Marine Pollution Bulletin xxx (2009) xxx–xxx

ARTICLE IN PRESS

Contents lists available at ScienceDirect

Marine Pollution Bulletin

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

Enzymatic and cellular responses in relation to body burden of PAHs in bivalvemolluscs: A case study with chronic levels of North Sea and Barents Sea dispersed oil

T. Baussant a,*, R.K. Bechmann a, I.C. Taban a, B.K. Larsen a,1, A.H. Tandberg a, A. Bjørnstad a, S. Torgrimsen a,A. N�vdal a, K.B. Øys�d a, G. Jonsson a,2, S. Sanni a

a International Research Institute of Stavanger/IRIS-Biomiljø, Mekjarvik 12, 4070 Randaberg, Norway

a r t i c l e i n f o

Keywords:Dispersed oilMytilus edulisChlamys islandicaNorth SeaBarents SeaBiomonitoringBiomarkers

0025-326X/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.marpolbul.2009.08.007

* Corresponding author. Tel.: +47 51875501; fax: +E-mail address: thierry.baussant@iris.no (T. Baussa

1 Present address: DTU-Aqua, Nordsøen Forskerpark2 Present address: Stavanger University Hospital, Sta

Please cite this article in press as: Baussant, T., ewith chronic levels of North Sea and Barents Se

a b s t r a c t

Mytilus edulis and Chlamys islandica were exposed to nominal dispersed crude oil concentrations in therange 0.015–0.25 mg/l for one month. Five biomarkers (enzymatic and cellular responses) were analysedtogether with bioaccumulation of PAHs at the end of exposure. In both species, PAH tissue residuesreflected the exposure concentration measured in the water and lipophylicity determined the bioaccu-mulation levels. Oil caused biomarker responses in both species but more significant alterations inexposed C. islandica were observed. The relationships between exposure levels and enzymatic responseswere apparently complex. The integrated biomarker response related against the exposure levels was U-shaped in both species and no correlation with total PAH body burden was found. For the monitoring ofchronic offshore discharges, dose- and time-related events should be evaluated in the selection of bio-markers to apply. From this study, cellular damages appear more fitted than enzymatic responses, tran-sient and more complex to interpret.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

An ongoing concern related to the development of offshore oiland gas operations in the North Sea is that long-term exposureto operational drilling and waste water may cause detrimental ef-fects to marine organisms. Currently, the amount of producedwater, the main waste water effluent, is increasing because ofaging of the fields and prospects predict a continuation of the in-crease in the coming years. There are reasons for environmentalconcern regarding a number of contaminants discharged in theproduced water effluents of the North Sea. These effluents are cur-rently being monitored for contaminants input and biological ef-fects (Hylland et al., 2008).

Offshore activities are also moving towards more remote sub-Arctic marine environments with unique aspects like the markedseasonal variation in light, colder temperature and the presenceof sea ice at some times of the year. These cold environments hosta wide variety of marine life, including some species which areimportant economical resources. In the Barents Sea, there is astrong official requirement to prevent long term environmentalimpacts. Presently, few data are available on chronic toxic effectsof offshore discharges in this relatively cold marine environment

ll rights reserved.

47 51875540.nt).

, Hirtshals, Denmark.vanger, Norway.

t al. Enzymatic and cellular respa dispersed oil. Mar. Pollut. Bu

and a zero discharge policy is applied in a conservative way. Forthe Barents Sea ecosystem, deficiency in sub-sea pipelines can leadto release of undetected chronic level of oil at times but the majorthreat remains related to massive discharges of oil contaminants inthe course of production, processing or transport.

A relevant issue to address is whether the monitoring tools andrisk assessment that currently prevail in the temperate regions canbe applied to determine the sensitivity and predict effects on Arcticmarine organisms. This can be appraised by understanding howbiological responses are modified under the conditions wherethese organisms are living. In the past, monitoring was basedmainly on analyses of chemicals like polycyclic aromatic hydrocar-bons (PAH) and alkylphenols, two classes of toxic pollutants foundin waste water. Determining when contamination has resulted inadverse effects in the natural environment requires a combinationof chemical and biological measurements. The need to use inte-grated measurements of actual effects on biota has been strength-ened during the last decades. Environmental monitoringprogrammes including both chemical analyses of seawater for con-taminant levels and the measurements of a battery of so-calledbiomarkers are presently recommended (Cajaraville et al., 2000;Galloway et al., 2004; Moore et al., 2004). Although the use of bio-markers for ecological risk assessment is still debated (Forbes et al.,2006), biomarkers are presently becoming part of the healthassessment and management of aquatic ecosystems in additionto the more traditional water chemical analyses (Hagger et al.,2008; Allan et al., 2006). Discussions within ICES/OSPAR are also

