxicol 46:43-54.
Pohjanvirta R, Tuomisto J, 1994. Short-term toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin
in laboratory animals: effects, mechanisms, and animal models. Pharmacol Rev 46:483–549.
Poland A, Glover E, 1973a. 2,3,7,8-Tetrachlorodibenzo-p-dioxin: potent inducer of
aminolevulinic acid synthetase. Science 179:476-477.
Poland A, Glover E, 1973b. Chlorinated dibenzo-p-dioxins: potent inducers of deltaaminolevulinic
acid synthetase and aryl hydrocarbon hydroxylase. II. A study of the structureactivity
relationship, Mol Pharmacol 9:736-747.
Poland A, Glover E, 1974. Comparison of 2,3,7,8-tetrachlorodibenzo-p-dioxin, a potent
inducer of aryl hydrocarbon hydroxylase, with 3-methyl-cholanthrene. Mol Pharmacol
10:349-359.
Poland A, Glover E, Robinson JR, Nebert DW, 1974. Genetic expression of aryl hydrocarbon
hydroxylase activity. Induction of monooxygenase activities and cytochrome P1-450
formation by 2,3,7,8-tetrachlorodibenzo-p-dioxin in mice genetically ‘nonresponsive’ to other
aromatic hydrocarbons. J Biol Chem 248:5599-5606.

237
REFERENCES
Poland A, Glover E, 1975. Genetic expression of aryl hydrocarbon hydroxylase by 2,3,7,8tetrachlorodibenzo-p-dioxin:
evidence for a receptor mutation in genetically non-responsive
mice. Mol Pharmacol 11:389-398.
Poland A, Glover E, Kende AS, 1976. Stereospecific, high-affinity binding of 2,3,7,8tetrachlorodibenzo-p-dioxin
by hepatic cytosol. J Biol Chem 251:4936–4946.
Poland A, Glover E, 1977. Chlorinated biphenyl induction of aryl hydrocarbon hydroxylase
activity: a study of the structure–activity relationship. Mol Pharmacol 13:924–938.
Poland A, Knutson JC, 1982. 2,3,7,8-Tetrachlorodibenzo-p-dioxin and related halogenated
aromatic hydrocarbons: examination of the mechanism of toxicity. Annu Rev Pharmacol
Toxicol 22:517–554.
Poland A, Glover E, 1987. Variation in the molecular mass of the Ah receptor among
vertebrate species and strains of rats. Biochem Biophys Res Commun 146:1439-1451
Poulicek M, 1982. Coquilles et autres structure squelettiques des mollusques. Composition
chimique, biomasse et biodégradation en milieu marin. PhD Thesis, Liège University,
Powell-Coffman JA, Bradfield CA, Wood WD, 1998. Caenorhabditis elegans orthologs of
the aryl hydrocarbon receptor and its heterodimerization partner the aryl hydrocarbon receptor
nuclear translocator, Proc Natl Acad Sci USA 95:2844-2849.
Pruell RJ, Taplin BK, McGovern DG, McKinney R, Norton SB, 2000. Organic contaminant
distribution in sediments, polychaetes (Nereis virens) and American lobster (Homarus
americanus) from a laboratory food chain experiment. Mar Environ Res 49:19-36
Quensen JF, Boyd SA, Tiedje JM, 1990. Dechlorination of four commercial polychlorinated
biphenyl mixtures (Aroclors) by anaerobic microorganisms from sediments. Appl Environ
Microbiol 56:2360–2369.
Quiniou F, Guillou M, Judas A, 1999. Arrest and delay in embryonic development in sea
urchin populations of the bay of Brest (Brittany, France): link with environmental factors.
Mar Pollut Bull 38(5):401-406
Radenac G, Fichet D, Miramand P, 2001. Bioaccumulation and toxicity of four dissolved
metals in Paracentrotus lividus sea-urchin embryo. Mar Environ Res 51:151-166
Rapaport RA, Eisenreich SJ, 1988. Historical atmospheric inputs of high molecular weight
chlorinated hydrocarbons to eastern North America. Environ Sci Technol, 22:931 –941.
Raulf M, König W, 1991. Induction of mediator generation in human inflammatory cells by
polychlorinated biphenyls. In New Trends in Allergy III (J. Ring and B. Przybilla, eds), pp.
489-499. Springer, Berlin, Germany.
Regoli F, Principato G, 1995. Glutathione, glutathione-dependent and antioxidant enzymes in
mussel, Mytilus galloprovincialis, exposed to metals under field and laboratory conditions:
implications for the use of biochemical biomarkers. Aquat Toxicol, 31:143–164.
Regoli F, Gorbi S, Frenzilli G, Nigro M, Corsi I, Focardi S, Winston GW. 2002. Oxidative
stress in ecotoxicology: from the analysis of individual antioxidants to a more integrated
approach. Mar Environ Res 54:419-423
Reichle DE, 1967. radioisotope turnover and energy flow in terrestrial isopod populations.
Ecology 48:351-366
Reichle DE, Dunaway PB, Nelson DJ, 1970. Turnover and concentration of radionuclides in
food chains. Nucl Saf 11:43-55

238
REFERENCES
Reinfelder JR, Fisher NS, 1991. The assimilation of elements ingested by marine copepods.
Science 251:794-796
Rijnsdorp AD, Vethaak AD, Van Leeuwen PI, 1992. Population biology of dab Limanda
limanda in the southeastern North Sea. Mar Ecol Prog Ser 91:19-35.
Riley JP, Segar DA. The distribution of the major and some of the minor elements in marine
animals. I: Echinoderms and Coelenterates. J Mar Biol Ass UK 1970; 50:721-730
Robles R, Morita Y, Mann KK, Perez GI, Yang S, Matikainen T, Sherr DH, Tilly JH, 2000.
The aryl hydrocarbon receptor, a basic helix-loop-helix transcription factor of the PAS gene
family, is required for normal ovarian germ cell dynamics in the mouse. Endocrinology
141:450-453.
Rodhouse PG, Nigmatullin CM, 1996. Role as consumers. Phil Trans Roy Soc London B
351:1003-1022
Roesijadi G, Brubacher LL, Unger ME, Anderson RS, 1997. Metallothionein mRNA
induction and generation of reactive oxygen species in molluscan hemocytes exposed to
cadmium in vitro. Comp Biochem Physiol 118(C), 171 /176
Rouleau C, Pelletier E, Tjälve H, 1993. The uptake and distribution of 203 HgCl 2 and CH 3HgCl 2
in the sea star Asterias rubens after 24-h exposure studied by impulse counting and whole
body autoradiography. Aquat Toxicol 26:103-116.
Rygg B, 1985. Effect of sediment copper on benthic fauna. Mar Ecol Prog Ser 25:83-89
Safe S, 1990. Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs), dibenzofurans
(PCDFs), and related compounds: environmental and mechanistic considerations which
support the development of toxic equivalency factors (TEFs). CRC Crit Rev Toxicol: 21:51-
88
Safe S, 1994. Polychlorinated biphenyls (PCBs): Environmental impact, biochemical and
toxic responses, and implications for risk assessment. Crit Rev Toxicol 24:87-149.
Safe S, 1995. Modulation of gene expression and endocrine response pathways by 2378tetrachlorodibenzo-p-dioxin
and related compounds. Pharmac Ther 67(2):247-281
Safe S, Bandiera S, Sawyer T, Robertson L, Safe L, Parkinson A, Thomas PE, Ryan DE, Reik
LM, Levin W, Denomme MA, Fujita T, 1985. PCBs: Structure-function relationships and
mechanism of action. Environ Health Perspect 60:47-56.
Sager DB, Girard DM Long-term effects on reproductive parameters in female rats after
translactational exposure to PCBs. Environ Res. 1994;66:52-76.
Saier B, 2001. Direct and indirect effects of seastars Asterias rubens on mussel beds (Mytilus
edulis) in the Wadden Sea. J Sea Res 46:29-42.
Sanders G, Jones KC, Hamilton-Taylor J, Dorr H, 1992. Historical inputs of polychlorinated
biphenyls and other organochlorines to a dated lacustrine sediment core in rural England.
Environ Sci Technol 26:1815 –1821
Sarkar A, Everaarts JM, 1995. DNA strandbreaks as a biological response to chlorinated
biphenyls congeners CB28, CB 52 and CB77, and benzo(a)pyrene: in vitro studies with sea
star DNA (Asterias rubens). In: Everaarts JM (ed), Contaminants in the Marine Environment:
their Fate in the Abiotic and Biotic Compartments with Emphasis on Biological Responses
(biomarkers) of Organisms. First annual progress report of the EC Indo-Dutch research
project. Netherlands Institute of Sea Research/NIOZ, Den Berg, Texel, the Netherlands, pp.
34-37.

239
REFERENCES
Satar AA, 2000. The immune system as a potential target for environmental estrogens
(endocrine disrupters): a new emerging field. Toxicology 150:191-206
Savas U, Griffin KJ, Johnson EF, 1999. Molecular mechanisms of cytochrome P-450
induction by xenobiotics: an expanded role for nuclear hormone receptors. Mol Pharmacol
56:851–857.
Sawyer T, Safe S, 1982. PCB isomers and congeners: induction of aryl hydrocarbon
hydroxylase and ethoxyresorufin-o-deethylase enzyme activities in rat hepatoma cells.
Toxicol Lett 13:87-94.
Schlebusch H, Wagner U, van der Ven H, Al-Hasani S, Krebs D, 1989. Polychlorinated
biphenyls: the main occurrence of the main congeners in follicular and sperm fluids. J Clin
Chem Clin Biochem 27:663-667.
Schlenk D, 1996. The role of biomarkers in risk assessment. Human Ecol Risk Assess,
2:251-256
Schlenk D, Perkins EJ, Layher WG, Zhang YS, 1996. Correlating metrics of fish heaklth with
cellular indicators of stress in Arkansas bayou. Mar Environ Res 42:247-251
Schmidt JV, Su GH-T, Reddy JK, Simon MC, Bradfield CA, 1996. Characterization of a
murine Ahr null allele: involvement of the Ah receptor in hepatic growth and development.
Proc Natl Acad Sci USA 93, pp. 6731-6736
Schramm M, Müller E, Triebskorn R, 1998. Brown trout (Salmo trutta f. Fario) liver
ultrastructure as a biomarker for assessment of small stream pollution. Biomarkers 3:93-108.
Schreitmüller J, Vigneron M, Bacher R, Ballschmiter K, 1994. Pattern analysis of
polychlorinated biphenyls (PCBs) in marine air of the Atlantic ocean. Intern J Environ Anal
Chem 57:33-52
Schultz-Bull DE, Petrick G, Kannan N, Duinker JC, 1995. Distribution of individual
chlorobiphenyls (PCB) in solution and suspension in the Baltic Sea. Mar Chem 48:245-270
Schulz DE, Petrick G, Duinker JC, 1989. Complete characterization of polychlorinated
biphenyl congeners in commercial Aroclor and Clophen mixtures by multidimensional
electron capture detection. Environ Sci Technol 23:852–859.
Schweitzer LE, Hose JE, Suffet IH, Bay SM, 1997. Differential toxicity of three PCB
congeners in developing sea urchin embryos. Environ Toxicol Chem 16(7):1510-1514
Schweitzer LE, Bay SM, Suffet IH, 2000. Dietary assimilation of a polychlorinated biphenyl
in adult sea urchins (Lytechinus pictus) and maternal transfer to their offspring, Environ
Toxicol Chem 19(7):1919-1924
Schwetz BA, Norris JM, Sparschu GL, Rowe VK, Gehring P, Emerson JL, Gerbig CG, 1973.
Toxicology of chlorinated dibenzo-p-dioxins. Environ Health Perspect 5:87-89
Seiler P, Fischer B, Lindenau A, Beier HM, 1994. Effects of persistent chlorinated
hydrocarbons on fertility and embryonic development in the rabbit. Hum Reprod. 9:1920-
1926
Shaw GR, Connell DW, 1987. Comparative kinetics for bioaccumulation of polychlorinated
biphenyls by the polychaete (Capitella capitella) and Fish (Mugil cephalus), Ecotoxicol
Environ Saf 13:84-91
Shen GT, Boyle EA, 1988. Determination of lead, cadmium, and other trace metals in
annually-banded corals. Chem Geol 67:47-62

240
REFERENCES
Shina A, Gross J, Lifshitz A, Sklarz B, 1978. Carotenoids of the invertebrates of the Red Sea
(Eilat shore) – II. Carotenoid pigments in the gonads of the sea urchin Tripneustes gratila
(Echinodermata). Comp Biochem Physiol 61(B):123-128.
Shugart LR, Bickham J, Jackim G, McMahon G, Ridley W, Stein J, Steinert S, 1992. DNA
alterations. In: Huggett RJ, Kimerle RA, Mehrle PM, Bergman HL (eds): Biomarkers:
biochemical, physiological and histological markers of anthropogenic stress. Chelsea MI:
Lewis publishers, pp. 125-163
Shugart LR, Peakall DB, Walkes CH, Everaarts JM, 1993. Overview. In: DB Peakall and LR
Shugart (Eds): Biomarkers: Research and application in the assessment of environmental
health. NATO ASI Series. Series H: Cell Biology 68:XII-XIX
Singer SC, March PE, Goncoulin F, Lee RF, 1980. Mixed function oxygenase activity in the
blue crab Callinectes sapidus: characterization of enzyme activity from stomach tissue.
Comp Biochem Physiol 65C:129-134
Sipes IG, Schnellmann RG, 1987. Biotransformation of PCBs: metabolic pathways and
mechanisms. In: Safe S (ed), Polychlorinated biphenyls: mammalian and environmental
toxicology, Springer Verlag, Berlin, pp. 97-110
Skei JM, 1994. Tiltaksorienterte miljøundersøkelser i Sørfjorden og Hardangerfjorden 1992.
NIVA-report (in Norwegian) 2, N° 3038,
Skerfving E, Svensson BG, Asplund L, Hagmar L, 1994. Exposure to mixtures and congeners
of polychlorinated biphenyls. Clin Chem 40:1409-1415
Sleiderink HM, Oostigh I, Goksøyr A, Boon JP, 1995. Sensitivity of cytochrome P450 1A
induction in dab (Limanda limanda) of different age and sex as a biomarker for
environmental contaminants in the southern north sea. Arch Environ Contam Toxicol 28:
423-430.
Smale MJ, 1996. Cephalopods as prey. IV. Fishes. Philosophical Transactions of the Royal
Society of London B 351, 1067-1081
Smith AG, 2000. How toxic is DDT. Lancet 356:267–268
Smith AG, Dinsdale D, Cabral JRP, Wright AL, 1987. Goitre and wasting induced in
hamsters by hexachlorobenzene. Arch Toxicol 60:343-349.
Smith AG, Gangoli SD, 2002. Organochlorine chemicals in seafood: occurrence and health
concerns. Food Chem Toxicol 40:767–779
Smith CR, Barker CM, Barker LF, Jessen-Eller K, Reinish CL, 1999. Polychlorinated
biphenyls (PCBs) selectively disrupt serotogenic cell growth in the developing Spisula
embryo. Toxicol Sci 50:54-63
Smith JE, 1968. Toxicity experiments. In: Torrey Canyon pollution and marine life. Smith JE
(ed), Cambridge Univ. Press, London, pp. 115-149
Snyder MJ, 1998. CYP4 Cytochrome P450 enzymes belonging to the CYP4 family from
marine invertebrates. Biochem Biophys Res Commun 249:187-190.
Snyder MJ, 2000. Cytochrome P450 enzymes in aquatic invertebrates: recent advances and
future directions. Aquat Toxicol 48:529-547.
Snyder MJ, Mulder EP, 2001. Environmental endocrine disruption in decapod crustacean
larvae: hormone titers, cytochrome P450, and stress protein responses to heptachlor exposure.
Aquat Toxicol 55:177-190

241
REFERENCES
Sørensen M, Bjerregaard P, 1991. Interactive accumulation of mercury and selenium in the
sea star Asterias rubens. Mar Biol 108:269-270.
Spies RB, Rice DWJr, Felton J, 1988. Effects of organic contaminants on reproduction of the
starry flounder Phatichthys stellatus in San Francisco Bay, I. Hepatic contamination and
mixed-function oxidase (MFO) activity during the reproductive season. Mar Biol 98:181-189.
Spies RB, Stegeman JJ, Hinton DE, Woodin B, Smolowitz R, Okihiro M, Shea D, 1996.
Biomarkers of hydrocarbon exposure and sublethal effects in embiotocid fishes from a natural
petroleum seep in the Santa Barbara channel. Aquat Toxicol 34:195-219.
Spromberg JA, John BM, Landis WG, 1998. Metapopulation dynamics: indirect effects and
multiple distinct outcomes in ecological risk assessment. Environ Toxicol Chem 17:1640-
1649.
Stagg R, McIntosh A, 1998. Biological effects of contaminants: determination of CYP1Adependent
mono-oxygenase activity in dab by fluorimetric measurement of EROD activity.
ICES Tech Mar Environ Sc 23: 16pp.
Stebbing ARD, Dethlefsen V, Carr M (eds), 1992. Biological effects of contaminants in the
North Sea. Results of the ICES/IOC Bremerhaven workshop. Mar Ecol Prog Ser (whole
issue), 361 pp
Stegeman JJ, 1985. Benzo(a)pyrene oxidation and microsomal enzyme activity in the mussel
(Mytilus edulis) and other bivalve mollusc species from the Western North Atlantic. Mar Biol
89:21-30
Stegeman JJ, 1995. Diversity and regulation of cytochrome P450 in aquatic species. In: Arinç
E, Hodgson E, Schenkman JB (eds): Molecular aspects of oxidative drug metabolizing
enzymes: their significance in environmental toxicology, chemical carcinogenesis and health.
NATO-ASI series. New-York. Springer-Verlag. pp 135-158
Stegeman JJ, Lech JJ, 1991. Cytochrome P450 momooxygenase systems in aquatic species:
carcinogen metabolism and biomarkers for carcinogen and pollutant exposure. Environ
Health Perspect 90:101-109.
Stegeman JJ, Brouwer M, Di Giulio RT, Förlin L, Fowler BA, Sanders BM, Van Veld PA,
1992a. Enzyme and protein synthesis as indicators of contaminant exposure and effect. In:
Hugget RJ, Kimerle RA, Mehrle PM Jr, Bergman HL (eds) Biomarkers - Biochemical,
Physiological and Histological Markers of Anthropogenic Stress. Lewis Publ., Chelsea, MI.
pp. 235-335
Stegeman JJ, Brouwer M, DiGiulio RT, Forlin L, Fowler BM, Sanders BM, Van Veld PA,
1992b. Molecular responses to environmental contamination: enzyme and protein systems as
indicators of contaminant exposure and effect. In: Huggett RJ (ed), Biomarkers for chemical
contaminants, CRC Press, pp. 87-206
Stegeman JJ, Hahn ME, 1994. Biochemistry and molecular biology of monooxygenases:
current perspectives on forms, functions, and regulation of cytochrome P450 in aquatic
species, in Ostrander GK & Malins D (eds), Aquatic Toxicology: Molecular, biochemical and
cellular perspectives, Lewis Pubs, Boca Raton, FL, pp. 87-95
Stegeman JJ, Schlezinger JJ, Craddock KE, Tillitt DE, 2001. Cytochrome P4501A expression
in midwater fishes: potential effects of chemical contaminants in remote oceanic zones.
Environ Sci Tech 35:54-62