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

2 T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx

ARTICLE IN PRESS

ongoing to determine protocols of application and select whichbiological markers to use for different biota (ICES, 2008a). Theserecommendations serve as a good basis on how water columnmonitoring surveys around platforms are applied (Gorbi et al.,2008; Hylland et al., 2008) and more generally for the environmen-tal management of seawater where offshore industries are present.

In the Biosea I JIP program, a biomarker-based approach hasbeen chosen. Laboratory-controlled experiments were performedto implement and evaluate a suite of relatively well establishedbiomarkers on several marine organisms for their primary use inoffshore oil and gas monitoring programmes in the North Seaand in the Barents Sea. Long-term exposures to several concentra-tions of dispersed crude oil were conducted with different species.Fish and invertebrates (echinoderms, crustaceans and bivalve mol-luscs) were selected. Here, we present the data obtained withmolluscs.

Although bivalve molluscs are more associated with benthos,they are extensively used in biomonitoring programmes and forexample deployed in caging experiments around platforms tomonitor both chemical concentrations and biological responses(Hylland et al., 2008). Their sessile, filter feeding lifestyle and lowenzymatic degradation rate render them capable of retaining rela-tively higher levels of organic molecules compared to other organ-isms (Meador et al., 1995).

Two mollusc species were used in this study. For the North Sea,we used the blue mussel Mytilus edulis and the Icelandic scallopChlamys islandica was selected for the Barents Sea. Blue musselshave been largely deployed for monitoring contaminant concentra-tions in the water as well as to evaluate a number of selected bio-markers in the laboratory or in real field conditions (Bocqueneet al., 2004; Bodin et al., 2004; Orbea et al., 2002; Porte et al.,2001; Regoli et al., 2002; Aarab et al., 2004). Ecotoxicological stud-ies with Icelandic scallop are relatively scarce and few report onbiomarkers. However, some authors have used C. islandica to studybiological responses in conditions where this species lives (Camuset al., 2002; Regoli et al., 2000); due to their relatively wide distri-bution in Arctic waters, Icelandic scallops represents a promisingsubstitute sentinel species for the northern marine areas wheremussels are not present.

One mechanism of toxicity after exposure to PAH appears to re-late to the production of reactive oxygen species (ROS) as by-prod-ucts of metabolism (Digiulio et al., 1989; Livingstone, 2003). Thesehighly potent molecules can react with important macromoleculeslike DNA, proteins and lipids. The consequences of ROS for cellcomponents and their functioning can be at times critical for indi-viduals. Aerobic cells have developed a variety of defences to pro-tect themselves against these oxyradical products. For example,several enzymes exist that can counteract the pro-oxidant chal-lenge, but overwhelming is also possible (Regoli et al., 2004). Toevaluate the mussels’ antioxidant status, common biomarkers in-clude the level of glutathione, gluathione-S-transferase, superoxidedismutase, catalase, glutathione peroxidase and glutathione reduc-tase. Here, we selected two of these biomarkers to estimatechanges in the redox status: gluthathione-S-transferase (GST) andcatalase (CAT). Also, the total oxidative scavenging capacity (TOSC)was used as an overall redox status parameter.

Catalase is considered a true oxidative scavenger enzyme whichremoves the potential damage of hydrogen peroxide by catalysingthe reaction 2H2O2 ? 2H2O + O2 (Digiulio et al., 1989; Orbea et al.,2002) while GST is more involved in the phase II detoxificationprocess by adding gluthathione to reactive metabolite compoundsthereby facilitating their excretion. GST contributes significantly inthe defence against oxidative damage of DNA and lipids, and isused as a biomarker of organic contamination in bivalves (Bocqu-ene et al., 2004; Devier et al., 2005; Moreira and Guilhermino,2005). Compared to individual antioxidant enzymes, the TOSC as-

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

say can readily be related to different forms of ROS and has a betterpredictive value to evaluate the overall redox status (Regoli et al.,1998, 2002). The changes in TOSC have been associated with dam-ages at the cell level such as DNA damage and lysosome membranestability in fish and in mussels (Regoli et al., 2003, 2004).

At the cellular level, the lysosomal membrane stability assayedby the neutral red retention time (NRRT) is easy to perform on hae-mocytes of mussels. It is used as a general low-cost marker to de-tect impairments of the functional integrity of cells (Livingstoneet al., 2000). Lysosomes are cellular organelles and in mussels, theyplay an important role in several biological pathways, includingthe immunological defence system (Moore and Willows, 1998).Furthermore, there appears to be a good link between the impair-ment of lysosomal membrane, oxidative stress, genotoxicity andfitness parameters (Moore et al., 2004). The comet assay is a rela-tively simple and sensitive technique for determining DNA strandbreaks in individual cells. The assay is used as a marker of geneticimpairment and has been used to assess DNA damage in variousmarine organisms exposed to various types of environmental tox-icants (Lee and Steinert, 2003; Mamaca et al., 2005; Mitchelmoreand Chipman, 1998; Perez-Cadahia et al., 2004).

In this paper, the individual responses of each of the above-mentioned biomarkers were first investigated in the two molluscspecies and related to several exposure concentrations of crudedispersed oil. Statistically significant responses compared to con-trol individuals were analysed. Secondly, multivariate analysiswas applied as a way to integrate all biomarker responses into aoverall ‘‘health” assessment index and to examine the combinedbiomarker patterns for the two species.

The aims of this study were: (i) to examine the biomarker re-sponses obtained in M. edulis and C. islandica using appropriate lab-oratory conditions for the two species and (ii) to conclude on thesuitability of the selected biomarkers for environmental monitor-ing in marine regions where oil and gas activities exist or are pros-pected. The two species were exposed to dispersed crude oiloriginating from the North Sea (M. edulis) or the Barents Sea(C. islandica).

2. Materials and methods

2.1. Origin of the organisms

Blue mussels Mytilus edulis were collected in December 2002 ata farm located close to the Lysefjord (Rogaland county, Norway)and purchased at Aspøy Skjell og Produktutvikling AS (Hundvåg,Norway). At the research centre facility (Akvamiljø AS, Randaberg,Norway), they were transferred to a 350 l tank with running 34‰

seawater at a temperature of 7 ± 0.5 �C. Mussels were used in theexperimental system after ca. 2 weeks. No mortality was recorded.

The Icelandic scallops C. islandica were collected by a bottomdragger during a cruise with R/V Jan Mayen around Spitzbergencarried out in September 2003. After sampling, they were kepton board the ship with running sub-surface seawater. Thereafter,they were shipped by plane to our facility and immediately trans-ferred to a tank filled with 200 l running seawater at 4 ± 0.5 �C. C.islandica were kept �4 weeks before they were used in the exper-iment. During this quarantine period the mortality was 13% ofwhich the largest proportion died during the week followingarrival.

The seawater was pumped directly from 75 m depth in the Byfj-ord close to our facility and sand-filtered before use in the experi-mental system. Both molluscs were fed regularly a mixture of liveIsochrysis galbana, Rhodomonas baltica and Skeletonema costatumduring the maintenance period.

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx 3

ARTICLE IN PRESS

2.2. Experimental system

2.2.1. Dispersion of oil in seawaterThe continuous flow system described in Sanni et al. (1998) was

used with some modifications to create a dispersion of oil in sea-water. Crude oil was placed at the top of a two-compartment glasscylinder separated by a Teflon� piston. At the bottom of the cylin-der, distilled water was used to push the piston upwards at the de-sired flow and oil was then evacuated at the top of the cylinder atthe same flow to the stream of seawater. The dispersion wasmechanically created using a mixing valve placed upstream fromthe oil injection point. The valve was adjusted to a position that in-creased the pressure in the seawater pipe to six bars, allowing thefragmentation of oil to droplets of a mean size of 10 lm.

A dispersion equivalent to a nominal concentration of 5 mg l�1

was first made by injection of crude oil in the flow of seawater(7 l min�1) using a precision syringe pump (ISCO model 260D,USA). Seawater containing oil droplets was then conveyed to a20 l header tank. Peristaltic pumps (Watson-Marlow model 205and 505, UK) were then used to dilute it with adequate volumesof seawater into 10 l Duran glass flasks connected to the exposuretanks in order to obtain the desired nominal exposure concentra-tions i.e. 0.25 (high), 0.06 (medium) and 0.015 (low) mg l�1. Thesame concentrations of dispersed oil were used for both speciesof molluscs in two experiments.