242
REFERENCES
Strathmann MF, 1987. Reproduction and development of marine invertebrates of the
Northern Pacific Coast. Chapter 26. Phylum Echinodermata, Class Asteroidea. University of
Washington Press, Seattle and London, pp. 535-555.
Stronkhorst J, Ariese F, van Hattum B, Postma JF, de Kluijverc M, den Besten PJ, Bergman
MJN, Daan R, Murk AJ, Vethaak AD, 2003. Environmental impact and recovery at two
dumping sites for dredged material in the North Sea. Environ Pollut 124:17-31
Sturesson U, 1976 Lead enrichment in shells of Mytilus edulis. AMBIO 5:253-256
Suedel BC, Dillon TM, Benson WH, 1997. Subchronic effects of five di-ortho PCB
congeners on survival growth and reproduction in the fathead minnow Pimephales promelas.
Environ Toxicol Chem 16(7):1526–1532
Sulaiman N, George S, Burke MD, 1991. Assessment of sublethal pollutant impact on
flounders in an industrialised estuary using hepatic biochemical indices. Mar Ecol Prog Ser
68: 207-212.
Suter GW, 1993. Ecological risk assessment. Boca Raton, FL: Lewis Publishers,
Svensson BG, Nilsson A, Hansson M, Rappe C, Akesson B, 1991. Exposure to dioxins and
dibenzofurans through the consumption of fish. New Engl J Med 324: 8–12
Tanabe S, 1988. PCB problems in the future : foresight from current knowledge. Environ Poll
50:5-28.
Tanabe S, Tanaka H, Tatsukawa R, 1984. Polychlorobiphenyls, DDT, and
hexachlorocyclohexane isomers in the western North Pacific ecosystem. Arch Environ
Contam Toxicol 13:731-738
Tanabe S, Tatsukawa R, 1992. Chemical modernization and vulnerability of cetaceans:
increasing toxic threat of organochlorine contaminants. In: Walker CH, Livingstone DR,
editors. Persistent pollutants in marine ecosystems. Pergamon pp. 161-177.
Taylor MG, Simkiss K, 1984. Inorganic deposits in invertebrates tissues. Environ Chem
3:102-138
Telli-Karakoç F, Tolun L, Henkelmann B, Klimm C, Okay O, Schramm KW, 2002.
Polycyclic hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs) distributions in the
Bay of Marmara sea: Izmit Bay. Environ Pollut 119:383-397
Temara A, Ledent G, Warnau M, Paucot M, Jangoux M, Dubois Ph, 1996. Experimental
cadmium contamination of Asterias rubens (Echinodermata). Mar Ecol Prog Ser 140:83-90.
Temara A, Warnau M, Jangoux M, Dubois Ph, 1997a. Factors influencing the concentrations
of heavy metals in the asteroid Asterias rubens L. (Echinodermata). Sci Tot Environ 203:51-
63
Temara A, Nguyen, QA, Hogarth AN, Warnau M, Jangoux M, Dubois Ph, 1997b. High
sensitivity of skeletogenesis to Pb in the asteroid Asterias rubens (Echinodermata). Aquat
Toxicol 40:1-10.
Temara A, Aboutboul P, Warnau M, Jangoux M, Dubois Ph, 1998a. Uptake and fate of lead
in the common asteroid Asterias rubens (Echinodermata). Water Ail- Soil Poll 102:201-208
Temara A, Skei JM, Gillan D, Warnau M, Jangoux M, Dubois Ph, 1998b. Validation of the
asteroid Asterias rubens (Echinodermata) as a bioindicator of spatial and temporal trends of
Pb, Cd, and Zn contamination in the field. Mar Environ Res 45, 341-356

243
REFERENCES
Temara A, Warnau M, Dubois Ph, 2002. Heavy metals in the sea star Asterias rubens
(Echinodermata): basis for the construction of an efficient biomonitoring program. In:
Fernandez JM, Fichez R (eds) Environmental changes and Radioactive tracers. IRD Editions,
Paris, pp. 71-91
Thurmond TS, Silverstone AE, Baggs RB, Quimby FW, Staples JE, Gasiewicz TA, 1999. A
chimeric aryl hydrocarbon receptor knockout mouse model indicates that aryl hydrocarbon
receptor activation in hematopoietic cells contributes to the hepatic lesions induced by
2,3,7,8-tetrachlorodibenzo-p-dioxin, Toxicol Appl Pharmacol 158:33-40.
Thurmond TS, Staples JE, Silverstone AE, Gasiewicz TA, 2000. The aryl hydrocarbon
receptor has a role in the in vivo maturation of murine bone marrow B lymphocytes and their
response to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol Appl Pharmacol 165:227-236.
Tijet NC, Feyereisen HR, 2001. The cytochrome P450 gene superfamily in Drosophila
melanogaster: annotation, intron-exon organization and phylogeny, Gene 262:189-198.
Torreilles J, Guérin MC, Roch P, 1999. Modified Alsever’s solution is not a good medium for
reactive oxygen metabolite study in bivalves. Fish Shellfish Immunol 8:65-69.
Triebskorn R, Köhler HR, Honnen W, Schramm M, Adams SM, Müller EF, 1997. Induction
of heat shock proteins, changes in liver ultrastructure, and alterations of fish behaviour: are
these biomarkers related and are they useful to reflect the state of pollution in the field. J
Aquat Ecosyst Stress Rec 6:57-73.
Trieff NM, Cipollaro M, Corsale, G, Esposito A, Ragucci E, Giordano GG, Ramanujam SVN,
Livingstone DR, Pagano G, 1988, Aroclor 1254 toxicity in sea urchin embryos and gametes.
Exp Oncol (Life Science Adv) 7: 57-64
Tryphonas H, 1998. The impact of PCBS and dioxins on children’s health: immunological
considerations. Can J Public Health 89:S49-S52.
Ueda T, Nakahara M, Ishii T, Suzuki Y, Suzuki H, 1979. Amounts of trace elements in
marine cephalopods. J Radiat Res 20:338-342
Ueno D, Inoue S, Ikeda K, Tanaka H, Yamada H, Tanabe S, 2003. Specific accumulation of
polychlorinated biphenyls and organochlorine pesticides in Japanese common squid as a
bioindicator. Environmental Pollution 125:227-235
UN-ECE, 1998. Draft Protocol to the Convention on Long-range Air Pollution on Persistent
Organic Pollutants, (EB.AIR:1998:2), Convention on Long-range Transboundary Air
Pollution, United Nations Economic and Social Council, Economic Commission for Europe,
UNEP/IOC/IAEA/FAO, 1990. Reference Method No 57. Contaminant monitoring programs
using marine organisms: Quality assurance and good laboratory practice. UNEP,
UNEP/FAO/IAEA/IOC, 1991. Reference Method No 12 Rev.2. Sampling of selected marine
organisms and sample preparation for the analysis of chlorinated hydrocarbons. UNEP,
UNEP/FAO/IOC/IAEA, 1993. Reference Method No 6. Guidelines for monitoring chemical
contaminants in the sea using marine organisms. UNEP,
UNEP, 2001. The Stockholm Convention on Persistent Organic Pollutants. United Nations
Environmental Programme: http://www.chem.unep.ch/sc/default.htm
US EPA (Environment Protection Agency), 2000a. National Center for Environmental
Assessment. Draft Dioxin Report: www.epa.gov/ncea/pdfs/dioxin/dioxreass.htm.

244
REFERENCES
US EPA, (Environment Protection Agency), 2000b. Method 4425: screening extracts of
environmental samples for planar organic compounds (PAHs, PCBs, PCDD/Fs) by a reporter
gene on a human cell line. EPA Office of Solid Waste, SW 846 Methods, Update IVB,
Vallack HW, Bakker DJ, Brandt I, Brorstrom-Lunden E, Brouwer A, Bull KR, Gough C,
Guardans R, Holoubek I, Jansson B, Koch R, Kuylenstierna J, Lecloux A, Mackay D,
McCutcheon P, Mocarelli P, Taalman RDF, 1998. Controlling persistent organic pollutants —
what next? Environ Toxicol Pharmacol 6:143-175.
Valsaraj KT, 1994. Hydrophobic compounds in the environment: adsorption equilibrium at
the air–water interface. Water Res 28:819-830
Van Alphen JSLJ, 1990. A mud balance for Belgian-Dutch coastal waters between 1969 and
1986. Neth J Sea Res 25:19-30.
Van den Berg M, Birnbaum L, Bosveld ATC, Brunström B, Cook P, Feeley M, Giesy JP,
Hanberg A, Hasegawa R, Kennedy SW, Kubiak T, Larsen JC, Van Leeuwen FXR, Djien AK
D, 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 wildlife. Environ Health Persp 106 (12): 775-792.
Van der Oost R, Opperhuizen A, Satumalay K, Heida H, Vermeulen NPE, 1996.
Biomonitoring aquatic pollution with feral eel (Anguilla anguilla) I. Bioaccumulation: biotasediment
ratios of PCBs, OCPs, PCDDs and PCDFs. Aquat Toxicol 35:21-46.
Van der Oost R, Beyer J, Vermeulen NPE, 2003, Fish bioaccumulation and biomarkers in
environmental risk assessment: a review. Environ Toxicol Pharmacol 13:57-49
Vanden Bossche JP, Jangoux M, 1976. Epithelial origin of starfish coelomocytes. Nature
261:227-228.
Verlaan PAJ, Donze M, Kuik P, 1998. Marine vs. fluvial suspended matter in the Scheldt
estuary. Est Coast Shelf Sci 46:873-883
Vethaak AD, Rheinallt T, 1992. Fish disease as a monitor for marine pollution: the case of the
North Sea. Rev Fish Biol Fish 2:1-32.
Viarengo A, 1985. Biochemical effects of trace metals. Mar Pollut Bull 16:153-158
Vigano L, Arillo A, Bagnasco M, Bennicelli C, Melodia F, 1993. Xenobiotic metabolizing
enzymes in uninduced and induced rainbow trout (Oncorhyncus mykiss): efffects of diets and
food deprivation. Comp Biochem Physiol 104(C): 51-55
Ville P, Roch P, Cooper EL, Masson P, Narbonne JF, 1995. PCBs increase molecular-related
activities (Lysozyme, antibacterial, hemolysis, proteases) but inhibit macrophage-related
functions (phagocytosis, wound healing) in earthworms. J Invert Pathol 65:217-224.
Voie OA, Fonnum F. 2000. effects of polychlorobiphenyls on production of reactive oxygen
species (ROS) in rat synaptosomes. Arch Toxicol 73:588-593
Voie OA, Wiik P, Fonnum F, 1998. Ortho-substituted polychlorinated biphenyls activate
respiratory burst measured as luminol-amplified chemoluminescence in human granulocytes.
Toxicol. Appl. Pharmacol. 150, 369–375.
Voldner EC, Li YF, 1995. Global usage of selected persistent organochlorines. Sci Total
Environ 160/161:201-210.
Von Boletzky S, 1983. Sepia officinalis. In : Boyle PR (ed) Cephalopod Life Cycles, Vol. 1.
Species accounts. Academic Press, London, pp. 31-52

245
REFERENCES
Voogt PA, Van Ieperen S, Van Rooyen BJCB, Wynne HJ, Janssen M, 1991. Effect of
photoperiod on steroid metabolism in the sea star Asterias rubens L. Comp Biochem Physiol
100(B):37-43
Vorderstrasse BA, Steppan LB, Silverstone AE, Kerkvliet NI, 2001. Aryl hydrocarbon
receptordeficient mice generate normal immune responses to model antigens and are resistant
to TCDD-induced immune suppression. Toxicol Appl Pharmacol 171:157-164.
Vos JG, 1980. Immunotoxicity assessement: screening and function studies. Arch Toxicol
Suppl 4:95-108
Vos JG, De Heer C, Van Loveren H, 1998. Immunotoxic effects of TCDD and toxic
equivalency factors. Teratogenesis, Carcinogenesis, and Mutagenesis 17:275–284.
Walker MK, Peterson RE, 1991. Potencies of polychlorinated dibenzo-p-dioxin,
dibenzofuran, and biphenyl congeners, relative to 2,3,7,8-tetrachlorodibenzo-p-dioxin, for
producing early life stage mortality in rainbow trout (Oncorhynchus mykiss). Aquat Toxicol
21:219-238
Walker CH, Livingstone DR (eds), 1992. Persistent pollutants in marine ecosystems.
Pergamon, pp. 161-177.
Walker CH, Hopkin SP, Sibly RM, Peakall DB, 1996. Principles of ecotoxicology. Taylor &
Francis, London, 321 pp.
Walker MK, Peterson RE, 1994. Aquatic toxicity of dioxins and related chemicals. In:
Schecter A (ed), Dioxins and health, Plenum Press, NY pp. 347-387
Wallace WG, Lopez GR, 1997. Bioavailability of biologically sequestered cadmium and the
implications of metal detoxification. Mar Ecol Prog Ser 147:149-157
Wang WX, Fisher NS, Luoma SN, 1996. Kinetic determination of trace element
bioaccumulation in the mussel Mytilus edulis. Mar Ecol Prog Ser 140:91-113
Wania F, Daly GL, 2002. Estimating the contribution of degradation in air and deposition to
the deep sea to the global loss of PCBs. Atmospheric Environment 36:5581–5593
Wania F, Mackay D, 1996. Tracking the distribution of persistent organic pollutants. Environ
Sci Technol 30(9):390-396
Warnau M, Pagano G, 1995a. Developmental toxicity of PbCl 2 in the echinoid Paracentrotus
lividus (Echinodermata) Bull Environ Contam Toxicol 53:434-441
Warnau M, Ledent G, Temara A Jangoux M, Dubois Ph, 1995b. Experimental cadmium
contamination of the echinoid Paracentrotus lividus: influence of exposure mode and
distribution of the metal in the organism. Mar Ecol Prog Ser 116: 117-124
Warnau M, Ledent G, Temara A, Alva V, Jangoux M, Dubois Ph, 1995c. Allometry of heavy
metal bioconcentration in the echinoid Paracentrotus lividus (Echinodermata), Arch Environ
Contam Toxicol 29:393-399
Warnau M, Iaccarino M, De Biase A, Temara A, Jangoux M, Dubois Ph, Pagano G, 1996a.
Spermiotoxicity and embryotoxicity of heavy metals in the echinoid Paracentrotus lividus,
Environ Toxicol Chem 15:1931-1936
Warnau M, Teyssié JL, Fowler SW, 1996b. Biokinetics of selected heavy metals and
radionuclides in the common Mediterranean echinoid Paracentrotus lividus: seawater and
food exposures. Mar Ecol Prog Ser 141:83-94

246
REFERENCES
Warnau M, Biondo R, Temara A, Bouquegneau JM, Jangoux M, Dubois Ph, 1998.
Distribution of heavy metals in the echinoid Paracentrotus lividus from the Mediterranean
Posidonia oceanica ecosystem: seasonnal and geographical variations. J Sea Res 39:267-280
Warnau M, Fowler SW, Teyssié JL, 1999. Biokinetics of radiocobalt in the asteroid Asterias
rubens: seawater and food exposures. Mar Pollut Bull 39: 159-164
Warnau M, Jangoux M, 1999. In vitro and in vivo investigations of the excretory function of
the rectal caeca in the asteroid Asterias rubens (Echinodermata). Comp Biochem Physiol
123(A):263-267
Waxman DJ, 1999. P450 gene induction by structurally diverse xenochemicals: central role of
nuclear receptors CAR, PXR, and PPAR. Arch Biochem Biophys 369:11–23.
Weeks BA, Anderson DP, DuFour AP, Fairbrother A, Goven AJ, Lahvis GP, Peters G, 1992.
Immunological biomarkers to assess environmental stress. In: Huggett RJ, Kimerle RA,
Mehrle PM, Bergman HL (eds). Biomarkers: biochemical, physiological and histological
markers of anthropogenic stress. Lewis publishers, Boca Raton, London, Tokyo, pp. 211-234
Weis P, Weis JS, 1991. The developmental toxicity of metals and methalloids in fish. In:
Newman MC, McIntosh AW (eds), Metal Ecotoxicology: Concepts and Applications. Lewis
Publ., Chelsea, MI, pp. 145-169
Weisberg S, Francisco C, Hallock D, eds, 1996. Bioaccumulation and toxicity of
polychlorinted biphenyl in sea urchins exposed to contaminated sediments, Southern
California Coastal Water Research Project Annual Report 1996, Westminster, CA, USA,
Weisberg S, Francisco C, Hallock D, eds, 1997, Relative toxicity of PCB congeners to sea
urchin embryos, Southern California Coastal Water Research Project Annual Report 1996,
Westminster, CA, USA,
Weisbrod A, Shea D, Leblanc G, Moore M, Stegeman JJ, 2000. Organochlorine
bioaccumulation and risk for whales in a Northwest Atlantic food web. Mar Environ Res
50:431-441
Weisbrod A, Shea D, Moore M, Stegeman JJ, 2001. Species, tissue and gender-related
organochlorine bioaccumulation in white-sided dolphins, pilot whales and their common prey
in the Northwest Atlantic. Mar Environ Rese 51:29-50
Weisglas-Kuperus N, Patandin S, Berbers GA, Sas TC, Mulder PG, Sauer PJ, Hooijkass H,
2000. Immunologic effects of background exposure to polychlorinated biphenyls and dioxins
in Dutch preschool children. Environ Health Persp 108:1203–1207
Wenning RJ, Di Giulio ET, Gallagher FP, 1988. Oxidant-mediated biochemical effects of
paraquat in the ribbed mussel, Geukensia demissa. Aquat Toxicol 12:157-170.
West CW, GT Ankley, JW Nichols, GE Elonen and DE Nessa. 1997. Toxicity and
bioaccumulation of 2,3,7,8-tetrachlorodibenzo-p-dioxin in long term tests with the freshwater
benthic invertebrates Chironomus tentans and Lumbriculus variegatus. Environ Toxicol
Chem 16:1287-1294
Wester PW, Vethaak D, van Muiswinkel WB, 1994. Fish as biomarkers in
immunotoxicology. Toxicology 86:213-232.
Whicker FW, Schultz V, 1982. Radioecology: nuclear energy and the environment, Vol. 2.
CRC Press, Boca Raton FL,
White E, Prives C, 1999. DNA damage enables p73. Nature 399:734-737