2.2.2. Experiment with North Sea crude oilApproximately 500 mussels were placed in 60 l tanks with

flow-through for both the control and exposed groups. A NorthSea crude oil was used for the exposed groups and the exposurelasted for a period of 7 months after which the remaining musselswere used to perform an embryo-larval assay. Sampling of musselsfor chemistry and biological measurements was done after1 month and 7 months. However, here we will present only the re-sults obtained after 1 month exposure. Temperature in the tanksduring that period was 7.5 ± 0.5 �C.

2.2.3. Experiment with Barents Sea crude oilBetween 30 and 40 scallops were placed in 140 l tanks isolated

with polystyrene plates on the outside to maintain a stable coolingof the inlet seawater. The oil used originated from the Barents Sea.The exposure was carried out for 1 month and sampling took placethereafter with no further exposure as done for mussels. The mea-sured temperature in the tanks with scallops was 5 ± 0.5 �C.

In both experiments, control and exposed individuals of the twospecies were fed the same mixture of microalgae used in the main-tenance period.

2.3. Chemical analysis

PAH analyses of seawater and biota samples were performed byGas Chromatography (HP5890, Hewlett Packard, USA) connectedto a Mass Spectrometer (Finnigan SSQ7000, USA) and analysed inselected ion mode (GC/MS-SIM) as described previously in Baus-sant et al. (2001) and Jonsson et al. (2004). The analyses includedthe 16 PAHs recommended by the EPA and the alkylated congeners(C1 to C2/C3) of naphthalene, phenanthrene, chrysene and dibenz-othiphene. These were divided into three classes according to theiroctanol water partition coefficient (log Kow) to facilitate the resultdescription (Table 1). The quantification limit for each PAH mea-sured in seawater and in biota was based on EPA method 610and the method from Douglas et al. (1994). For seawater, this ran-ged between 0.005 and 0.01 lg/l (uncertainty ± 20%). In biota, thiswas between 1 and 2 lg/kg (uncertainty ± 20%).

A GC–MS analysis of PAH composition of the two crude oils wasmade prior to the experiments. There was relatively more of the

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

light molecular weight PAHs in the Barents Sea oil and more ofthe heavy molecular weight PAHs in the North Sea oil. In theBarents Sea oil, we found 92%, 5% and 0.2%, respectively, for the2-ring, 3-ring and 4-ring PAHs. For the North Sea oil we found88%, 8% and 0.8%, respectively, for the 2-ring, 3-ring and 4-ringPAHs. Larger PAHs than 4-ring were also detected in the two oilsbut none were quantifiable. Overall, the sum of PAH represented1.8% of the Barents Sea oil and 1.6% of the North Sea oil.

Seawater samples were collected in 1 l Duran glass bottles con-taining hydrochloric acid to ensure low pH (<2) and prepared foranalysis within 48 h after sampling.

PAH content in whole organisms was analysed to determinePAH body burden in both species and at the three exposure con-centrations. The whole soft tissue was dissected, excess waterwas wiped off on a paper towel and then it was transferred intoa glass vial (previously heated to 500 �C) with Teflon seal. Sampleswere stored at �80 �C until analysis. For blue mussels, we used 2–3pooled individuals per replicate and three replicates per exposuregroups. For Icelandic scallop, only one individual was used for eachreplicate and again three replicates in each group were analysed.

Bioaccumulation factors (BAFs) at the end of the exposure per-iod were estimated based on the PAH measured by GC–MS in biotaand in seawater as:

BAF ¼ PAHbiotaPAHseawater

:

2.4. Enzymatic and cellular responses

Table 2 shows the biomarkers selected in this study for the twospecies.

2.4.1. Enzymatic-level responsesGluthathione-S-transferase (GST), catalase and total oxidative

scavenging capacity (TOSC) towards peroxyl radicals were mea-sured in the cytosolic fraction of frozen digestive glands after sam-ple preparation. Results were normalised to the concentration oftotal protein determined by the Bradford absorption spectroscopyassay (Bradford, 1976). For these parameters, 10 individuals of M.edulis were used for each exposure group and the control. For C.islandica, we used fifteen individuals per treatment group.