Whitlock JP, 1993. Mechanistic aspects of dioxin action. Chem Res Toxicol 6:754-760
247
REFERENCES
Whitlock JP, 1999. Induction of cytochrome P4501A1. Annu Rev Pharmacol Toxicol
39:103-125.
WHO International Programme on Chemical Safety (IPCS), 1993. Biomarkers and risk
assessment: concepts and principles. Environmental Health Criteria 155, World Health
Organization, Geneva.
WHO, 1997. Derivation of toxic equivalency factors (TEFs) for dioxin-like compounds in
humans and wildlife. Organohalogen Compd 34:237–40.
WHO, 1999. Dioxins and their effects on human health. Fact sheet n°225 (www.who.int/inffs/en/fact225.html)
WHO/EURO, 1991. Consultation on tolerable daily intake from food of PCDDs and PCDFs:
Summary report. Bilthoven, The Netherlands, 4–7 December 1990. EUR/IPC/PCS 030(5).
World Health Organization, Regional Office for Europe, Copenhagen,
Widdows J, Donkin P, 1992. Mussels and environmental contamination: bioaccumulation
and physiological aspects. In: Gosling E (ed), The mussel Mytilus: Ecology, physiology,
genetics and culture, Developments in aquacuolture and fisheries science, Vol 25, Elsevier
science publishers, Amsterdam, The Netherlands,
Wiesner L, Powell WH, Karchner SI, Franks DG, Cooper EL, Kauschke E, Hahn ME, 2001.
cDNA cloning of aryl hydrocarbon receptor (AhR) homologues in zebra mussel (Dreissena
polymorpha) and blue mussel (Mytilus edulis), in: Pollutant Responses in Marine Organisms
(PRIMO 11), Plymouth, England,
Wilbrink M, Treskes M, De Vlieger TA, Vermenlen NPE, 1991. Comparative toxicokinetics
of 2,2’- and 4,4’-dichlorobiphenyls in the pond snail Lymnea stagnalis (L.). Arch Environ
Contam Toxicol 19:565-571
Wilkinson L, 1988. Systat: the system to statistics. Systat Inc, Evanston, IL,
Windal I, 2001. Développement de méthodes rapides d’analyse des dioxines dans les
poussières d’incinérateurs et destruction des dioxines par eaux subcritique. Ph.D. Thesis,
University of Liège, 135 pp.
Winston GW, Di Giulio RT, 1991. Prooxidant and antioxidant mechanisms in aquatic
organisms. Aquat Toxicol 19:137-161
Wood L, O’Keefe P, Bush B, 1997. Similarity analysis of PAH and PCB bioaccumulation
patterns in sediment-exposed Chironomus tentans larvae. Environ Toxicol Chem 16:283-292
Xhoffer C, Wouters L, Grieken R, 1992. Characterization of individual particles in the North
Sea surface microlayer and underlying seawater: comparison with atmospheric particles.
Environ Sci Technol 26:2151–2162
Yamada H, Takayanagi K, Tateishi M, Tagata H, Ikeda K, 1997. Organotin compounds and
polychlorinated biphenyls of livers in squid collected from coastal waters and open ocean.
Environ Pollut 96(2):217-226
Yasuda M, Katsuki M, Fujii-Kuriyama Y, 1997. Loss of teratogenic response to 2,3,7,8tetrachlorodibenzo-p-dioxin
(TCDD) in mice lacking the Ah (dioxin) receptor, Genes Cells
2:645–654.

248
REFERENCES
Yawetz A, Manelis R, Fishelson L, 1992. The effects of Aroclor 1254 and petrochemiocal
pollutants on cytochrome p450 from the digestive gland microsomes of four species of
Mediterranean molluscs. Comp Biochem Physiol 103(C):607-614
Zabel AW, Cook PM, Peterson RE, 1995. Toxic equivalency factors of PCDD, PCDF, and
PCB congeners based on early life stage mortality in raibow trout (Oncorhynchus mykiss).
Aquat Toxicol 31:315-328
Zahner H, Fernandez-Salgero PM, Letterio J, Sheikh MS, Fornace AJ Jr, Roberts AB,
Gonzalez FJ, 1998. The involvement of aryl hydrocarbon receptor in the activation of 350
transforming growth factor-ß and apoptosis. Mol Pharmacol 54:313-321
Zar JH, 1996. Biostatistical analysis. Simon & Schuster/A Viacom company. Upper Saddle
River. NJ 07458. 3rd edition, 850 pp.

VII. ANNEX STUDIES
249

250

Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
VII.1 Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus:
seawater and food exposures to a 14 C-radiolabelled congener (PCB 153).
Environmental Pollution (in press)
Danis B a , Cotret O b , Teyssié JL b , Bustamante P c , Fowler SW b & Warnau M b
a. Laboratoire de Biologie Marine, Université Libre de Bruxelles, Belgium
b. International Atomic Energy Agency, Marine Environmental Laboratory, Monaco
c. Laboratoire de Biologie et d'Environnement Marins, Université de La Rochelle, France
251

ABSTRACT
Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
Adult Paracentrotus lividus were exposed to a 14 C-labelled PCB congener (PCB 153) using
two different exposure modes: (1) the surrounding seawater and (2) the food (viz. the
phanerogam Posidonia oceanica and the brown alga Taonia atomaria). Uptake kinetics from
water and loss kinetics after single feeding were followed in four body compartments of the
sea urchins (body wall, spines, gut and gonads). Results indicate that PCB bioaccumulation in
P. lividus varies from one body compartment to another, with the exposure mode and the
nature of the food. The echinoids accumulate PCB 153 more efficiently when exposed via
water than via the food (the transfer efficiency is higher by one order of magnitude). Target
body compartments of PCB 153 were found to be body wall and spines when individuals
were exposed via water, and gut when they were exposed via food. It is concluded that P.
lividus is an efficient bioaccumulator of PCB and that it could be considered as an interesting
indicator for monitoring PCB contamination in the marine environment.
KEYWORDS
PCB; bioaccumulation; seawater; food; echinoderm; Paracentrotus lividus
252

INTRODUCTION
Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
Polychlorinated biphenyls (PCBs) are hydrophobic contaminants that have been shown to
cause various adverse effects on a wide variety of living species (see e.g. Harding & Addison
1986). Because of their toxic potential, their production and use have been strictly regulated
and then banned in many countries since the mid-seventies (Metcalfe 1994). However, due to
their extreme resistance to physicochemical and biological degradation, PCBs have become
widely spread in the environment, and particularly in the marine environment.
Information on PCB bioaccumulation in marine benthic organisms is scarce and is generally
limited to experiments using laboratory-contaminated sediments as a source of contamination
(Meador et al. 1995, Weisberg et al. 1996, Boese et al. 1997). In addition, few studies have
taken into account essential species, upon which depends the ecosystem structure and/or
functionning (Fowler & Oregioni 1976, Phillips 1976).
The echinoid Paracentrotus lividus qualifies as an excellent bioindicator species of PCB
contamination in the Mediterranean sea. It is indeed a widely distributed, sedentary and
abundant species that plays key roles in various Mediterranean ecosystems (Hayward &
Ryland 1990), including seagrass meadows where usual indicators such as mussels are poorly
represented. Its value as bioindicator species for metal contamination is well documented by
numerous laboratory and field studies (e.g. Warnau et al. 1995b,c, 1996a,b, 1998). Several
studies have investigated the toxicological effects of PCBs on sea urchin early development
(e.g. Trieff et al. 1988, Kobayashi 1995, Weisberg et al. 1997, Schweitzer et al. 1997, 2000)
or immune system (Coteur et al. 2001), but virtually nothing is known about PCB
bioaccumulation kinetics in echinoderms in general and in P. lividus in particular.
The aim of this study was to further assess the value of P. lividus as an indicator species of
environmental PCB contamination. Therefore, PCB bioaccumulation was investigated in P.
lividus exposed through its two main contamination pathways: seawater and food.
Congener IUPAC#153 (2,2’,4,4’,5,5’ hexachlorobiphenyl) was selected as representative of
PCBs for this study: it is the most abundant congener in marine biota (Stebbing et al. 1992)
and has been shown to be an excellent indicator congener in PCB monitoring programmes
(e.g., Metcalfe 1994, Atuma et al. 1996). Finally, in order to study environmentally-realistic
simulated contaminant levels, the selected PCB congener was 14 C-labelled and measured
using a sensitive radiodetection technique (liquid scintillation).
253

MATERIALS AND METHODS
Sampling
Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
The echinoid Paracentrotus lividus (Lamarck), the phanerogam Posidonia oceanica (L.) and
the brown alga Taonia atomaria were collected in June 1999 by SCUBA diving between 5
and 10 m depth in a P. oceanica meadow off "la Pointe des Douaniers" (Cap d'Ail, France).
Prior to experimentation, specimens were acclimated to laboratory conditions for 1 month
(constantly aerated open circuit aquaria, salinity 36 ‰, 17 ± 0.5°C, 12/12 h dark/light cycle).
Radiotracer
14 C-labelled 2,2',4,4',5,5' hexachlorobiphenyl (purity > 95%) was purchased from Sigma
Chemicals, USA. Specific activity was 925 mCi mmol -1 . Stock solutions were prepared in
acetone at a concentration of 1 µg ml -1 .
Sample treatment and liquid scintillation counting
Water samples (2 ml) were directly transfered to 20 ml glass scintillation vials (Packard,
USA) and to 10 ml of Ultima Gold XR ® (Packard Instruments) scintillation liquid were
added. Subsamples of vegetal and echinoid tissues were placed in a vial containing 2 ml of
Acetonitrile ® in an ultrasonic bath for 10 min. Acetonitrile ® was then collected and replaced
by another 2 ml of Acetonitrile ® and the ultrasonic operation was repeated for a further 10
min. This treatment gave 4 ml of liquid phase (the extraction) and a residue. The residue was
digested overnight at 70°C with 2 ml of Soluene ® , and 10 ml of Hionic Fluor ® scintillation
liquid were then added . The liquid phase (4 ml) was added to 16 ml of filtered seawater and
extracted twice using 2 ml of n-hexane (Sigma, USA) under constant agitation. The organic
phase (4 ml) and the aqueous phase (20 ml) were treated separately. The entire organic phase
and 2 ml of the aqueous phase were each added separately to 10 ml of Ultima Gold XR ®
scintillation liquid.
14 C-radioactivity was measured using a 1600 TR Liquid Scintillation Analyzer (Packard),
compared to standards of known activities, and corrected for quenching, background and
physical decay of the tracer. Counting times were adjusted to obtain couting rates with
relative errors lower than 5%. PCB concentrations were eventually expressed on a total lipid
content basis. Lipids were determined according to the method of Barnes & Blackstock
(1973).
Experimental procedure
Seawater exposure
10 sea urchins (diameter: 50 ± 7 mm) were placed for 9 days in a 20 l glass aquarium
(constantly aerated closed circuit aquarium, salinity 36 ‰, 17 ± 0.5°C, 12/12 h dark/light
254

Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
cycle) containing natural seawater spiked with 14 C-labelled PCB 153. One day before starting
the experiment, two 5 l glass beakers were filled with filtered seawater (36 ‰, 17 ± 0.5°C),
spiked with the radiolabelled PCB, and constantly stirred for 24 h using an orbital agitation
plate. Contaminated water was then poured into the glass aquarium and uncontaminated
seawater was added to obtain a final volume of 20 l. Seawater and radiotracer were renewed
daily throughout the experiment. Activity was checked before and after each renewal to
assess the stability of the labelled PCB concentration in seawater (Table 42). The echinoids
were fed unlabelled fresh Posidonia oceanica leaves every second day, just before the water
renewal. After 2 h, uningested leaves were removed in order to avoid as much as possible
PCB incorporation via the food. At different times (0, 2, 5 and 8 days) echinoids (n = 3) were
collected, dissected into 4 body compartments (body wall, spines, gut, gonads), and
radioanalyzed to determine the body distribution of the PCB.
Food exposure
Table 42. Characteristics of the background and added concentrations of PCB
153. Background concentrations were measured in seawater the day before
starting the experiments; added concentrations were measured in samples of
seawater, Posidonia oceanica shoots, and Taonia atomaria thallia regularly
collected in the experimental microcosms throughout the experiment.
Compartment PCB 153 concentration
Background Seawater 0.026 ng l -1 (n = 6)
Added
Seawater
(dissolved +
particulate)
31.4 ± 15.6 ng l -1 (n = 36)
P.oceanica 5.5 ± 0.5 ng g -1 wet wt (n = 12)
T. atomaria 17.9 ± 5.6 ng g -1 wet wt (n = 12)
Shoots of Posidonia oceanica and thallia of Taonia atomaria were exposed for 14 d in two
separate glass aquaria containing 10 l natural seawater spiked with 14 C-labelled PCB 153
(Table 42). Seawater and tracer were renewed daily. Two groups of 20 sea urchins (diameter:
52 ± 6 mm) were placed in two 20 l polyvinylchloride aquaria (constantly aerated open
circuit, salinity 36 ‰, 17 ± 0.5°C, 12/12 h dark/light cycle) and allowed to feed overnight on
the previously contaminated food. In parallel, two groups of 3 echinoids were placed in
separate parts in both aquaria, and fed with uncontaminated food to serve as a control for
possible cross-contamination through seawater. After a 18 h feeding period, remaining
labelled food was removed and echinoids were fed ad libitum with uncontaminated
P.oceanica or T. atomaria until the end of the experiment. Faeces were removed twice a day
255

Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
in order to avoid as much as possible any cross contamination via PCB leaching from the
faeces. At different times (days 2, 4, 9, 11, 14, and 17 after feeding), individuals (n = 3) were
dissected into 3 body compartments (body wall, gut, gonads) in order to determine loss
kinetics and body distribution of ingested PCB.
Data analyses
Uptake of PCB 153 from seawater was expressed as change in PCB concentration (ng g -1 total
lipids) over time. Uptake kinetics were described by using the following exponential model
(Equation 10):
k t
Equation 10: C(t) = A e
where C(t) is the PCB concentration taken up at time t (d) (ng g -1 total lipids) and k is the rate
constant (d -1 ).
Regarding elimination of the radiotracer ingested with food, loss kinetics were best described
using a simple exponential model (Equation 11):
-k t
Equation 11: C(t) = C(0) e
where C(0) and C(t) are the 14 C-PCB 153 concentrations (ng g -1 total lipids) at time 0
(beginning of the loss experiment) and at time t (d), and k is the rate constant (d -1 ).
Constants of the model and their statistics were calculated by iterative adjustment and Hessian
matrix computation, respectively, using the nonlinear curve-fiting routines in the Systat ® 5.2.1
software (Wilkinson 1988).
Differences between PCB 153 concentrations in the different body compartments of the
echinoid were tested by one-way ANOVA and the multiple comparison test of Tukey (Zar
1996). Changes in PCB distribution among body compartents were tested for significance
using the G-test (adapted from the log-likelihood ratio test) for 2xk contingency tables (Zar
1996). Prior to this test, data were arcsin-transformed, using the correction of Freeman-Tukey
(1950) as described in Zar (1996). The level of significance for statistical tests was always set
at a = 0.05.
RESULTS
Seawater experiment
The uptake of PCB #153 by P. lividus exposed to contaminated seawater was followed in four
body compartments (body wall, spines, gut and gonads) (Fig. 55). Parameters and statistics of
256

Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
the uptake kinetics are given in Table 43. Estimated uptake rate constants were quite
homogeneous among the different body compartments. This indicates that bioaccumulation
efficiency was similar, particularly in body wall and spines and in gut and gonads.
PCB conentration (ng g -1 lipids)
PCB conentration (ng g -1 lipids)
300
200
100
Body wall
0
0 2 4 6 8 10
400
300
200
100
Spines
Time (d)
0
0 2 4 6 8 10
Time (d)
Figure 55. Paracentrotus lividus. Uptake kinetics of 14 C-PCB 153 from seawater in
4 body compartments of the sea urchin (mean concentration ng g -1 total lipids ± SD,
n=3)
Table 43. Paracentrotus lividus. Parameters and statistics of the equation fitting
the uptake of 14 C-PCB 153 in the body compartments of echinoids exposed via
seawater: C(t) = A e k.t . C(t): 14 C-PCB 153 concentration (ng g -1 lipids) at time t
(d); A: ordinate at the origin (ng g -1 lipids); k: rate constant (d -1 ); ASE:
asymptotic standard error; R 2 : determination coefficient.
A (ASE) k (ASE) R 2
Body wall 7.05 (4.8) 0.45 (0.09) 0.93
Spines 20.5 (5.2) 0.34 (0.03) 0.97
Gut 0.53 (0.88) 0.53 (0.21) 0.80
Gonads 0.86 (1.4) 0.53 (0.21) 0.80
Table 44 presents the concentrations of stable PCB 153 corresponding to the radiolabelled
congener incorporated at different times over the exposure period (Table 44A) and the
corresponding distribution among the considered body compartments (Table 44B). The body
wall and spines concentrated C 14 -PCB 153 to the highest degree, reaching mean
257
PCB conentration (ng g -1 lipids)
PCB conentration (ng g -1 lipids)
50
40
30
20
10
80
70
60
50
40
30
20
10
Digestive tract
0
0 2 4
Time (d)
6 8 10
Gonads
0
0 2 4 6 8 10
Time (d)

Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
concentrations up to 262 and 319 ng g -1 lipids, respectively. These concentrations were one
order of magnitude higher than in gut and gonads (p Tukey test < 0.0001; Table 44A). Body
distribution of the 14 C-PCB 153 varied significantly (G-test; p < 0.05) among sampling days
(Table 44B). However, spines and, secondarily, body wall always displayed the highest
proportion (29-62%) of total body load of incorporated PCB.
Table 44. Paracentrotus lividus. Concentrations and distribution of PCB 153 incorporated in the different body
compartments of the echinoids exposed to the congener via seawater. A. PCB 153 concentrations (mean ng g -1
lipids ± SD, n = 3). Mean concentrations sharing the same superscript do not differ significantly between each
other. B. PCB 153 distribution (mean % ± SD, n = 3) among body compartments.
A.
Body wall Spines Gut Gonads
Day 2 27.6 a ± 20.8 23.5 a ± 4.480 1.43 a ± 1.71 1.56 a ± 2.39
Day 5 62.7 b ± 25.8 130 c ± 7.450 7.35 d ± 3.16 12.0 d ± 5.15
Day 8 262 e ± 54.3 319 e ± 35.90 35.4 f ± 15.3 57.6 f ± 24.9
B.
Body wall Spines Gut Gonads
Day 2 46.3 ± 23.3 50.5 ± 26.3 1.8 ± 2.5 2.0 ± 2.9
Day 5 28.7 ± 7.8 62.3 ± 8.1 8.7 ± 2.7 5.7 ± 2.1
Day 8 38.6 ± 5.2 47.3 ± 1.8 9.9 ± 4.6 9.2 ± 4.8
Food experiment
Two sets of sea urchins were allowed to feed overnight either on P. oceanica or on T.
atomaria previously exposed to 14 C-labelled PCB 153. Sea urchins were then placed in
uncontaminated conditions in order to determine the loss kinetics of the ingested PCB.
Analysis of control animals showed that there was no significant cross contamination through
seawater due to food leaching.
A latency time of 2 to 4 days was observed before the contaminant concentration reached a
maximum in the body compartments (Fig 56, Table 45). Loss kinetics were calculated taking
into account the period between that maximal value and the end of the experiment.
258