2.4.1.1. Gluthathione-S-transferase. The thawed cytosol was diluted(1:5) with an appropriate amount of cold phosphate buffer(100 mM, pH 7.4) prior to analysis. Total GST activity was deter-mined in diluted samples using 20 mM of the artificial substrate1-chloro-2,4-dinitrobenzene (CDNB), which is conjugated by mostGST isoforms. GST was adapted to a Tomtec Quadra 96-320 liquidhandling system to dispense the different solutions into 96-wellmicro-plates. The increase in absorbance due to the formation ofthe conjugate product GST-CDNB was recorded on a microplatereader (Multiskan model 354, LabSystem) with absorbance at340 nm for 1 min. To calculate GST activity, the blank value wasdeducted from the sample readings and a molar extinction coeffi-cient (2) for glutathione-CDNB of 9.6 mM�1 cm�1 was applied:

GST activityðlmol=min=mg proteinsÞ

¼ DDO=minðenzymaticÞ9:6� ½proteins in the cell� ;

where DDO/min is the total DDO/min in the assay cell minus theDDO/min due to the spontaneous conjugation of the substrate(blank without sample).

2.4.1.2. Catalase. The catalase protocol used was based on themethod of Claiborne (1985). After dilution of the cytosol (1:5), a

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

Table 1List of molecules analysed by GC–MS in this study and their abbreviations.

Class Compound Abbreviation

logKow 3–4 Naphthalene naCl-naphthalene c1-naC2-naphthalene c2-na

log Kow 4-5 C3-naphthalene c3-naAcenaphthylene acyAcenaphthene aceFluorene flPhenanthrene phAnthracene anDibenzothiophene dbtC1-dibenzothiophene c1-dbt

logKow >5 C1-phen/anthr c1-phC2-phen/anthr c2-phC2-dibenzothiophene c2-dbtFluoranthene fluoPyrene pyBenzo(a)anthracene baaChrysene chC1-chrysene c1-chC2-chrysene c2-chBenzo(b)fluoranthene bbfBenzo(k)fluoranthene bkfBenzo(a)pyrene bapIndeno(1.2.3.cd)pyrene ipBenzo(g.h.i)perylene bghipDibenzo(a.h)anthracene daha

4 T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx

ARTICLE IN PRESS

reaction mixture consisting of Phosphate buffer at 100 mM andH2O2 (30%) at 500 mM were added to a quartz cuvette and the in-crease in absorbance was measured immediately at 240 nm for30 s on a Perkin-Elmer Model Lambda 2S UV/VIS spectrometer.

The catalase activities were defined in terms of moles of H2O2

consumed per minute per mg protein sample using a molar extinc-tion coefficient (2) for H2O2 0.04 mM�1 cm�1.

Activity of catalase ðlmol=min=mg proteinsÞ

¼ DOD=min0:040 �mg=ml protein in the cuvette

where 0.040 mM�1 cm�1 is a constant value for the molar extinc-tion coefficient (e) for H2O2 and DOD = (Max OD sample �Min ODsample).

2.4.1.3. Total oxidative scavenging capacity. The principle of theTOSC-assay is based on the measurement of the efficiency of se-lected biological sample preparations to scavenge specific ROS.Here, we used a method to assess the scavenging capacity of per-oxyl radicals. The method used was based in Winston et al.(1996) and Regoli and Winston (1999). Buffer solutions were ad-justed for marine invertebrates. Peroxyl radicals can oxidize thesubstrate a-keto-c-methiolbutiryc acid (KMBA) to ethylene gaswhich is measured by gas chromatography. Reactions were carriedout in 10 ml rubber septa sealed vials in a final volume of 1 ml. Eth-ylene production was measured by gas-chromatographic analysisof 200 ll taken from the head space of the reaction vials. Ethylene

Table 2List of biological markers used in M. edulis and C. islandica.

Biological markers Abbr.

Glutathione-S-transferase GSTCatalase CATTotal oxidative scavenging capacity TOSCLysosomal membrane stability NRRTComet assay COM

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

formation was monitored for 96 min with a Hewlett Packard (HP5890 series II) gas chromatograph equipped with a supelco SPB-1capillary column (30 m � 0.32 mm � 0.25 lm) and a flame ioniza-tion detector (FID).

2.4.2. Cellular-level responsesThe neutral red retention time (NRRT) in lysosomes and the

DNA comet assay were used to assess damages at the cellular lev-els. Both NRRT and comet were determined on hemolymph sam-ples collected from the anterior muscle in M. edulis and theadductor muscle in C. islandica. A 1 ml syringe containing physio-logical saline solution was used to obtain a 1:1 saline:haemolymphsample. For both species, the NRRT assay and the comet assay wereperformed on fresh biological material using, respectively, 10 and 9individuals per group.

2.4.2.1. Neutral red retention time. Impairments of lysosomal mem-branes were assayed by the NRRT method based on that from Loweand Pipe (1994). The principle of this assay is based on the fact thatlysosomes in healthy cells can retain neutral red after its initial up-take for longer periods than cells exhibiting stress. Sub-cellularretention of neutral red is assessed at timed intervals (up to180 min) using light microscopy. Slides with 30–50 ll of cell sus-pension were prepared. The test for each slide was terminatedwhen dye loss was evident in 50% of the granular haemocytes.

2.4.2.2. Comet assay. The comet assay was used to examine DNA al-kali–labile strand breaks at the individual cell level after themethod by Singh et al. (1998) and as described in Taban et al.(2004) and Mamaca et al. (2005). Due to preparation and analyticaltime for this assay, one control group and one exposed group wereanalysed each day over three consecutive days following sampling.DNA strand breaks were measured in 50 cells analysed randomlyfor each individual and the percentage DNA in the ‘‘comet tail”was scored using an image analysis package (Kinetic Imaging soft-ware, Komet version 4.0, Liverpool, UK).

2.5. Analytical methods of biomarkers

Differences from the control were tested with a non-paramatricKruskal–Wallis one-way analysis of variance. The level of statisti-cal significance for rejection of H0 ‘‘No difference” was set atp < 0.05.

Also, multivariate analysis was used to assess the integrated re-sponse based on biomarker data obtained from each species. Datawere analysed using a graphical method described in Beliaeff andBurgeot (2002) to combine the different biomarkers into an inte-grated biomarker response (IBR) calculated as:

IBR ¼Xn

i¼1

Ai;

where Ai is the area represented by biomarker i on a star plotgraphic.

Additionally, the standardized data (scores used to computeIBR) of biomarkers obtained in each treatment were used in a prin-

Tissue or cell type Unit

Digestive gland nmol/min/mg proteinDigestive gland lmol/min/mg proteinDigestive gland Tosc/mg proteinHaemocyte minHaemocyte %DNA in comet tail

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx 5

ARTICLE IN PRESS

cipal component analysis (PCA) to compare the combination ofbiomarkers characterizing best each exposure concentration andevaluate the differences or similarities in the two species. ThePCA results were visualized by two-dimensional plots of the firstthree components.

The IBR calculations were performed with Excel software(Microsoft, WA, USA). The Kruskal–Wallis test and the PCA analysiswere performed using JMP software (Windows, ver. 5, SAS institute,Cary, NC, USA).

3. Results

3.1. Chemical analyses

3.1.1. SeawaterFor the two experiments, each with different crude oil, we

found the sum of PAHs ranging from 0.2 (low exposure) to 4 (highexposure) lg/l. At the low and high exposure concentrations, thesum of PAHs in the two experiments was similar. A larger differ-ence was obtained at the medium concentration where the sumof PAHs in the North Sea oil experiment was half that analysedin the Barents Sea oil experiment (Table 3). As in crude oil, PAHswith log Kow 3–4 (naphthalene and alkylated forms) representedthe major fraction (60%) in the exposure seawater in the twoexperiments (Table 4). PAHs larger than 4-aromatic rings were de-tected but a quantification of these PAHs in seawater was not pos-sible. Also, the composition of the quantifiable PAHs analysed inseawater was very similar in the two experiments (Fig. 1). Overall,the seawater PAH analyses indicated quite comparable exposureconditions in the two experiments.

3.1.2. BiotaThe percentage of dominant PAHs in body tissues of both spe-

cies was different to that analysed in seawater (see Table 4). PAHswith log Kow 4–5 represented 50%. Some of the larger PAHs with 5-to 6-aromatic rings like benzo[a]pyrene, benzo[g,h,i]perylene anddibenzo[a,h]anthracene were found. PAHs with log Kow > 5 repre-sented between 20% and 30% of the total PAHs. This observation re-vealed the presence of these large molecular weight PAHs inseawater, although under the detection limit, hence their accumu-lation to some extent by the organisms where they were detectedand quantified. The pattern of individual PAHs in the two organ-isms was very similar (Fig. 2). Bioaccumulation factors estimatedafter one month in M. edulis and C. islandica were also similar (Ta-ble 5). Based on the sum of PAHs in biota and the sum of PAHs inseawater, we found BAF (wet weight) values close to 3000 in M.edulis. BAF in the low and medium exposure concentrations in C.islandica were slightly higher (3700) than in M. edulis. BAF at thehigh exposure concentration in C. islandica was approximately40% below that estimated in the low and medium exposures andbelow the values estimated for M. edulis. Fig. 3 indicates that BAFswere significantly (r2 = 0.85; p < .0001) correlated to the lipophy-licity of PAHs in the two organisms regardless of the type of oiland conditions used in the two experiments.