15
10
5
Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
Figure 56. Paracentrotus lividus. Loss kinetics of 14 C-PCB 153 (mean concentration ng g -1 total lipids ± SD, n =
3) in three body compartments of the sea urchin after a single feeding on radiolabelled food (Posidonia oceanica,
(left) or Taonia atomaria (right)).
Table 45. Paracentrotus lividus. PCB concentrations (ng g -1 lipids; mean ± SD, n = 3)
measured in the sea urchin body compartments after a single feeding, using two different
food (Posidonia oceanica vs Taonia atomaria). Mean concentrations sharing the same
superscript do not differ significantly among each other (p Tukey test > 0.05).
P. oceanica
Body wall
0
0
30
5 10
Time (d)
15 20
20
10
Gut
0
0 5 10
Time (d)
15 20
30
20
10
Gonads
0
0 5 10
Time (d)
15 20
Body wall Gut Gonads
Day 2 10.0 a ± 1.74 10.8 a ± 2.03 2.26 a ± 0.064
Day 4 11.8 c ± 2.59 20.8 c ± 4.47 25.9 c ± 8.79
Day 9 4.58 e ± 0.69 4.60 e ± 1.95 3.28 e ± 0.96
Day 11 5.51 g ± 2.07 3.87 g ± 0.52 3.31 g ± 0.81
Day 14 2.37 h ± 1.08 0.095 h ± 0.106 1.00 h ± 0.64
Day 17 1.83 i ± 1.40 0.16 i ± 0.057 0.59 i ± 0.13
259
30
20
10
0
0 5 10 15 20
10
Time (d)
8
6
4
2
Body wall
Gut
0
0 5 10 15 20
Time (d)
3
2
1
0
0 5 10 15 20
Time (d)
Gonads

T. atomaria
Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
Body wall Gut Gonads
Day 2 19.6 j ± 6.22 5.94 j ± 0.26 2.75 j ± 0.23
Day 4 12.7 k ± 2.02 7.90 k ± 1.35 2.59 k ± 1.73
Day 9 4.86 l ± 1.96 6.13 l ± 0.90 1.65 l ± 0.23
Day 11 3.01 m ± 1.82 5.74 m ± 1.14 1.14 m ± 0.071
Day 14 3.26 o ± 1.30 1.57 o ± 0.35 1.31 o ± 0.26
Day 17 0.77 p ± 0.43 1.09 p ± 0.26 1.12 p ± 0.47
Loss kinetics were similar for both food considered (P. oceanica and T. atomaria). The main
differences between the two feedings was the decrease of radioactivity in the gut contents:
after P. oceanica feeding, 14 C-PCB activity decreased exponentially in the gut contents,
whereas, with T. atomaria, it decreased linearly as a function of time (data not shown).
Loss of ingested 14 C-PCB 153 followed a one-component exponential model in each body
compartments (Table 46). The kinetics were characterized by rapid loss of ingested PCB:
calculated biological half-lives (Tb 1/2) ranged between 2.5 and 9.2 days.
Table 46. Paracentrotus lividus. Parameters and statistics of the equation describing the loss of PCB
#153 from the sea urchin body compartments after a single feeding on Posidonia oceanica and
Taonia atomaria. Equation is C(t) = C(0) e -k t ; where C(t) and C(0) are 14 C-PCB 153 concentrations
(ng g -1 lipids) at time t (d) and time 0, respectively, and k is the rate constant (d -1 ). ASE: asymptotic
standard error; R 2 : corrected determination coefficient; Tb 1/2: Biological half-life (d).
P. oceanica
C(0) (ASE) k (ASE) R 2
Tb1/2 Body wall 11.7 (0.99) 0.145 (0.023) 0.83 4.8
Gut 20.8 (1.30) 0.282 (0.042) 0.94 2.5
Gonads 29.9 (1.14) 0.376 (0.049) 0.98 1.8
T. atomaria
C(0) (ASE) k (ASE) R 2
Tb1/2 Body wall 19.3 (1.43) 0.193 (0.031) 0.87 3.6
Gut 8.5 (0.84) 0.095 (0.019) 0.71 7.3
Gonads 2.49 (0.46) 0.075 (0.031) 0.42 9.2
DISCUSSION
The present study reports the first experimental data on the bioaccumulation kinetics of a key
PCB congener in the sea urchin Paracentrotus lividus, a common species widely distributed
in the Mediterranean Sea and on the NE Atlantic coast.
260

Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
The biokinetic experiments carried out in this study were performed using a 14 C-labelled PCB
congener (2,2',4,4',5,5' hexachlorobiphenyl) and were designed in order to expose sea urchins
to moderate to high PCB concentrations as commonly found in the marine environment.
PCB 153 was shown to be efficiently accumulated from seawater by the sea urchin. This
observation matches other uptake experiments using for example polychaetes and fish
exposed to Aroclor (Fowler et al. 1978, Shaw & Connell 1987). All the considered body
compartments (body wall, spines, gut, gonads) accumulated the congener following
exponential uptake kinetics. Even if the exposure period was rather short (8 d), concentration
factors (ratio between 14 C-PCB in body compartments and in surrounding seawater) were
quite elevated and ranged between 10 3 (in the soft tissues: gut and gonads) and 10 5 (in the
calcified tissues: body wall and spines). This indicates the efficiency of the sea urchin organs
as bioaccumulator compartments and, hence, their usefulness as tools for the survey and
biomonitoring of PCB contamination in the marine environment. Being easily dissected and
constituting ca. 90% of the total body weight, body wall and spines are of particular interest
with respect to field studies, and they should be recommended as body compartments to
monitor.
After feeding with common food of radiolabelled, loss of PCB 153 displayed exponential
kinetics. The loss was quite rapid in each compartment. Biological half-lives of the PCB
congener ranged between 2.5 and 9 d. This indicates a low retention of the PCB taken up
through the trophic chain. Nevertheless, over the long term, this pathway could contribute
significantly to the total load of PCB in the sea urchin. It is notewothy that, even if PCB 153
was 3 times more concentrated in T. atomaria thallia given as food than in P. oceanica leaves
(see Table 42), the congener concentration incoporated in the sea urchin soft tissues (gut and
gonads) were higher when the animals were exposed via P. oceanica. (see Tables 45 and 46).
This would indicate a higher bioavailability of the PCB congener when it is incorporated in
the P. oceanica tissues than in those of T. atomaria. Conversely, retention of ingested PCB
was 3 to 5 fold stronger in soft tissues of sea urchins fed T. atomaria (see Table 46).
While this work constitutes the first report on PCB bioaccumulation kinetics in an adult sea
urchin, previous studies have used radiolabelled 14 C-PCB to examine bioaccumulation in
other aquatic organisms (e.g., Goerke et al. 1973, Gooch & Hamdy 1982, Schweitzer et al.
1997). However, these studies are few and mostly concern PCBs as Aroclor mixtures (see e.g.
Butcher et al. 1997). The main advantage of the 14 C approach to measure PCB
bioaccumulation in aquatic biota is obviously the high sensitivity and the rapidity of the
detection, compared to analytical techniques using gas chromatography. Furthermore, it
261

Bioaccumulation of PCBs in the sea urchin Paracentrotus lividus
allows working with low, realistic PCB concentrations, and assessing uptake in individual
organs which are often too small to be analyzed by classical chemical without pooling.
ACKNOWLEDGEMENTS
The IAEA Marine Environment Laboratory operates under a bipartite agreement between the
International Atomic Energy Agency and the Government of the Principality of Monaco. B.D.
is holder of a FRIA doctoral grant. M.W. is a Honorary Research Associate of the National
Fund for Scientific Research (NFSR, Belgium). Research was partially supported by a
Belgian Federal Research Programme (SSTC, Contract MN/11/30) and a NFSR fellowship to
M.W.
262

263
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
VII.2 Bioaccumulation of PCBs in the cuttlefish Sepia officinalis from
seawater, sediment and food pathways.
Environmental Pollution (submitted)
Danis B a , Bustamante P b , Cotret O c , Teyssié JL c , Fowler SW c & Warnau M c
a. Laboratoire de Biologie Marine (CP 160-15), Université Libre de Bruxelles, 50 Av. F.D.
Roosevelt, B-1050 Brussels, Belgium
b.Laboratoire de Biologie et Environnement Marins, UPRES-EA 3168, Université de La
Rochelle, 22 Av. Michel Crépeau, F-17042 La Rochelle Cedex, France
c. Marine Environment Laboratory - International Atomic Energy Agency, 4 Quai Antoine I er ,
MC-98000 Monaco

ABSTRACT
264
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
The cuttlefish Sepia officinalis was selected as a model cephalopod to study PCB
bioaccumulation via seawater, sediments and food. Newly hatched, juvenile cuttlefish were
exposed for 17 days to environmentally realistic concentrations of 14 C-labelled 2,2',4,4',5,5'-
hexachlorobiphenyl (PCB 153) (18 ng PCB l -1 seawater; 30 ng PCB g -1 dry wt sediments;
Artemia salina exposed to 18 ng PCB l -1 seawater). Accumulation of PCB 153 was followed
in three body compartments: digestive gland, cuttlebone and the combined remaining tissues.
Results showed that (1) uptake kinetics were source- and body compartment-dependent, (2)
for each body compartment, the accumulation was far greater when S. officinalis was exposed
via seawater, (3) the cuttlebone accumulated little of the contaminant regardless of the source,
and (4) the PCB congener showed a similar distribution pattern among the different body
compartments following exposure to contaminated seawater, sediment or food with the lowest
concentrations in the cuttlebone and the highest in the remaining tissues. The use of
radiotracer techniques allowed delineating PCB kinetics in small whole organisms as well as
in their separate tissues. The results underscore the enhanced ability of cephalopods to
concentrate organic pollutants such as PCBs, and raise the question of potential risk to their
predators in contaminated areas.
KEYWORDS
Cephalopods, persistent organic pollutants, kinetics, transfer, distribution

INTRODUCTION
265
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Among persistent organic pollutants (POPs), polychlorobiphenyls (PCBs) are a class of
anthropogenic contaminants of long-standing environmental concern (Livingstone et al.,
2000). Represented by 209 congeners, PCBs are widespread, highly conservative and are
readily accumulated in living organisms. PCBs entering marine waters become available for
marine organisms whose biology can be affected at various organizational levels (Metcalfe
1994). Moreover, these contaminants may also be biomagnified in food webs raising a
potential risk for high trophic level predators. Marine invertebrates take up these
contaminants via three main routes: ambient seawater (through gills and body surfaces), direct
contact with sediments, and ingestion of food. Previous studies using sea stars and sea urchins
(Danis et al. Chap. III.2, VII.1) have shown that the efficiency of PCBs bioaccumulation
depends on the source of contamination (viz. seawater is probably the main contamination
source) and on the considered congener (Danis et al. Chap. III.1).
Cephalopods, including benthic (octopus), nectobenthic (cuttlefish), neritic and oceanic
(squids) species, are widely distributed in the world oceans. They generally display short life
span and high growth rates. Owing to the economic value of many cephalopod species, they
are of high commercial interest for fisheries (Forsythe & Heukelem 1987, Navarro &
Villanueva 2000). Moreover, cephalopods play key roles in marine ecosystems, being both
active predators of fish and crustaceans (Castro & Guerra 1990, Rodhouse & Nigmatullin
1996) and important prey items for marine mammals, seabirds and fish (Clarke 1996, Croxall
& Prince 1996, Klages 1996, Smale 1996). Therefore, cephalopods can be considered as an
important vector for transferring potentially hazardous contaminants to top marine predators
(Bustamante et al. 1998, Weisbrod et al. 2000, 2001).
Cephalopods are known to accumulate numerous contaminants among which are POPs such
as organochlorine pesticides or PCBs (Tanabe et al. 1984, Kawano et al. 1986, Yamada et al.
1997, Weisbrod et al. 2000, 2001, Ueno et al. 2003). However, very little is known about
bioaccumulation capacity depending upon the routes of exposure to these contaminants; such
data are needed to further assess the potential impact of organic pollutants on cephalopod
populations. For these reasons, experiments were designed to study the bioaccumulation of
PCBs by the common cuttlefish Sepia officinalis following exposure via seawater, sediments
and food.
PCB biokinetics were determined using radiotracer techniques in order to measure fluxes at
environmentally realistic concentrations (Danis et al. Chap. III.2, VII.1). These techniques
have already proven useful when examining kinetic behaviour in organs that are too small for

266
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
employing classical PCB analytical measurements have to be considered. The PCB congener
#153 (2,2’,4,4’,5,5’ hexachlorobiphenyl) was selected as a model PCB, since it is the most
abundant in marine biota and has been recognized as an indicator of total PCB contamination
(see e.g., Stebbing et al. 1992, Atuma et al. 1996).
MATERIALS AND METHODS SECTION
Organisms
Seven- to nine-day old newly hatched cuttlefish (hereafter called juveniles) were used in the
experiments. Cuttlefish eggs were spawned in the laboratory by adults collected by net fishing
off Monaco and were maintained in aquaria until hatching. Young cuttlefish were kept in a
separate aquarium (open circuit, 20 l h -1 flow rate, constant aeration, 36 p.s.u., 16.5 ± 0.5 °C,
12h/12h light/dark cycle) and fed brine shrimp (Artemia salina) until used in the experiments.
Radiotracer
The PCB radiotracer ( 14 C-labelled 2,2',4,4',5,5'-hexachlorobiphenyl; purity ≥ 95%) was
purchased from Sigma Chemicals, USA. Specific activity was 925 MBq mmol -1 and stock
solutions were prepared in acetone at a concentration of 1 µg ml -1 and stored at –20°C until
used.
Liquid scintillation counting
Water samples (2 ml) containing the radiotracer were directly transferred to 20 ml glass
scintillation vials (Packard, USA) and mixed with 10 ml of scintillation liquid (Ultima Gold
XR ® , Packard, USA). Sediment and biota samples (cuttlefish tissues and brine shrimp) were
ultrasonified twice for 10 min, each time with 2 ml of Acetonitrile ® (Packard, USA). This
treatment produced a liquid phase (4 ml) containing the extracted 14 C-PCB and a residue. The
residue was digested overnight at 70°C with 2 ml of Soluene ® (Packard, USA) and mixed
with 10 ml of scintillation liquid (Hionic Fluor ® , Packard, USA). The liquid phase was added
to 16 ml of filtered seawater and extracted twice using 2 ml of n-Hexane (Sigma, USA) under
agitation. The organic phase (4 ml) and the aqueous phase (20 ml) were treated separately.
The whole organic phase, and 2 ml of the aqueous phase were mixed separately with 10 ml of
Ultima Gold XR ® scintillation liquid. The treatment is summarized in Figure 57.

267
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Figure 57. Diagrammatic representation of sample processing prior to
liquid scintillation counting.
ß-emission of the tracer was measured in the samples using a 1600 TR Liquid Scintillation
Analyzer (Packard). The counting time was selected to obtain counting rates with relative
propagated errors less than 5% (maximum counting duration: 2 h). Radioactivity measured in
the samples was compared to standards and corrected for quenching, background and physical
decay of the tracer. PCB concentrations were expressed on a total lipid content basis where
lipids were determined according to the method of Barnes & Blackstock (1973).
Experimental procedures.
Uptake from seawater
Juvenile cuttlefish (mean wet wt ± SD = 0.141 ± 0.039 g; n = 25) were placed for 20 d in a 70
l glass aquarium containing natural seawater spiked with 18 ng 14 C-PCB 153 l -1 seawater. This
concentration corresponds to a moderate level of contamination in the North Sea (Stebbing et
al. 1992). One day before starting the experiment, four 5 l glass beakers were filled with
filtered seawater (36 p.s.u., 16.5 ± 0.5°C), spiked with the PCB stock solution, and constantly
stirred using an orbital agitation plate. Contaminated seawater was poured into the glass
aquarium which was subsequently filled to a final volume of 70 l with uncontaminated

268
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
seawater. This operation (spiked seawater preparation) was performed daily throughout the
duration of the experiment and radiolabelled seawater was renewed daily in the aquarium.
Radioactivity was checked before and after each seawater renewal to assess the stability of the
labelled PCB concentration. All cuttlefish were fed twice a day with Artemia salina and were
periodically sampled. After 1 h, uningested brine shrimp were removed to limit as much as
possible incorporation of PCB through the food. At each sampling time, 3 individuals were
dissected to determine the distribution of the radiotracer among digestive gland, cuttlebone
and the remaining tissues.
Uptake from sediments
Sediments (2.5 kg dry wt) from the North Sea (Audresselles, Pas-de-Calais, France) were
contaminated for 4 d with the 14 C-labelled PCB using the rolling jar method (Murdoch et al.,
1997). Sediments were then placed in a 70 l glass aquarium and maintained overnight under
flowing seawater (open circuit, 20 l h -1 flow rate, 36 p.s.u., 16.5 ± 0.5 °C) to allow any loosely
bound contaminant to leach. The seawater level in the aquarium was then reduced so that a 2
cm layer of natural seawater was running over the 3 cm layer of spiked sediments. This was
done in order to minimize cuttlefish movements required for feeding and to optimize their
contact time with sediments. Juvenile cuttlefish (mean wet wt ± SD = 0.124 ± 0.046 g; n =
25) were then placed for 17 d in the aquarium during which time the sediment and seawater
radioactivity was periodically checked. These measurements showed that 14 C-PCB
radioactivity and concentration (9.49 ± 1.14 ng g -1 dry wt) remained constant in the sediments
throughout the experiment, and that no radioactivity could be detected in seawater. Cuttlefish
were fed twice daily with Artemia salina and any uningested food was removed after 1 h and
ß-counted. No activity could be detected in the brine shrimp. Cuttlefish were periodically
sampled to follow PCB uptake kinetics and distribution among the tissues over time.
Uptake from food
Before the feeding experiment, brine shrimp were exposed for 7 d in a glass aquarium
containing 4 l of filtered seawater spiked with 18 ng 14 C-PCB 153 l -1 (24-h water spiking
procedure as described above for seawater). Radiolabelled seawater was renewed daily and
brine shrimp were regularly fed a mixture of phytoplankton. After 7 d, brine shrimp were
used to feed the juvenile cuttlefish. (Fresh brine shrimp were labelled daily to have 7-d
labelled food available each day throughout the experiment.) Juvenile cuttlefish (mean wet wt
± SD = 0.130 ± 0.029 g; n = 25) were placed in individual plastic containers (10 cm diameter,
5 cm height) and held in a 70 l glass aquarium (open circuit, 20 l h -1 flow rate, 36 p.s.u., 16.5
± 0.5 °C). During the entire experiment (17 d), individuals were allowed to ingest

269
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
radiolabelled A. salina twice a day for 1 h; food was given in excess and any non-ingested
shrimp remaining after the feeding period were immediately removed). Cuttlefish were
periodically sampled over time to follow PCB uptake kinetics and distribution among the
tissues.
Data analyses
Uptake of the 14 C-PCB congener from seawater, sediments and food was expressed as change
in PCB concentration (ng g -1 total lipids) over time. Radiotracer uptake kinetics were
described either by using a linear model (Equation 12), a saturation exponential model
(Equation 13), or a combined model (logistic plus exponential) (Equation 14):
Equation 12: C(t) = C o +k.t
Equation 13: C(t) = C ss (1-e -k.t )
Equation 14: C(t) = C ss (1-e -k.t ) / 1+e -k.(t-I)
where C(t), C o, and C ss are the 14 C-PCB concentrations (ng g -1 total lipids), respectively, at
time t (d), at time 0, and at steady-state, k is the rate constant (d -1 ) and I is the time (d) at the
inflexion point. The model showing the most accurate fit (based on the calculation of the
determination coefficient, R 2 , and examination of the residuals) was used. Constants and
statistics of the different models were estimated by iterative adjustment of the models and
Hessian matrix computation, respectively, using the nonlinear curve-fitting routines in Systat ®
5.2.1 (Wilkinson 1988). Differences among 14 C-PCB concentrations in the different body
compartments were tested using one-way ANOVA and the multiple comparison test of Tukey
(Zar 1996). After arcsine-transformation of data (using the correction of Freeman-Tukey
1950; in Zar 1996), changes in 14 C-PCB body distribution were tested for significance using
the G-test (adapted from the log-likelihood ratio test) for 2 x k contingency tables (Zar 1996).
The level of significance for statistical tests was always set at a = 0.05.
RESULTS
Uptake kinetics of 14 C-PCB 153 by S. officinalis were determined by exposing juveniles via
three different pathways: seawater, sediments and food. Accumulation was followed in three
body compartments: digestive gland, cuttlebone and remaining tissues (which included all the
other tissues and organs). Accumulation was also considered in whole-body organisms using
reconstituted data of the separate tissues.