Table 3Sum PAH analysed in seawater of the North Sea oil and Barents Sea oil experiments atthe three exposure concentrations.

Exposure treatment Sum PAH (lg/l)

North Sea Barents Sea

Low (L) 0.154 0.204Medium (M) 0.812 1.60High (H) 3.758 3.84

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

3.2. Biomarker responses

Mean GST and TOSC measured in control individuals wererespectively, 5-fold and 2-fold higher in C. islandica than in M. edu-lis. For catalase, the activities were similar in the control individu-als of the two species. NRRT in haemocytes of control individualswas also similar. Concerning the comet tail scores in haemocytesof control individuals, there was no significant variation for M. edu-lis during the three consecutive analytical days and a mean valuewas estimated. A significant variation was found for C. islandicaand the %DNA in comet tail in the control group is shown specifi-cally for the day when the low, medium and high exposure groupswere analysed (Table 6).

In M. edulis, we found significant difference of GST activity com-pared to the control only in the high exposure concentration. Therewere no statistical differences in any of the two other antioxidantparameters. In mussel haemocytes, the lysosomal membrane sta-bility assayed by NRRT did not reveal significant impairment. How-ever, DNA integrity of haemocytes was clearly impaired comparedto control as shown by the comet assay. The %DNA in comet tailwas significantly more elevated in all exposed individuals at thethree exposure concentrations. We found no significant differencein the %DNA in comet tail of the cells analysed in the medium (14%)and the high (13%) concentrations despite a 4-fold difference inbody burden.

Conversely to M. edulis, GST was only significantly different tothe control in the low exposure concentration in C. islandica. Also,there were significant impairments in CAT at all exposure concen-trations. Likewise, TOSC was significantly enhanced but showed asignificant difference from the control only in the low and mediumexposures. At the cellular level, both NRRT and the comet assaygave significant differences to the control in the medium and inthe high exposure concentrations only.

3.3. Multivariate analyses

The overall biomarker responses obtained at the different con-centrations in M. edulis and in C. islandica were first explored withthe integrated biomarker response (IBR). The results are shown inFig. 4. Compared to control individuals, IBR values were all ele-vated in exposed individuals of both species and the relationshipwith exposure dose was U-shaped. Hence, no good correspondencebetween IBR values and whole tissue PAH body burden wasshown. In both species, IBR values were markedly elevated atthe lower concentration whereas body burden of PAHs was thelowest. In fact, IBR was the highest at the lowest concentrationin M. edulis.

An indication of the contribution of each biomarker to the IBRplot is given in Table 7. For M. edulis, relatively high scores forCAT, TOSC and also NRRT appear to explain the high IBR value atthe low exposure concentration. Relatively high GST and cometscores are found in the IBR value of the high exposure. In C. islan-dica, GST and TOSC appeared to contribute dominantly in the IBRindex at the low concentration. The highest IBR value is shown inthe high exposure group corresponding to high scores for NRRTand comet.

The scores obtained for each biomarker to create the IBR plotwere further used in a PCA to improve identification of the dom-inant combination of biomarkers that characterized the re-sponses in M. edulis and C. islandica at the differentconcentrations (Fig. 5). The first three principal components(PC) represented 93% of the total variance in the data set. Gen-erally, the PCA did not allow separating clearly the two speciesbased on the biomarker loadings plotted in the co-ordinates ofthe three PCs. The different exposures were nevertheless charac-terized by distinct biomarker patterns in both species. PC1 (58%)

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

Table 4Comparison of the percentage of the sum of the three PAH categories in biota and seawater in the high exposure group in North Sea oil and Barents Sea oil experiments. The sameis also indicated for the two crude oils.