Uptake from seawater
270
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Bioaccumulation kinetics and their parameters in the different body compartments of the
cuttlefish following exposure to environmentally realistic seawater PCB concentrations (18
ng l -1 ) are shown in Fig. 59 and Table 47.
Figure 58. Sepia officinalis. Uptake of 14 C-PCB 153 from seawater in the different body
compartments and in whole-body juvenile cuttlefish (mean ng g -1 total lipids ± SD; n = 3)
Table 47. Sepia officinalis. Parameters and statistics of the equations describing the uptake of 14 C-PCB #153 in
different body compartments of the cuttlefish exposed via seawater, sediments and food.
L (linear model): C(t) = C 0.+ k.t
S (saturation model): C(t) = Css.(1-e -k.t );
C (combined model): C(t) = Css.(1-e -k.t )/(1+e -k.(t-I) );
C 0, C(t), Css: PCB concentrations (ng g -1 lipids) at, respectively, time 0, time t (d) and steady-state; k: rate
constant (d -1 ); I: time (d) at the inflexion point; ASE: asymptotic standard error; R 2 : corrected
determination coefficient.
Body compartment Model C 0 (ASE) Css (ASE) k (ASE) I (ASE) R 2
Seawater
Digestive gland L 0 1.25 (0.05) 0.92
Cuttlebone -
Remaining tissues L 0 7.45 (0.22) 0.95
Whole body L 3.86 (3.76) 6.36 (0.38) 0.94

271
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Sediments
Digestive gland C 6.94 (0.52) 4.40 (16.6) 2.22 (0.82) 0.75
Cuttlebone -
Remaining tissues S 47.9 (3.11) 0.44 (0.09) 0.78
Whole body S 39.0 (1.94) 0.49 (0.08) 0.77
Food
Digestive gland C 6.71 (0.44) 0.97 (0.55) 2.85 (0.54) 0.82
Cuttlebone -
Remaining tissues L -1.14 (0.69) 1.25 (0.07) 0.91
Whole body L -1.28 (0.64) 0.94 (0.07) 0.91
Except for cuttlebone, uptake of the contaminant was best described using linear models.
Cuttlebone took up little, if any, PCB congener. Among the compartments, the remaining
tissues concentrated PCB 153 to the greatest degree, almost one order of magnitude higher
than in the digestive gland and two orders of magnitude higher than in the cuttlebone (p Tukey test
≤ 0.0001). Concentration factors (ratio between PCB concentration in organism and in an
equal weight of seawater) were calculated for the compartments considered and for the
whole-body (Table 48).
Table 48. Sepia officinalis. Concentration factors and transfer factors (CF, TF; mean ± SD; n = 3) in the body
compartments and in whole-body juvenile cuttlefish at the end of the experimental exposures.
CFs are calculated as the ratio between PCB 153 concentration in the body compartments (ng g -1 total lipids) and
its concentration in seawater (ng g -1 ). TFs are calculated as the ratio between PCB 153 concentration in the body
compartments (ng g -1 total lipids) and its concentration in sediments (ng g -1 dry wt) or food (ng g -1 lipids).
Digestive gland Cuttlebone Remaining tissues Whole body
Seawater 527 ± 120 37.0 ± 24.7 3280 ± 513 2990 ± 449
Sediments 0.74 ± 0.26 0.00 ± 0.00 5.31 ± 0.77 4.65 ± 0.47
Food 1.81 ± 0.24 0.01 ± 0.03 3.72 ± 0.71 3.45 ± 0.61
The tissue distribution of 14 C-PCB 153 varied significantly (log-likelihood ratio, G-test)
throughout the experiment (Fig. 59). At the beginning of the experiment, all of the
contaminant (100 %) was found in the remaining tissues. Then, during the first days of
exposure, a progressive transfer to the digestive gland took place, reaching a peak of about
15% of the total load after 4 d. The remaining tissues always contained the major fraction (80-
100%) of the total 14 C-PCB body load, whereas very low proportions were found in the
cuttlebone (0-2.5%).

100%
80%
60%
40%
20%
0%
Uptake from sediments
272
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Digestive Gland Cuttlebone Remaining Tissues
day1 day2 day4 day7 day11 day14 day17
Figure 59. Sepia officinalis. 14 C-PCB 153 distribution (mean %) among the
different body compartments during the seawater experiment.
As observed in the seawater experiment, cuttlebone did not appear to accumulate 14 C-PCB
during sediment exposure. Accumulation kinetics in the digestive gland were best described
by a combined model (logistic + exponential components), whereas accumulation in the
remaining tissues and in whole organisms was best fitted by a saturation exponential model
(Fig. 60, Table 47). Similar to the seawater experiment, the body compartment that took up
PCB 153 to the greatest degree was the remainder. Nevertheless, transfer factors between
sediments and organisms remained low, reaching maximum values of 5.3 (Table 48).

273
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Figure 60. Sepia officinalis. Uptake of 14 C-PCB 153 from sediments in the different body
compartments and in whole-body juvenile cuttlefish (mean ng g -1 total lipids ± SD; n = 3)
The tissue distribution of 14 C-PCB was determined at different times during the experiment
(Fig. 61). The remaining tissues contained the major fraction (84-99%) of the total PCB body
load, whereas the lowest proportions were found in the cuttlebone (0-1.3%). At day 1, the
remaining tissues contained 99 % of the radiotracer, with a progressive transfer to the
digestive gland taking place during the first days of exposure and reaching a peak of about 13
% of the total load after 4 d.
100%
80%
60%
40%
20%
0%
20
15
10
5
0
0 5 10 15 20
2.0
1.5
1.0
0.5
Digestive Gland
C(t)=6.94.(1-e -4.40.t )/(1+e -4.40.(t-2.22) )
Cuttlebone
Time (d)
0.0
0 5 10 15 20
0
0 5 10 15 20
Digestive Gland Cuttlebone Remaining Tissues
day1 day2 day4 day7 day11 day14 day17
Figure 61. Sepia officinalis. 14 C-PCB 153 distribution (mean %) among the
different body compartments during the sediment experiment.
100
60
50
40
30
20
80
60
40
20
Remaining tissues
C(t)=47.9.(1-e -0.44.t )
R 2 =0.75 R 2 =0.78
Whole-body
Time (d)
C(t)=39.0.(1-e -0.49t )
10
0 5 10 15 20
Time (d) Time (d)
R 2 =0.77

Uptake from food
274
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Throughout the experiment, juvenile cuttlefish were fed radiolabelled brine shrimp ad libitum
for 1 h, twice a day. Uptake kinetics of 14 C-PCB 153 ingested with food followed the
combined (logistic + exponential) model in the digestive gland and the linear model in the
remaining tissues and in whole organisms (Fig. 62, Table 47). The separate body
compartments (except cuttlebone) displayed PCB concentrations and transfer factors of the
same order of magnitude (Fig. 62, Table 48).
Figure 62. Sepia officinalis. Uptake of 14 C-PCB 153 in the different body compartments
and in whole-body juvenile cuttlefish (mean ng g -1 total lipids ± SD; n = 3) following
ingestion of radiolabelled food (Artemia salina).
As expected, the distribution of the contaminant among cuttlefish tissues determined at
different times showed that at the beginning of the experiment, the ingested radiotracer was
entirely associated with the digestive gland (Fig. 63). Over time, the proportion of 14 C-PCB
activity significantly decreased in the digestive gland and increased in the remaining tissues
(G test, p < 0.05).
15
10
5
0
0 5 10 15 20
5
4
3
2
1
Digestive Gland
Cuttlebone
Time (d)
0
0 5 10 15 20
Time (d)
25
20
15
10
5
0
0 5 10 15 20
20
15
10
5
Remaining tissues
C(t)=6.71.(1-e -0.97.t )/(1+e -0.97.(t-2.85) ) C(t)=-1.14+1.25.t
R 2 =0.82
Whole-Body
Time (d)
C(t)=-1.28+0.94.t
0
0 5 10 15 20
Time (d)
R 2 =0.91
R 2 =0.91

DISCUSSION
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
275
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Digestive Gland Cuttlebone Remaining Tissues
day1 day2 day4 day7 day11 day14 day17
Figure 63. Sepia officinalis. 14 C-PCB 153 distribution (mean %) among the
different body compartments during the food experiment.
Cephalopods are well-known for their capacity to accumulate radioactive, metallic or organic
contaminants to relatively high levels (see e.g. Ueda et al. 1979, Miramand & Bentley 1992,
Yamada et al. 1997, Bustamante et al. 2000). However, only few studies have aimed at
investigating the concentrations and distribution of PCBs in these molluscs, and most of these
studies have only considered whole organisms as a source of contaminant for their predators
(Tanabe et al. 1984, Kawano et al. 1986, Weisbrod et al. 2000, 2001). Some studies have
employed cephalopods as bioindicators of local or global PCB contamination (Butty &
Holdway 1997, Yamada et al. 1997), and other ones have demonstrated cellular or tissue
effect of organic pollutants (Mann et al. 1988, Cheah et al. 1995). Nevertheless, a review of
available literature clearly shows a general lack of information regarding bioaccumulation
processes in cephalopods.
The present experimental work, exposing the common cuttlefish S. officinalis to
environmentally realistic concentrations of a radiolabelled PCB congener, demonstrated the
potential of these organisms to bioconcentrate PCBs to high levels. The bioconcentration
efficiency was far greater when cuttlefish were exposed through seawater compared to either
the food or sediment pathways. Indeed, PCB 153 concentration factors (CFs) from seawater
were 2 to 3 orders of magnitude higher than transfer factors from sediment or food. These
differences are in close agreement with observations from sea stars and sea urchins exposed
to the same contaminant under similar conditions (Danis et al. Chap. III.2, VII.1).

276
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
The distribution of the radiolabelled PCB congener was examined in three body
compartments (the digestive gland, cuttlebone and the remaining tissues of juvenile cuttlefish)
and the design of the experiments allowed following the uptake kinetics of the PCB in the
separate body compartments. The digestive gland was selected, since this organ plays a major
role in the energetic metabolism of cephalopods (Boucaud-Camou & Boucher-Rodoni 1983)
and has been documented to accumulate PCBs to high levels (Yamada et al. 1997, Ueno et al.
2003). The cuttlebone was also examined as this calcareous compartment acts as an internal
skeleton and represents ca. 3 % of the total body weight of the juveniles. Finally, the third
compartment (remaining tissues) comprises the rest of the animal (ca. 90 % of the total body
weight) and mainly consists of muscles. The tracer experiments showed a similar distribution
pattern between the three compartments irrespective of the source of exposure: fractions of
the radiolabelled PCB were always much higher in the remaining tissues, followed by the
digestive gland and with very little amounts in the cuttlebone. Following exposure via
seawater, sediments and food, the cuttlebone did not significantly concentrate the PCB
congener. In fact, the cuttlebone contained less than 2.5 % of the total contaminant load and
only for a limited period of time. This is likely due to the internal localization of the
cuttlebone which appears to have no direct contact with ambient seawater and to the low fat
content of this tissue.
In contrast to the cuttlebone, the digestive gland readily accumulates PCB 153. The highest
concentrations in the digestive gland were reached following seawater exposure. After 17 d of
exposure, juvenile cuttlefish continued to concentrate the PCB in their digestive gland in a
linear fashion. On the other hand, PCB 153 uptake in the digestive gland displayed saturation
kinetics following both sediment and food exposures. These results are particularly surprising
for the digestive gland as it (a) has no direct contact with ambient seawater and (b) plays a
major role in the digestive processes, including nutrient absorption. However, this organ is
also involved in detoxification processes of xenobiotics in cephalopods (Cheah et al. 1995,
Bustamante et al. 2002). Therefore, elevated concentrations in the digestive gland could result
more from transfers from other organs accumulating compounds to be detoxified than from
actual bioconcentration in the digestive gland itself. At this stage, it is difficult to explain the
precise role of the cuttlefish digestive gland in the metabolism of PCBs; however, complex
PCB redistribution processes clearly do occur among the tissues of this cephalopod.
Therefore, it would be of major interest to characterize precisely the distribution of PCBs in
adult cephalopods by making a finer separation of all tissues and organs.

277
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
Among the three compartments examined in these experiments, most of the PCB 153 taken
up (≥ 80 %) was located in the remaining tissues regardless of the exposure pathway.
However, the PCB concentration factors (CFs) calculated in the remaining tissues following
the seawater experiment was three orders of magnitude higher than transfer factors (TFs)
computed at the end of the sediment or food experiments. The fact that the remaining tissues
are in direct contact with seawater could explain such a very high degree of accumulation
following seawater exposure. However, cuttlefish also spend a substantial part of their time in
very close contact with sediments; therefore, differences between CFs and TFs most probably
also reflect differences in PCB bioavailability between the seawater and sediment pathways.
The uptake kinetics of PCB 153 in the remaining tissues followed a linear model when
cuttlefish were exposed via seawater or food, and a saturation model when exposed via
sediments. Saturation concentrations were relatively low if compared to studies with sea stars
exposed to sediments spiked with similar PCB 153 concentrations (Danis et al. Chap. III.2).
Sea stars displayed the same uptake kinetics but reached much higher PCB 153
concentrations, especially in compartments in direct contact with sediments (≥ 3,000 ng g -1
lipids in the body wall).
CONCLUSIONS
Although the cuttlefish used in this study were early juveniles, the results following seawater,
sediment and food exposures have shown that PCBs are incorporated to high levels in their
tissues. Even if this study was a first approach to understanding the pathways of PCB
contamination in cephalopods, it highlights the general lack of knowledge concerning
mechanisms underlying PCB bioaccumulation, distribution and depuration. Finally, the main
finding from this study is that seawater appears to be the main route for PCB incorporation in
juvenile cuttlefish. Therefore, cuttlefish might be useful bioindicators of ambient water PCB
contamination. Assuming that cuttlefish are representative of cephalopods in general, many
species from contaminated areas could play a major role in the transfer of these persistent
organic pollutants to their predators.
ACKNOWLEDGEMENTS
We thank Prof. E. Boucaud-Camou for advice on rearing cuttlefish. The IAEA Marine
Environment Laboratory operates under a bipartite agreement between the International
Atomic Energy Agency and the Government of the Principality of Monaco. BD is holder of a

278
Bioaccumulation of PCBs in the cuttlefish Sepia officinalis
FRIA and of a Van Buuren doctoral grants. MW is a Honorary Research Associate of the
National Fund for Scientific Research (NFSR, Belgium). Research was supported by a
Belgian Federal Research Programme (SSTC, Contract MN/11/30) and by a NFSR short-term
fellowship to M. Warnau.

279
Measurement of EROD activity
VII.3 Measurement of EROD activity: Caution on the spectral properties of
the standards used
Marine Biotechnology (in press)
Radenac G a , Coteur G b , Danis B b , Dubois Ph b & Warnau M c
a. Laboratoire de Biologie & Environnement Marins, EA 3168 – Université de La Rochelle,
La Rochelle, France.
b. Laboratoire de Biologie Marine, Université Libre de Bruxelles, Brussels, Belgium.
c. International Atomic Energy Agency - Marine Environment Laboratory, Principality of
Monaco.

ABSTRACT
280
Measurement of EROD activity
The activity of the enzyme EROD has been extensively used in biomonitoring studies for
more than a decade. Although the analytical procedure is quite simple, it is often poorly
characterized. In this study, spectral properties of particular standard compounds used for
EROD activity measurement (viz. ethoxyresorufin and resorufin, standards from Molecular
Probes®) were tested in order to optimise excitation and emission wavelengths to be used in
the fluorimetric assay of EROD activity. The optimal excitation wavelength for the detection
of resorufin was 560 nm. Indeed, at this wavelength, the excitation represents only 37% of its
maximum level for ethoxyresorufin while it represents 86% for resorufin. This allows
discrimination between the fluorescence emitted by both standards favouring the formed
resorufin. Our results demonstrate that any analytical work using spectrofluorometry to
measure EROD activity should be preceded by a precise determination of the spectral
characteristics of each set of standards used.
KEYWORDS
EROD, ethoxyresorufin, resorufin, spectral characteristics, fluorimetry, biomarker

INTRODUCTION
281
Measurement of EROD activity
The use of the enzyme 7-ethoxy-resorufin-O-deethylase (EROD) activity as an environmental
biomarker was suggested some thirty years ago (Burke & Mayer 1974). From that time, the
main objective of the numerous studies dealing with EROD activity has been to assess the
effects of contamination by specific pollutants on target organisms in the marine environment
(Addison & Edwards 1988, Galgani et al. 1991, Holdway et al. 1994). Indeed, a few organic
compounds such as polyaromatic hydrocarbons (PAHs) and polychlorinated biphenyls
(PCBs) induce the EROD activity of the enzyme CYP1A, an enzyme that is involved in the
detoxication of these organic pollutants (Sulaiman et al. 1991). Because of the good
sensitivity of this biomarker in marine vertebrates, it has been widely used in biomonitoring
studies for more than a decade (Galgani et al. 1992).
The analytical procedure for EROD activity measurement is quite simple and consists in
determining (generally using spectrofluorimetry) the efficiency of a given biological sample
to convert an experimentally-added substrate (ethoxy-resorufin) into a specific product
(resorufin) (Grzebyk & Galgani 1991). However, several studies highlighted the lack of
knowledge on the "natural" variability of this enzymatic activity (Galgani & Payne 1991).
More recently, it has been shown that factors such as season (Eggens et al. 1995), sexual
maturity stage in female individuals (Burgeot et al. 1994) or growth (Sleiderink et al. 1995)
strongly influence EROD activity. In addition, it is now well known that experimental
parameters (e.g., storage temperature, pH of extraction buffers, incubation temperature)
governing in the laboratory during sample treatments may also affect this activity (Grzebyk &
Galgani 1991, Burgeot et al. 1994). Surprisingly, except in some cases (e.g. Burke & Mayer
1974, 1983) optimal wavelengths for the measurement of this enzymatic activity are not
systematically characterized and reported in published studies , even if several technical
reports questioned the purity of the standards used (Eggens & Galgani 1992, Munkittrick et
al. 1993, Stagg & McIntosh 1998).
When testing for EROD activity, low amounts of resorufin are formed compared to the
amounts of ethoxyresorufin added to the samples. In addition, these two products have similar
chemical (and thus spectral) properties. This implies that even slight departure from optimal
spectral parameters could easily result in inaccurate measurement of the fluorescence due
specifically to the end-product resorufin. Therefore, the aim of the present work was to test
the spectral properties of the reference standards (viz. ethoxyresorufin and resorufin) used for
EROD activity determination, in order to determine the optimal excitation and emission

282
Measurement of EROD activity
wavelengths which should be used to measure accurately the apparition of the resorufin in the
incubation mixture.
MATERIALS AND METHODS
The standards tested in this study (R–352 for ethoxyresorufin –EthR– and R–363 for resorufin
–Res –; references from Molecular Probes®) were stored under optimal conditions:
refrigerated (4°C) and protected from light in order to avoid any degradation.
Stock solutions were prepared in DMSO (Stagg and McIntosh, 1998); their concentrations
were first determined by spectrophotometry using an extinction coefficient of 54.0 cm -1 mM -1
for Res (absorbance read at l = 572 nm) and of 16.0 cm -1 mM -1 for EthR (absorbance read at
l = 494 nm) (Molecular Probes-®, unpublished data). Homogenisation of the Res stock
solution was extended overnight and was facilitated using ultrasonication during 10 minutes
at room temperature (Kennedy and Jones, 1994). Working solutions were freshly prepared in
a 0.1M phosphate buffer, pH = 7.6 and kept at 4°C in the dark. The final concentrations of
EthR and Res used in this study were 2µM and 2.5 nM, respectively. These concentrations
were selected since they represent the less favorable conditions to detect the fluorescence due
specifically to Res. Indeed, 2.5 nM Res corresponds to the lowest concentration that can be
measured accurately with this method and 2 µM EthR is the concentration that displays the
maximum of fluorescence activity (Stagg and McIntosh, 1998).
Fluorimetric measurements were performed during the first minute of incubation within a 2ml
cuvette at a temperature of 20°C, using a RF-5001 PC Shimadzu fluorimeter. Excitation and
emission spectra of the tested standards (EthR and Res) were carried out in order to determine
the maximum excitation and emission wavelengths for each chemical.
RESULTS AND DISCUSSION
A wide range of excitation wavelengths for EROD activity measurement are reported in
published studies: e.g. 510 nm (Galgani & Payne 1991), 530 nm (Holdway et al. 1994,
Brumley et al. 1995), 535 nm (Stagg & McIntosh 1998), 537 nm (Hewitt et al. 1998), 538 nm
(Gunther et al. 1997), 544 nm (Burke & Mayer 1974, Grzebyk & Galgani 1991, Burgeot et al.
1994). In order to determine the optimal excitation wavelength to be used for the detection of
Res in a mixture of EthR and Res, excitation spectra of both compounds were recorded with a
fixed emission wavelength corresponding to the peak emission of Resorufin (i.e. 584 nm).

283
Measurement of EROD activity
Maximum excitation was reached at l = 494 nm for EthR and l = 572 nm for Res (Fig. 64).
The shape of each spectrum differed quite significantly. Res showed a narrow excitation
spectrum whereas, for EthR, the spectrum displayed a large plateau from 494 nm (the
maximum of excitation) to 535 nm: at this wavelength, the excitation of EthR still represented
86% of its maximum value.
900
800
Fluorescence intensity (F.U.)
700
600
500
400
300
200
100
100%
0
450 470 490
494
510 530
535
550
560
570 572
Figure 64. Excitation spectra of ethoxyresorufin (2 µM) and resorufin (2.5 nM) standards from Molecular
Probes® - Emission values at 584 nm are expressed as relative values compared to the maximum of fluorescence
intensity. Y-scales of both curves are independent.
This plateau was not detected in another study (Galgani & Payne 1991) testing the same
products purchased from another company (Sigma®). These authors recorded a narrower
peak characterized by a maximum of excitation at 457 nm while less than 20% of the
maximum excitation remained at 535 nm. In our case, it appears that an excitation of 535 nm
(which is the most commonly used wavelength) would be most inappropriate to detect Res in
the reaction mixture. Indeed, EthR would be the main excited (and thus detected) compound.
This hypothesis was tested by comparing emission spectra of both molecules excited at 535
nm (Fig. 65). They showed very close wavelengths of maximal emission: 576 nm for EthR
and 584 nm for Res. This is consistent with the similarity of their chemical conformation.
Moreover, the emission at l = 584 nm from the EthR was greater than the one from the
product to be detected, i.e. Res (Fig. 65).
86%
41%
Excitation wavelenght (nm)
Ethoxyresorufin
Resorufin
86%
37%
100%

Fluorescence intensity (F.U.)
800
700
600
500
400
300
200
100
0
576
111 %
584
284
Measurement of EROD activity
540 560 580 600 620 640 660
Emission wavelenght (nm)
Figure 65. Emission spectra of ethoxyresorufin (2 µM) and resorufin (2.5 nM) standards from Molecular
Probes® - Excitation values at 535 nm are expressed as relative values compared to the maximum of
fluorescence intensity. Same Y-scales for both curves.
It appears that an optimal detection of the reaction product would consist of exciting at a
wavelength close to the peak of excitation of Resorufin (572 nm) but not close to the emission
peak (584 nm) in order to prevent any interference by the excitation beam. For this reason,
standards were excited at l = 560 nm; emission at l = 584 nm due to EthR was about four
times lower (23%) than that of resorufin (Fig. 66). At this wavelength, the excitation
represents only 37% of its maximum level for EthR and 86% for Res (Fig. 64). Therefore,
although close to the emission wavelength, this excitation wavelength (560 nm) would be
efficient to discriminate between the set of standards tested here. To solve the problem of
wavelength proximity, slits were reduced, thereby preventing any interference between
excitation and emission beams (3 and 5 nm respectively).
Ethoxyresorufin
Resorufin

Fluorescence intensity (F.U.)
120
100
80
60
40
20
0
285
Measurement of EROD activity
Figure 66. Emission spectra of ethoxyresorufin (2 µM) and resorufin (2.5 nM) standards from Molecular
Probes® - Excitation values at 560 nm are expressed as relative values compared to the maximum of
fluorescence intensity. Same Y-scales for both curves.
CONCLUSIONS
23 %
584
540 560 580 600 620 640 660
In addition to common analytical controls, spectral characterization of standards used should
also be carefully and systematically checked before any assay of EROD activity using a new,
unknown batch of EthR as substrate. The lack of such control could question the accuracy of
the assay. For the standard used in the present study (references from Molecular Probes®),
optimum excitation wavelength was determined to be 560 nm, a wavelength which is much
higher than those used in previous works.
Emission wavelenght (nm)
These recommendations are particularly relevant for users of plate-reader fluorimeters.
Indeed, these very easy-to-use and powerful equipements are more and more commonly used
for EROD assays. However, most of the plate-readers are equipped with filters of limited
wavelength range. They are thus not able to run scans over a wide range of wavelengths and
hence to check the spectral characteristics of the substrate.
Resorufin
Ethoxyresorufin

ACKNOWLEDGEMENTS
286
Measurement of EROD activity
We gratefully thank the team of the Organic Chemistry Laboratory (ULB) for providing
access to analytical facilities and for insightful advises. Ph.D. and M.W. are, respectively,
Research Associate and Honorary Research Associate of the National Fund for Scientific
Research (NFSR, Belgium). G.C. and B.D. were holders of a FRIA doctoral grant. This work
was supported by the SSTC covenant MN-11-30 (Prime Minister Services, Belgium). This is
a contribution of the "Centre interuniversitaire de Biologie marine (CIBIM)". The IAEA
Marine Environment Laboratory operates under a bipartite agreement between the
International Atomic Energy Agency and the Government of the Principality of Monaco.

Effects of PCBs on ROS production by Paracentrotus lividus
VII.4 Effects of PCBs on reactive oxygen species (ROS) production by the
immune cells of Paracentrotus lividus (Echinodermata).
Marine Pollution Bulletin 42(8):667-672
Coteur G a , Danis B a , Fowler SW b , Teyssié JL b , Dubois Ph a & Warnau M b
a. Laboratoire de Biologie Marine, CP 160/15, Université Libre de Bruxelles, Av. F. D.
Roosevelt, 50, B-1050 Brussels, Belgium.
b. Marine Environment Laboratory, International Atomic Energy Agency, P.O. Box 800, 4
Quai Antoine 1 er , MC-98000 Monaco.
287

ABSTRACT
Effects of PCBs on ROS production by Paracentrotus lividus
The impact of four PCB congeners: 3,3’,4,4’-tetrachlorobiphenyl (IUPAC congener #77),
3,3’,4,4’,5-pentachlorobiphenyl (IUPAC #126), 2,2’,4,4’,5,5’-hexachlorobiphenyl (IUPAC
#153) and 3,3’,4,4’,5,5’-hexachlorobiphenyl (IUPAC #169) was investigated on the reactive
oxygen species (ROS) production by coelomocytes of the echinoid Paracentrotus lividus, an
important species in marine benthic ecosystems. PCBs were found to increase ROS
production and to delay the time of peak production. These effects were stronger on bacteria-
stimulated cells and were congener-specific: coplanar congeners (#77, 126 and 169) had more
effect than the non-coplanar PCB 153. Among coplanar congeners, PCB 169 showed dose-
dependent effects whereas PCB 77 and 126 were more toxic at high and low doses,
respectively. The relative immunotoxicity of the different PCB congeners is discussed in the
light of their structural properties and biological affinities.
KEYWORDS
Polychlorinated biphenyls, immune system, reactive oxygen species, coelomocytes, congener-
specific immunotoxicity, echinoderms, chemiluminescence.
288

INTRODUCTION
Effects of PCBs on ROS production by Paracentrotus lividus
Echinoderms are a major phylum of deuterostome invertebrates that includes a number of
species playing key-roles in numerous marine ecosystems (Harrold & Pearse 1987, Birkeland
1989, Menge et al. 1994). By their occurrence in coastal and estuarine waters, they are
directly exposed to anthropogenic contaminants. A number of these contaminants have been
shown to affect several aspects of the physiology of echinoderms such as their reproduction
(den Besten et al. 1989), early development (Kobayashi 1995, Warnau et al. 1996a), somatic
growth (Temara et al. 1997b), and neurophysiology (Mallefet et al. 1994). Surprisingly, the
impact of contaminants on the immune system of echinoderms has never been assessed.
Echinoderms lack a specific adaptive immune system and rely on innate responses involving
both humoral and cellular components (Chia & Xing 1996). Effector cells are the so-called
coelomocytes, i.e. free-circulating cells found in the coelomic cavities. Five different types of
coelomocytes (amoebocytes, spherule cells, vibratile cells, crystal cells and progenitor cells)
are involved in responses such as encapsulation (Jans et al. 1996), phagocytosis (Bertheussen
1981), hydrolytic enzyme secretion (Canicatti, 1990) and reactive oxygen species (ROS)
production (Ito et al. 1992). These processes appear very efficient and constitute the main line
of internal defence against non-self material. ROS production has recently received much
attention in the field of invertebrate immunology and appears highly sensitive to xenobiotic
exposure (Anderson et al. 1997). ROS production most probably occurs via the activation of a
membrane-associated NAD(P)H-oxidase univalently reducing molecular oxygen to
superoxide anion ( . -
O2 ) which in turn can lead to the production of other oxidants (Babior
1984).
By their very long persistence in the environment and the fact that they are readily
bioaccumulated and highly toxic for organisms, polychlorinated biphenyls (PCBs) are among
the contaminants of highest concern in marine ecosystems. Most of the studies on PCB
immunotoxicity in invertebrates have concerned the impact of commercial mixtures (e.g.
Aroclor®) in annelids (Ville et al. 1995; Burch et al. 1999). However, it is now widely
accepted that the toxicity of such mixtures is mainly due to a few congeners, viz. the non-
ortho-substituted and mono-ortho-substituted congeners (Duinker et al. 1989, Safe 1990,
Metcalfe 1994). Moreover, the relative proportions of PCB congeners in these mixtures are
very different from those found in contaminated environments (Metcalfe 1994). We thus
chose to work on a congener-specific basis, with the most toxic (the non-ortho-substituted
289

Effects of PCBs on ROS production by Paracentrotus lividus
coplanar 3,3’,4,4’-tetrachlorobiphenyl [IUPAC congener #77], 3,3’,4,4’,5-
pentachlorobiphenyl [#126] and 3,3’,4,4’,5,5’-hexachlorobiphenyl [#169]; Safe et al. 1985)
and the most abundant (2,2’,4,4’,5,5’-hexachlorobiphenyl [#153]; Muir et al., 1988)
congeners. PCBs were injected into the coelomic cavity in order to obtain final concentrations
comparable to those found in natural seawater in moderately contaminated environments
(around 1 ng l -1 for oceanic water; for a review, see Fowler 1990). The impact of these
congeners was evaluated on the ROS production (measured as peroxidase/luminol – enhanced
chemiluminescence) in echinoderms. The edible echinoid Paracentrotus lividus, a well-
known key-species of commercial importance in the Atlantic ocean and in the Mediterranean,
was used as a model for this study.
MATERIALS AND METHODS
Test organisms
Adult specimens of Paracentrotus lividus (5.8 ± 0.5 cm ambital diameter, 90 ± 5 g wet
weight) were collected at Cap d’Ail, France. They were immediately brought to the laboratory
and acclimated for 1 week in open circuit aquarium (capacity: 3000 l, salinity: 38‰,
temperature: 17 ± 0.5°C, flow rate: 50-100 l h -1 with constant aeration). They were fed ad
libitum the brown alga Taonia atomaria.
PCB exposures
Stock solutions of the different congeners (Promochem, Germany) were prepared according
to the method of Murdoch et al. (1997). Briefly, aliquots of PCB stock solutions (prepared in
acetone) were added to decanted natural seawater to obtain final concentrations of 5 µg l -1
(PCB 153) or 1.25 µg l -1 (PCBs 77, 126 and 169). Samples were constantly agitated for 3h to
allow for acetone evaporation and solution homogeneity, and finally serially diluted in
decanted natural seawater (with 3 h agitation between each dilution).
Eventually, 0.5 ml of the PCB solutions was injected into the coelomic cavity of the sea
urchins. A control group was injected with seawater alone; blank animals were not injected.
Injected doses of PCBs were: 2.5, 0.28 and 0.03 ng for PCB 153 and 0.625, 0.069 and 0.008
ng for PCBs 77, 126 and 169). Assuming a total volume of coelomic fluid of 50 ml per sea
urchin, coelomic PCB concentration range would be: 0.6 – 50 ng l -1 for PCB #153 and 0.16 –
12.5 ng l -1 for PCBs 77, 126 and 169. Four replicates (i.e. four animals) per treatment were
sampled 43 h after the PCB (or control) injection.
290

ROS production measurements
Effects of PCBs on ROS production by Paracentrotus lividus
An aliquot of 3 ml of coelomic fluid, obtained by cutting the peristomial membrane of the sea
urchins, was poured into 3 ml anticoagulant buffer (0.012 M EDTA in Ca-, Mg-free artificial
seawater -CMFASW-; Noble, 1970) at 4°C. The coelomocyte concentration of this
suspension was determined using a Thoma haemocytometer. The suspension was then
centrifuged (400xg for 10 min, 4°C) and the supernatant replaced by CMFASW to obtain a
final concentration of 1 ± 0.25 x 10 6 cells ml -1 . This concentration was double-checked as
described above.
Reactive oxygen species were measured by the luminol-enhanced chemiluminescence method
(Lambert & Nicolas 1998) with the following modification: horseradish peroxidase (HRP)
was added in order to increase the chemiluminescence signal and to render the reaction more
specific to peroxides (Pagano et al. 1997). For peroxidase/luminol-enhanced
chemiluminescence (PLCL) measurement, 500 µl of luminol (0.1 mg ml -1 ) / HRP (0.05 mg
ml -1 ) in artificial seawater (ASW, Sigma) was added to 100 µl of a suspension of autoclaved
bacteria (Micrococcus luteus, 2.5 x 10 9 bacteria ml -1 in ASW) (stimulated ROS production) or
an equivalent volume of ASW (unstimulated ROS production). The chemical background (i.e.
the chemiluminescence of the above-described medium) was measured and, subsequently,
400 µl of coelomocyte suspension (1 x 10 6 cells ml -1 ) was added. Samples were placed at
13°C and the PLCL (summed on a 5 sec period) was measured every 10 min over a 2 h period
using a EG&G LB-9507 luminometer equipped with a 100-fold attenuation filter.
Measurements were normalised with the actual coelomocyte concentration in each sample
and usually expressed either as the sum of all 10 min-interval measurements for 10 6 cells
(“total chemiluminescence”) or as the time needed to reach the peak PLCL value (“time-to-
peak”).
Statistical analyses
The occurrence of a dose-response relationship was tested by linear and non-linear
regressions. When regression statistics were not significant for a given congener, differences
among immune responses observed for the different doses of this PCB were tested for
significance using analysis of variance (one-way ANOVA) followed by the multiple
comparison test of Tukey (Zar 1996). All statistical analyses were carried out using the Systat
5.2.1‚ software (Wilkinson 1988).
291

RESULTS
Kinetics of reactive oxygen species (ROS) production
Effects of PCBs on ROS production by Paracentrotus lividus
In control and blank animals stimulated coelomocytes showed 4-5 times more ROS
production compared to unstimulated cells (Fig. 67). The peak chemiluminescence of
stimulated coelomocytes was reached after 30-50 min, the signal then decreased slowly and
levelled off at 90 min. The ROS production by coelomocytes of seawater-injected sea urchins
(controls) was never significantly different from non-injected animals (blanks) (p > 0.1);
results from both these controls were thus pooled for further statistical analyses.
Chemiluminescence
(RLU /10 6 cells)
250
200
150
100
Figure 67. Stimulated vs. unstimulated ROS production by coelomocytes
of P. lividus: chemiluminescence (mean ± SE, n=4) of bacteria-stimulated
(black squares) or unstimulated (white dots) coelomocytes over time.
Effects of PCB congeners on ROS production
50
0
0 50 100 150
In order to get an insight into the effects of PCB congeners, three different parameters were
analysed: total chemiluminescence of unstimulated and bacteria-stimulated coelomocytes and
time-to-peak of stimulated ROS production kinetics.
Concerning unstimulated chemiluminescence, none of the four PCB congeners tested showed
significant effects (Fig. 68). When coelomocytes were bacteria-stimulated, different
behaviours of ROS production were observed according to the congener tested (Fig. 69).
Whereas treatment with congener #153 showed no significant effect over the considered dose
range, congener #169 affected ROS production dose-dependently according to the following
linear relationship: Y = 5386 X + 3093 (R 2 =0.253, p regression=0.047), where Y is the
chemiluminescence (in Relative Light Units, RLU, per 10 6 cells) and X is the injected dose
292
Time (min)

Effects of PCBs on ROS production by Paracentrotus lividus
(ng). No dose-response relationships were significant (p>0.1) for PCBs #77 and 126.
However, congener #126 significantly affected ROS production (p Tukey=0.017) at the lowest
dose injected (0.008 ng), whereas congener #77 significantly affected ROS production
(p Tukey=0.029) only at the highest dose tested (0.625 ng). As for ROS production, effect on
time-to-peak also varied according to the considered congener (Fig. 70). Treatment with PCB
#153 showed no significant effect.
Unstimulated ROS production
(RLU / 10 6 cells)
3000
2500
2000
1500
1000
500
0
control 0.001 0.01 0.1 1 10
PCB dose injected (ng)
Figure 68. ROS production (total chemiluminescence) by
unstimulated coelomocytes of P. lividus according to the PCB dose
injected: PCB #77 (black dots), #126 (white squares), #153
(asterisks) and #169 (black triangles). Controls were injected with
seawater alone. (Mean ± SE, n=4).
Stimulated ROS production
(RLU/ 10 6 cells)
14000
12000
10000
8000
6000
4000
2000
*
0
control
0.001 0.01 0.1 1 10
PCB dose injected (ng)
Figure 69. ROS production (total chemiluminescence) by stimulated
coelomocytes of P. lividus according to the PCB dose injected: PCB
#77 (black dots), #126 (white squares), #153 (asterisks) and #169
(black triangles). Controls were injected with seawater alone. (Mean ±
SE, n=4). (*) significantly different from the control; (¥) significant
dose-response relationship.
293
*
¥

Stimulated time-to-peak (min)
90
80
70
60
50
40
30
20
10
*
*
Effects of PCBs on ROS production by Paracentrotus lividus
0
control 0.001 0.01 0.1 1 10
PCB dose injected (ng)
Figure 70. Time-to-peak of the ROS production kinetics of
stimulated P. lividus coelomocytes according to the PCB dose
injected: PCB #77 (black dots), #126 (white squares), #153
(asterisks) and #169 (black triangles). Controls were injected with
seawater alone. (Mean ± SE, n=4). (*) significantly different from
the control.
Congeners #77 and 126 failed to show dose–response relationships but both congeners
significantly increased the time-to-peak (p Tukey=0.02 and 0.001, respectively) at low PCB dose
(0.008 ng).
Regarding congener #169, the time-to-peak tended (p regression=0.086) to be affected according
to a dose-response relationship that was best described by the following logarithmic equation:
Y = 2.7 log(X) + 57.7 (R 2 =0.196), where Y is the time-to-peak (min) and X is the injected
dose (ng).
Table 49 summarises the effects of the four PCB congeners on parameters of ROS production
by coelomocytes of P. lividus. It appears that congeners #77, 126 and 169 have the most
potent immunomodulatory activity in contrast to PCB 153 that has none.
Table 49. Summary of the observed effects of different PCB congeners on ROS production by Paracentrotus
lividus coelomocytes. Legend: +, significant stimulation (p < 0.05); (+), significant stimulation (0.05 < p < 0.1);
X, no significant effect (p > 0.1); hc, effect observed at high concentration; lc, effect observed at low
concentration; dr, significant dose-response relationship.
ROS production parameter
PCB congener Unstimulated ROS Stimulated ROS Stimulated time-to-peak
production
production
#77 X +, hc +, lc
#126 X +, lc +, lc
#169 X +, dr (+), dr
#153 X X X
294

DISCUSSION
Effects of PCBs on ROS production by Paracentrotus lividus
In the echinoid P. lividus, PCBs stimulated the cellular response of ROS production while
delaying the time of peak production. PCBs are known to inhibit a number of cellular
responses in invertebrates such as the formation of secretory-rosettes and erythrocyte-rosettes,
wound healing and phagocytosis (Cooper 1992, Fitzpatrick et al. 1992, Goven et al. 1993,
1994). On the contrary, an increased respiratory burst in response to PCBs has been described
in human polymorphonuclear cells and granulocytes (Raulf & König 1991; Voie et al. 1998).
It was demonstrated that these latter effects were due to the stimulation of enzymatic activities
involved in signal transduction such as phospholipase and protein kinase activities (Voie et al.
1998). Interestingly, in invertebrates, the oxidative activity of the immune cells can be
triggered by the protein kinase C activator PMA (phorbol myristate acetate) (Dikkeboom et
al. 1987). Moreover, this immune response is calcium-dependent (Torreilles et al. 1999) and
most probably involves a NAD(P)H-oxidase (Noël et al. 1993). It thus seems that ROS
production in invertebrates is very similar to the respiratory burst of vertebrate phagocytes.
Therefore, we suggest that, in echinoderms, PCBs act on targets corresponding to those found
in vertebrates (i.e. the signal transduction pathway leading to the activation of the ROS-
producing oxidase). This is supported by the fact that PCBs act differentially according to the
activation status of coelomocytes: while bacteria-stimulated ROS production was increased,
unstimulated cells remained unaffected. While ROS production acts as an efficient cytotoxic
mechanism against non-self material, this increased production could be detrimental by
triggering peroxidation of self-tissues.
PCB congeners were found to affect differentially the immune response in P. lividus: non-
ortho-substituted coplanar congeners (#77, 126 and 169) enhanced and delayed ROS
production; the non-coplanar congener #153 failed to affect this response. The former ones
show close structural similarities with the well-known highly toxic 2,3,7,8-
tetrachlorodibenzo-p-dioxin (2,3,7,8-T 4CDD), whereas non-coplanar congeners are known to
interact to a lesser extent with physiological mechanisms (Metcalfe 1994, Schweitzer et al.
1997). Our results are thus in good agreement with previous studies on the toxicity of single
PCB congeners in relation to their chemical properties (for a review, see Metcalfe 1994).
Among coplanar PCBs, differences were found in their relative toxicity: PCB #126 stimulated
ROS production at low dose only; PCB 169 affected the ROS production according to a dose-
dependent relationships; and PCB 77 had intermediate effects. The particular response to
PCB126 could suggest the induction of antioxidant enzymes or molecules at higher doses of
this congener. Indeed, superoxide dismutase (SOD, scavenging superoxide anions), catalase
295

Effects of PCBs on ROS production by Paracentrotus lividus
(scavenging hydrogen peroxide) or glutathione peroxidase (GPx) activities were demonstrated
in a number of invertebrate phyla (for a review, see Livingstone 1991). They were showed to
be inducible by xenobiotics such as paraquat (Wenning et al. 1988). Although nothing is
known on these enzymes in echinoderms, occurrence of antioxidant molecules of the
carotenoid family were demonstrated in sea urchins (Shina et al. 1978). Alternatively, the
absence of ROS production stimulation at higher doses of PCB 126 could reflect the
induction of detoxication mechanisms: components of the cytochrome P450 monooxygenase
(MO) system, responsible for the biotransformation of aromatic compounds, have been
reported in seastars (den Besten et al. 1990b, 1993) and sea urchins (Payne & May 1979). It
was demonstrated that, among PCB congeners 118, 126 and 153, only the coplanar congener
#126 induced the cytochrome P450 monooxygenase (MO) system in the seastar Asterias
rubens (den Besten et al. 1993). In fact, PCB 126 is known to be a 3-methylcholanthrene (3-
MC)-type inducer of the mammalian MO system. The induction of a MO system in P. lividus
by PCB 126 could thus explain the disappearance of the effect of this congener at increasing
doses, although the concentration range of PCB 126 used in this study (corresponding to
2.6x10 -7 to 2.0x10 -5 µmole kg -1 wet weight animal) were much lower than what was
demonstrated as MO-inducing (10 µmole kg -1 ) by den Besten et al. (1993).
In conclusion, all of the tested PCB congeners showed different pattern of effects (Table 1):
PCB 153 shows no dose-response relationship nor effects at specific doses, PCB 77 shows no
dose-response relationship but affect ROS production at high dose, PCB 126 shows no dose-
response relationship but affect ROS production at low dose and eventually PCB 169 affects
ROS production according to a dose-response relationship. Therefore, the effect is dependent
on the congener tested and can thus be considered as congener-specific. Indeed, it has been
reported that PCB congeners act differentially in relation to their biological reactivity
(Borgman et al. 1990; Wilbrink et al. 1991).
It is concluded that PCBs affect (and possibly impair) the immune system of echinoderms in a
congener-dependent manner, thereby suggesting that PCBs, particularly coplanar congeners,
actually represent a threat to macrobenthos communities. Further work is needed to determine
the exact mechanism(s) underlying the enhancement of ROS production by these xenobiotics.
ACKNOWLEDGEMENTS
The IAEA Marine Environment Laboratory operates under a bipartite agreement between the
International Atomic Energy Agency and the Government of the Principality of Monaco. G.
Coteur is holder of a FRIA doctoral grant. Ph. Dubois and M. Warnau are Research
296

Effects of PCBs on ROS production by Paracentrotus lividus
Associates of the National Fund for Scientific Research (NFSR, Belgium). This research was
supported by a Belgian Federal Research Programme (SSTC, Contract MN/11/30) and by a
special funding of the National Bank of Belgium (BNB) and a NFSR (Belgium) short-term
fellowship to M. Warnau. Contribution of the “Centre Interuniversitaire de Biologie Marine”
(CIBIM).
297

298

APPENDIX I: CAPTIONS TO FIGURES
299
APPENDIX I : CAPTIONS TO FIGURES
Figure 1. Numbering system for sites of chlorine on a biphenyl molecule (Metcalfe 1994) .. 15
Figure 2. Molecular configuration of 2,3,7,8 TCDD and PCB 169 (Metcalfe 1994) ............. 18
Figure 3. Contaminants transfers between compartments in a coastal model (Moore et al.
2002) ........................................................................................................................... 20
Figure 4. Model describing the fate of lipophilic xenobiotics in organisms (Hodgson & Levi
1993) ........................................................................................................................... 22
Figure 5. Pathways for activation and detoxification of organic chemicals (Walker et al. 1996)
.................................................................................................................................... 23
Figure 6. Hypothesized induction mechanism of CYP1A (Bucheli & Fent 1995)................. 24
Figure 7. Diagram of the general water circulation in the North Sea (NSTF 1993) ............... 31
Figure 8. The common NE Atlantic sea star Asterias rubens L. (Hayward & Ryland 1996) . 34
Figure 9. Uptake kinetics of the sum of 10 PCB congeners (mean concentration in ng g –1 total
lipids) in bodywall and pyloric caeca of A. rubens exposed to spiked sediments........... 47
Figure 10. Uptake kinetics of 7 non coplanar PCB congeners (mean concentration in ng g –1
total lipids) in (A) bodywall and (B) pyloric caeca of A. rubens exposed to spiked
sediments..................................................................................................................... 48
Figure 11. Individual uptake kinetics of 3 c-PCB congeners (mean concentration in ng g –1
total lipids ± SD, n = 3) in (A) bodywall and (B) pyloric caeca of A. rubens exposed to
spiked sediments. ......................................................................................................... 48
Figure 12. Variation of the ratio between PCB 153 and c-PCBs 77, 126 and 169 in sediments
(yellow) and sea stars bodywall (orange) and pyloric caeca (green) at different times
during the exposure period. Time 0: background ratios (before spiking). To fit the figure,
PCB 153:169 ratio is divided by a factor 100 in sea stars bodywall, and by 1000 in
pyloric caeca. ............................................................................................................... 51
Figure 13. Comparison between c-PCB 77 (A), 126 (B) and 169 (C) concentrations (pg WHO
TEQ g -1 lipids; except PCB 169 in pyloric caeca: 100 pg WHO TEQ g -1 total lipids) in
bodywall (white bars) and pyloric caeca (grey bars), and ROS production by bacteriastimulated
amoebocytes (white dots; total chemiluminescence, RLU) of A. rubens
exposed to spiked sediments......................................................................................... 53
Figure 14. Comparison between ROS production by nonstimulated (black dots) and bacteriastimulated
(white dots) amoebocytes (total chemiluminescence, RLU) and c-PCB 77
(white squares) 126 (light grey squares) and 169 (dark grey squares) concentrations (ng
g -1 total lipids; except PCB 169 in pyloric caeca: pg g -1 total lipids) in (A) bodywall and
(B) pyloric caeca of A. rubens exposed to spiked sediments. ........................................ 54
Figure 15. Schematic representation of sample processing before ß-spectrometry analysis... 64
Figure 16. Asterias rubens-Seawater experiment. Uptake of 14 C-PCB 153 from seawater in
different body compartments of the sea star (mean concentration in ng g -1 total lipids ±
SD, n=3) ...................................................................................................................... 66
Figure 17. Asterias rubens-Sediment experiment. Uptake of 14 C-PCB 153 from sediments in
different body compartments of the sea star (mean concentration in ng g -1 total lipids ±
SD, n=3). ..................................................................................................................... 67
Figure 18. Schematic representation of sample processing before liquid scintillation counting
.................................................................................................................................... 77

300
APPENDIX I : CAPTIONS TO FIGURES
Figure 19. Asterias rubens. Uptake of 14 C-PCB 77 from seawater in different body
compartments of the sea star (mean concentration in ng g –1 total lipids ± SD, n = 3)..... 81
Figure 20. Linear regression between 14 C-PCB 77 concentrations (ng g –1 total lipids)
measured in oral and aboral body walls of sea stars exposed to contaminated seawater
for 15 d. r: correlation coefficient................................................................................. 83
Figure 21. Asterias rubens. Uptake of 14 C-PCB 77 from sediments in different body
compartments of the sea star (mean concentration in ng g –1 total lipids ± SD, n = 3)..... 84
Figure 22. Linear regression between 14 C-PCB 77 concentrations (ng g –1 total lipids)
measured in rectal and pyloric caeca of sea stars exposed for 34 d to contaminated
sediments. r: correlation coefficient.............................................................................. 85
Figure 23. Asterias rubens. Uptake of 14 C-PCB 77 from food (Mytilus galloprovincialis) in
different body compartments of the sea star (mean concentration in ng g –1 total lipids ±
SD, n = 3). ................................................................................................................... 86
Figure 24. Linear regression between 14 C-PCB 77 concentrations (ng g –1 total lipids)
measured in rectal and pyloric caeca of sea stars fed contaminated food for 34 d. r:
correlation coefficient. ................................................................................................. 86
Figure 25. Asterias rubens. ROS production (sum LCL; non-stimulated and bacteriastimulated
amoebocytes; mean±SD; n = 3) measured in sea stars exposed to 14 C-PCB
77 via (A) seawater, (B) sediments, or (C) food............................................................ 88
Figure 26 Asterias rubens. CYP1A IPP induction (ratio of experimental response to control
group; mean±SD; n = 3) measured using competitive ELISA in sea stars exposed to
14 C-PCB 77 via seawater, sediments or food................................................................. 89
Figure 27. Schematic representation of CYP1A IPP competitive ELISA method. .............. 100
Figure 28. Western blot SDS-PAGE gel. Lanes are: (A) molecular weight standards (kD), (B)
BNF-injected trout microsomes, (C) PCB 126-injected sea star PMS extract and (D)
seawater-injected sea star PMS extract....................................................................... 103
Figure 29 CYP1A IPP induction (induction fold, mean ± SD, n = 3) in sea stars injected with
coplanar (126) or non-coplanar (153) PCB................................................................. 104
Figure 30. CYP1A IPP induction (ratio of experimental response to control group) as a
function of injected PCB dose (ng g -1 FW; log scale) for two structurally contrasting
congeners. Curve fitting: linear regression of CYP1A induction as a function of injected
dose. .......................................................................................................................... 105
Figure 31. ROS production (total chemiluminescence) by non-stimulated amoebocytes as a
function of injected PCB dose (ng g -1 FW; log scale) for two structurally contrasting
congeners. Curve fitting: linear regression of ROS production as a function of injected
dose (excluding data of the highest dose); Baseline: ROS production value in control
group (n = 15). ........................................................................................................... 106
Figure 32. ROS production (total chemiluminescence) by bacteria-stimulated amoebocytes as
a function of injected PCB dose (ng g -1 FW; log scale) for two structurally contrasting
congeners. Curve fitting: linear regression of ROS production as a function of injected
dose (excluding data of the highest dose); Baseline: ROS production value in control
group (n = 15). ........................................................................................................... 107
Figure 33. Regressions between ROS production (total chemiluminescence) and CYP1A IPP
induction (ratio of experimental response to control group) in the case of non-stimulated
and bacteria-stimulated amoebocytes from PCB 77-injected sea stars. Curve fitting:
linear regression of CYP1A IPP induction as a function of ROS production; R 2 :
determination coefficient............................................................................................ 108
Figure 34. Sampling stations and transects along and off the Belgian coast........................ 118

301
APPENDIX I : CAPTIONS TO FIGURES
Figure 35 Correlation between ∑ 6PCBs (ng g -1 dry weight) and lipid content (g 100g -1 dry wt)
in four body compartments of the asteroid A. rubens for stations Nieuwpoort and 250.
.................................................................................................................................. 125
Figure 36. Correlation between concentration ratio (CR) of PCBs in the A. rubens pyloric
caeca (CR = ∑ 6 PCBs in asteroids -ng g -1 lipids-/ ∑ 6 PCBs in sediments -ng g -1 dry wt-)
and the percentage of the grain-size fraction 250-500µm in the sediments of the sampling
stations (%)................................................................................................................ 126
Figure 37. Asterias rubens. Concentrations (µg g -1 ; mean+SD; n=5) of heavy metals (Zn, Cu,
Cd, Pb) measured in the body wall of sea stars sampled in the southern North Sea.
Histogram bars sharing the same superscript do not differ significantly. Cu
concentrations do not differ significantly between stations. ........................................ 148
Figure 38. Asterias rubens. Concentrations (µg g -1 ; mean+SD; n=5) of heavy metals (Zn, Cu,
Cd, Pb) measured in the pyloric caeca of sea stars sampled in the southern North Sea.
Histogram bars sharing the same superscript do not differ significantly...................... 149
Figure 39. Asterias rubens. Concentration (ng g -1 total lipids; mean+SD; n=3) of PCB 153 and
the sum of 6 PCB congeners measured in the pyloric caeca and body wall of sea stars
sampled in the southern North Sea. Histogram bars sharing the same superscript do not
differ significantly...................................................................................................... 150
Figure 40. Asterias rubens. Linear regressions between PCB 153 and the other considered
PCB congeners (∑ 5PCB) concentrations (ng g -1 total lipids) measured in the body wall
(A) or pyloric caeca (B) of sea stars. .......................................................................... 151
Figure 41 Asterias rubens. ROS production (stimulated and non-stimulated; total
chemiluminescence, Relative Light Units (RLU) 10 -6 cells; mean+SD; n=5) measured in
sea stars sampled in the southern North Sea. Histogram bars sharing the same superscript
do not differ significantly........................................................................................... 152
Figure 42. Asterias rubens. CYP1A IPP induction (induction fold; mean+SD; n=5) measured
in sea stars sampled in the North Sea using competitive ELISA. Histogram bars sharing
the same superscript do not differ significantly........................................................... 152
Figure 43. Asterias rubens. Linear regressions between CYP1A induction (induction fold) and
PCB 153 or ∑ 6PCB concentrations (ng g -1 total lipids) measured in sea stars pyloric
caeca.......................................................................................................................... 153
Figure 44. Map of the southern North Sea showing the location of the sampling stations.
Arrows indicate the river mouths................................................................................ 163
Figure 45. Biological responses measured in Dutch and German stations. (a) Effects of
sediments on the early development of Asterias rubens and Psammechinus miliaris
(means + sd, n=6). (b) Immune responses of Asterias rubens: coelomic amoebocyte
concentration (CAC) and bacteria-stimulated amoebocyte ROS production (RLU:
Relative Light Units) (means + sd, n=10)................................................................... 174
Figure 46. Cluster tree of environmental and contamination variables. Variables showing
Pearson’s distance below 0.2 were grouped. Six groups were determined: metals levels
in the

302
APPENDIX I : CAPTIONS TO FIGURES
Figure 47. Relationships between biological, environmental and contamination variables. All
variables were used in a factor analysis using the principal component method. For
abbreviations, see legend of Fig. 3.............................................................................. 177
Figure 48. Contaminant levels (mean pg g -1 dry wt.; n=2) measured in sediments from the
different sampling stations. ........................................................................................ 191
Figure 49. Calculated ratios between the major congener of a given class of contaminant and
the sum of all the congeners of that given class, in sediments (white), mussel (light grey)
and sea stars (dark grey)............................................................................................. 194
Figure 50. Contribution to toxicity (%TEQ) of the different classes of contaminants (PCDDs,
PCDFs, cPCBs) in sediments (A), mussels (B) and sea stars (C) from the different
sampling stations........................................................................................................ 195
Figure 51. Contaminant levels (mean±SD, pg g -1 lipids) measured in mussels (Mytilusedulis)
from the different sampling stations. .......................................................................... 195
Figure 52. Contaminant levels (mean±SD, pg g -1 lipids), and CYP1A immunopositive protein
induction (mean±SD, n=9) measured in sea stars (Asteriasrubens) from the different
sampling stations........................................................................................................ 197
Figure 53. Bioconcentration factors (ratio between the mean concentration measured in sea
stars and the mean concentration measured in mussels) calculated for the different
congeners, in each sampling station............................................................................ 198
Figure 54. Regressions between mean CYP1A IPP induction (Time fold) and TEQs values of
the different DLC classes (pg TEQ g -1 lipids) determined in sea star pyloric caeca. R 2 :
corrected determination coefficient. ........................................................................... 199
Figure 55. Paracentrotus lividus. Uptake kinetics of 14 C-PCB 153 from seawater in 4 body
compartments of the sea urchin (mean concentration ng g -1 total lipids ± SD, n=3)..... 257
Figure 56. Paracentrotus lividus. Loss kinetics of 14 C-PCB 153 (mean concentration ng g -1
total lipids ± SD, n = 3) in three body compartments of the sea urchin after a single
feeding on radiolabelled food (Posidonia oceanica, (left) or Taonia atomaria (right)). 259
Figure 57. Diagrammatic representation of sample processing prior to liquid scintillation
counting..................................................................................................................... 267
Figure 58. Sepia officinalis. Uptake of 14 C-PCB 153 from seawater in the different body
compartments and in whole-body juvenile cuttlefish (mean ng g -1 total lipids ± SD; n =
3) ............................................................................................................................... 270
Figure 59. Sepia officinalis. 14 C-PCB 153 distribution (mean %) among the different body
compartments during the seawater experiment. .......................................................... 272
Figure 60. Sepia officinalis. Uptake of 14 C-PCB 153 from sediments in the different body
compartments and in whole-body juvenile cuttlefish (mean ng g -1 total lipids ± SD; n =
3) ............................................................................................................................... 273
Figure 61. Sepia officinalis. 14 C-PCB 153 distribution (mean %) among the different body
compartments during the sediment experiment........................................................... 273
Figure 62. Sepia officinalis. Uptake of 14 C-PCB 153 in the different body compartments and
in whole-body juvenile cuttlefish (mean ng g -1 total lipids ± SD; n = 3) following
ingestion of radiolabelled food (Artemia salina)......................................................... 274
Figure 63. Sepia officinalis. 14 C-PCB 153 distribution (mean %) among the different body
compartments during the food experiment.................................................................. 275
Figure 64. Excitation spectra of ethoxyresorufin (2 µM) and resorufin (2.5 nM) standards
from Molecular Probes® - Emission values at 584 nm are expressed as relative values
compared to the maximum of fluorescence intensity. Y-scales of both curves are
independent................................................................................................................ 283

303
APPENDIX I : CAPTIONS TO FIGURES
Figure 65. Emission spectra of ethoxyresorufin (2 µM) and resorufin (2.5 nM) standards from
Molecular Probes® - Excitation values at 535 nm are expressed as relative values
compared to the maximum of fluorescence intensity. Same Y-scales for both curves. 284
Figure 66. Emission spectra of ethoxyresorufin (2 µM) and resorufin (2.5 nM) standards from
Molecular Probes® - Excitation values at 560 nm are expressed as relative values
compared to the maximum of fluorescence intensity. Same Y-scales for both curves. 285
Figure 67. Stimulated vs. unstimulated ROS production by coelomocytes of P. lividus:
chemiluminescence (mean ± SE, n=4) of bacteria-stimulated (black squares) or
unstimulated (white dots) coelomocytes over time. .................................................... 292
Figure 68. ROS production (total chemiluminescence) by unstimulated coelomocytes of P.
lividus according to the PCB dose injected: PCB #77 (black dots), #126 (white squares),
#153 (asterisks) and #169 (black triangles). Controls were injected with seawater alone.
(Mean ± SE, n=4)....................................................................................................... 293
Figure 69. ROS production (total chemiluminescence) by stimulated coelomocytes of P.
lividus according to the PCB dose injected: PCB #77 (black dots), #126 (white squares),
#153 (asterisks) and #169 (black triangles). Controls were injected with seawater alone.
(Mean ± SE, n=4). (*) significantly different from the control; (¥) significant doseresponse
relationship.................................................................................................. 293
Figure 70. Time-to-peak of the ROS production kinetics of stimulated P. lividus
coelomocytes according to the PCB dose injected: PCB #77 (black dots), #126 (white
squares), #153 (asterisks) and #169 (black triangles). Controls were injected with
seawater alone. (Mean ± SE, n=4). (*) significantly different from the control. .......... 294

APPENDIX II: CAPTIONS TO TABLES
304
APPENDIX II : CAPTIONS TO TABLES
Table 1. Protective and non-protective responses to chemicals (Walker et al. 1996)............. 24
Table 2. Biomarkers at different organizational levels (Walker et al. 1996).......................... 28
Table 3. Sediments characteristics: total and organic carbon content (mg C g -1 ), and grainsize
distribution (%)..................................................................................................... 43
Table 4. Concentration of different PCB congeners in the sediments (mean concentration; ng
g -1 dry wt; n=3). ........................................................................................................... 49
Table 5. Parameters and statistics of the equations describing uptake kinetics of the noncoplanar
PCB congeners accumulated from sediments in two body compartments of
Asterias rubens. Model used: C(t)=C 0+C ss.(1-e -k.t ) where C(t) and C 0 are PCB
concentrations (ng g -1 lipids) respectively at time t (d) and at time 0 and C ss is the PCB
concentration incorporated at steady-state; k is the depuration rate constant (d -1 ); ASE is
the asymptotic standard error; C BKD is the background PCB concentrations, measured in
sea stars before starting the experiment; R 2
corr is the corrected determination coefficient.
.................................................................................................................................... 50
Table 6. PCB concentrations over time (d) (mean ± SD; ng g -1 lipids) in the bodywall (A) and
pyloric caeca (B) of the sea star.................................................................................... 50
Table 7. Characteristics of the background and added concentrations of PCB 153.
Background concentrations were measured in sediments, seawater and sea stars (body
wall and pyloric caeca) the day before starting the experiment; added concentrations
were measured in subsamples of sediments and seawater regularly collected in the
experimental microcosms throughout the experiment. Ranges of values of PCB 153
(unless specified) reported for sediments and seawater in the field are given for
comparison. ................................................................................................................. 62
Table 8. Asterias rubens. Parameters and statistics of the equations describing the uptake of
14 C-PCB #153 from seawater and sediments in the body compartments of the sea star. 67
Table 9 Asterias rubens. Concentration factors (CF; maximum, minimum and mean values) in
the body compartments of the sea star after 34 days of exposure via seawater (A).
Transfer factors (TF; maximum, minimum, and mean values) in the body compartments
of the sea star after 34 days of exposure via sediments (B). CFs are calculated as the ratio
between PCB 153 concentration in the sea star body compartments (ng g -1 total lipids)
and its concentration in seawater (ng g -1 ). TFs are calculated as the ratio between PCB
153 concentration in the sea star body compartments (ng g -1 total lipids) and its
concentration in sediments (ng g -1 dry wt). ................................................................... 68
Table 10 Asterias rubens. PCB distribution (mean % ± SD, n = 3) in the different body
compartments of the sea star after 34 days of exposure via seawater or sediments........ 68
Table 11 Asterias rubens. Comparisons among PCB #153 concentrations obtained in the
present study (background + incorporated concentrations) and previous field studies in
the North Sea. .............................................................................................................. 70
Table 12 Asterias rubens. Parameters and statistics of the equations describing the uptake of
14 C-PCB #77 in different body compartments of the sea star. L (linear model): C(t)=kt;
S (saturation model): C(t)=Css.(1-e -kt ); where C(t) and Css: 14 C-PCB #77 concentrations

305
APPENDIX II : CAPTIONS TO TABLES
(ng g -1 lipids) respectively at time t (d) and at steady-state; k: rate constant (d -1 ); ASE:
asymptotic standard error; R 2 : corrected determination coefficient. .............................. 81
Table 13 Asterias rubens. Concentration and Transfer factors, CF and TF (mean ± SD; n=24
for seawater exposure and n=33 for sediments and food exposures) in body
compartments at the end of exposure periods via seawater, sediments or food. CFs
calculated as ratio between PCB 77 concentration in body compartments (ng g –1 total
lipids) and its concentration in seawater (ng g -1 ). TFs calculated as ratio between PCB 77
concentration in body compartments (ng g –1 total lipids) and its concentration in
sediments (ng g -1 dry wt) or in food (ng g -1 total lipids). BW: bodywall, Pyl. Caec.:
pyloric caeca, Rect. Caec.: rectal caeca, C.D.S.: central digestive system ..................... 82
Table 14 Asterias rubens. Parameters and statistics of the equations describing the induction
of CYP1A immunopositive protein during the exposure experiments. Kinetics were
described using a saturation model: C(t)=Css(1-e -lt ) where l is the rate constant (d -1 ).
Other symbols as in Table 12. ...................................................................................... 89
Table 15. Optical densities measured in a representative ELISA (mean ± SD). Ab = rabbit
antitrout CYP1A antibody; SW = seawater injected sea stars (100 µg ml -1 ); AC =
acetone injected sea stars (100 µg ml -1 ); Tµ = trout microsomes (100 µg ml -1 ); B * Tµ =
biotinylated trout microsomes; Av-HRP = avidine-coupled horse radish peroxidase;
O.D. = optical density (mean ± SD, biological triplicates); - = no sample/treatment .. 104
Table 16. Positions and characteristics of the sampling stations ......................................... 118
Table 17. Certified and measured PCB concentrations (ng g -1 dry wt ± sd, n = 4) in a certified
material reference (sediments from the harbour “Nova Scotian” in East of Canada) ... 120
Table 18. Certified (C; mean value ± 95% CI) and measured (M; min and max values, n = 6)
metal concentrations (mg g -1 dry wt) of certified reference material (Mytilus edulis
tissues, CRM n°278; BCR). ....................................................................................... 121
Table 19. PCB concentrations (mean ± sd; mg g -1 dry wt, n = 6) in the bulk fraction of the
sediments. Superscripts indicate ranking of stations according to decreasing
concentrations of a given PCB congener (a>b>c>…). Stations with concentrations
sharing a common superscript are not significantly different from each other (multiple
comparison test of Tukey; a = 0.05). nm = not measured ........................................... 122
Table 20. PCB concentrations (mean ± sd; ng g -1 total lipids, n = 3 pools of 3) in the different
body compartments of the asteroid Asterias rubens. Superscripts indicate ranking of
stations according to decreasing concentrations of a given congener (a>b>c>…).
Stations with concentrations sharing a common superscript are not significantly different
from each other (multiple comparison test of Tukey; a = 0.05). nm = not measured... 123
Table 21. Grain-size distribution (mean % ± sd, n = 6) in the dried sediments for the 19
sampling stations. Predominant grain-size fractions are indicated in bold................... 126
Table 22 Metal concentrations (mean ± sd; mg g -1 dry wt, n = 6) in the different grain-size
fractions of the sediments. Superscripts indicate ranking of stations according to
decreasing concentrations of a given metal (a>b>c>…). Stations with concentrations
sharing a common superscript are not significantly different from each other (multiple
comparison test of Tukey; a = 0.05)........................................................................... 127
Table 23 Correlation coefficients (r) between metal concentrations measured in the different
grain-size fractions of the sediments (pcorrelation always < 0.0001 except when ns -non
significant correlation- is indicated) ........................................................................... 129
Table 24. Variability (%) in metal concentrations measured in sediments explained by the
factors considered (station, grain-size) and their interaction........................................ 130
Table 25 Metal concentrations (mean ± sd; mg g -1 dry wt, n = 5 pools of 3) in the different
body compartments of the sea star Asterias rubens. Superscripts indicate ranking of
stations according to decreasing concentrations of a given metal (a>b>c>…). Stations

306
APPENDIX II : CAPTIONS TO TABLES
with concentrations sharing a common superscript are not significantly different from
each other (multiple comparison test of Tukey; a = 0.05)........................................... 132
Table 26. Correlation coefficients (r) between metal concentrations measured in the different
sea star body compartments (p correlation always

307
APPENDIX II : CAPTIONS TO TABLES
oceanica shoots, and Taonia atomaria thallia regularly collected in the experimental
microcosms throughout the experiment. ..................................................................... 255
Table 43. Paracentrotus lividus. Parameters and statistics of the equation fitting the uptake of
14 C-PCB 153 in the body compartments of echinoids exposed via seawater: C(t) = A e k.t .
C(t): 14 C-PCB 153 concentration (ng g -1 lipids) at time t (d); A: ordinate at the origin (ng
g -1 lipids); k: rate constant (d -1 ); ASE: asymptotic standard error; R 2 : determination
coefficient.................................................................................................................. 257
Table 44. Paracentrotus lividus. Concentrations and distribution of PCB 153 incorporated in
the different body compartments of the echinoids exposed to the congener via seawater.
A. PCB 153 concentrations (mean ng g -1 lipids ± SD, n = 3). Mean concentrations
sharing the same superscript do not differ significantly between each other. B. PCB 153
distribution (mean % ± SD, n = 3) among body compartments................................... 258
Table 45. Paracentrotus lividus. PCB concentrations (ng g -1 lipids; mean ± SD, n = 3)
measured in the sea urchin body compartments after a single feeding, using two different
food (Posidonia oceanica vs Taonia atomaria). Mean concentrations sharing the same
superscript do not differ significantly among each other (p Tukey test > 0.05). .................. 259
Table 46. Paracentrotus lividus. Parameters and statistics of the equation describing the loss
of PCB #153 from the sea urchin body compartments after a single feeding on Posidonia
oceanica and Taonia atomaria. Equation is C(t) = C(0) e -k t ; where C(t) and C(0) are 14 C-
PCB 153 concentrations (ng g -1 lipids) at time t (d) and time 0, respectively, and k is the
rate constant (d -1 ). ASE: asymptotic standard error; R 2 : corrected determination
coefficient; Tb 1/2: Biological half-life (d).................................................................... 260
Table 47. Sepia officinalis. Parameters and statistics of the equations describing the uptake of
14 C-PCB #153 in different body compartments of the cuttlefish exposed via seawater,
sediments and food..................................................................................................... 270
Table 48. Sepia officinalis. Concentration factors and transfer factors (CF, TF; mean ± SD; n
= 3) in the body compartments and in whole-body juvenile cuttlefish at the end of the
experimental exposures.............................................................................................. 271
Table 49. Summary of the observed effects of different PCB congeners on ROS production by
Paracentrotus lividus coelomocytes. Legend: +, significant stimulation (p < 0.05); (+),
significant stimulation (0.05 < p < 0.1); X, no significant effect (p > 0.1); hc, effect
observed at high concentration; lc, effect observed at low concentration; dr, significant
dose-response relationship.......................................................................................... 294

APPENDIX III: CAPTIONS TO EQUATIONS
308
APPENDIX III : CAPTIONS TO EQUATIONS
Equation 1: C t = C 0 + C ss (1 - e -kt ),........................................................................................ 46
Equation 2: C(t) = Css (1-e -k.t ) .............................................................................................. 65
Equation 3: C(t) = C 0 e k.t ...................................................................................................... 65
Equation 4: Ct = Css (1-e -k.t ) / 1+e -k.(t-I) .................................................................................. 65
Equation 5: C(t) = Css (1-e -k e .t ) ............................................................................................. 80
Equation 6: C(t) = k u.t.......................................................................................................... 80
Equation 7: y = a . e b.CYPi .................................................................................................... 102
Equation 8: CYP i = 100 – (100 . A/A max)............................................................................ 102
Equation 9: CYP I = (a . e b.CYPi )/(CYP ic)............................................................................... 102
Equation 10: C(t) = A e k t ................................................................................................... 256
Equation 11: C(t) = C(0) e -k t .............................................................................................. 256
Equation 12: C(t) = C o +k.t ................................................................................................ 269
Equation 13: C(t) = C ss (1-e -k.t ) ........................................................................................... 269
Equation 14: C(t) = C ss (1-e -k.t ) / 1+e -k.(t-I) ............................................................................. 269