Sum PAH crude oil (%) Sum PAH seawater (%) Sum PAH biota (%)

North Sea Barents Sea North Sea Barents Sea M. edulis C. islandica

logKow 3–4 64 66 67 65 31 23logKow 4–5 28 28 27 30 48 48logKow >5 8 5 6 5 21 29

PAH

con

cent

ratio

n(µµ

g/L)

PAH

con

cent

ratio

n(µ

g/L)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

nac1

-nac2

-nac3

-nafl ph

c1-ph

c2-ph db

t

c1-db

t

c2-db

t py ch

North SeaBarents Sea

Fig. 1. Seawater PAH composition in North Sea oil and Barents Sea oil experiments(high exposure concentration). Only the quantifiable PAH are shown. Codes arethose shown in Table 1.

0

1000

2000

3000

4000

5000

nac1

-nac2

-nac3

-na phc1

-phc2

-ph dbt

c1-db

t

c2-db

t

PAH

in w

hole

tissu

e(µ

g/kg

)

M. edulisC. islandicaM. edulisC. islandica

0306090

pybaa ch

c1-chc2

-ch bbfbk

fbb

kfbap ip

bghip

0306090

pybaa ch

c1-chc2

-ch bbfbk

fbb

kfbap ip

bghip

Fig. 2. Individual PAH body burden in M. edulis and C. islandica at the high exposureconcentration in the experiments with North Sea and Barents Sea oil, respectively.

Table 5Bioaccumulation factor estimated from total PAHs in biota and seawater in the twoexperiments. L: low exposure; M: middle exposure; H: high exposure for North Seaoil and Barents Sea oil experiments.

Exposure group

L M H

North Sea 2838 3147 2647Barents Sea 3675 3694 2122

logBAF=0.8386logKow- 0.4049

logBAF=0.7638logKow - 0.10111

2

3

4

5

LogKow

LogB

AF

C.islandicaM.edulis

logBAF=0.8386logKow- 0.4049

logBAF=0.7638logKow - 0.10111

2

3

4

5

3 4 5 63 4 5 6LogKow

LogB

AF

C.islandicaM.edulis

Fig. 3. Relationship between log BAFand log Kow in M. edulis (n = 19) and in C.islandica (n = 24). The plot is based on data obtained at the high and mediumexposure concentrations for each of the two experiments. A linear fitting wasapplied and the resulting equation is shown.

6 T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx

ARTICLE IN PRESS

separated the low exposure groups of both species from thehigher exposure groups. This component was mainly associatedwith changes of the enzymatic responses and NRRT in the expo-sure gradient. Larger DNA alterations measured by the comet as-say in the high exposure groups of both species explained thegroup separation on PC2 (26%). The third component (PC3; 9%)was apparently associated with CAT response. For catalase, thetwo species appear to show an opposite pattern with exposureconcentrations.

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

4. Discussion

This study aimed to evaluate the responses of some commonlyused biomarkers together with chemical tissue uptake in the con-text of offshore monitoring from oil and gas industries. In recentyears, monitoring of the marine environment has shifted theemphasis away from mainly chemical analysis to an integratedassessment where biological and ecological approaches are imple-mented together with chemical data. A full incorporation of bio-markers into for example risk assessment is not yet achieved butmanagers and decision-makers have a more comprehensive panelof tools to base their assessment on compared to the approach tra-ditionally used.

Alternatives to using organisms to monitor chemical com-pounds in the water exist. For example, Vrana et al. (2005) have re-viewed a large number of passive sampling techniques formonitoring pollutants in water. However, biological assessment re-quires the use of organisms on which several metrics should betested and applied. Mytilus sp. is commonly used in water columnmonitoring for the temperate regions. Four to six weeks corre-sponds to the period of caged mussel deployment for the offshoremonitoring in these regions (Hylland et al., 2008; Gorbi et al., 2008)before sampling and analytical works are performed. This periodcorresponds also to the sampling time when the present data werecollected. For the northern areas, other species like C. islandica ap-pear more appropriate. To our knowledge, this study is one of thefirst to present results where a combination of chemistry, enzy-matic and cellular biomarkers are reported following chronic oilexposure to that species.

Compared to chemical data, our results tend to demonstratethat dose-to-effect responses are complex and strengthen the needfor a careful selection of biomarkers and interpretation when thetime-course of events is not known.

4.1. PAH bioaccumulation

Chemical analyses are the underpinning data on which to basedifferences in biomarker responses. The use of different oils was

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

Table 6Mean ± standard deviation in biomarker responses at the different concentrations and in control for the two species after 1 month exposure. When statistically significant is found(+++), the p-value is shown in parenthesis. n.s. = no significant difference. L, M, H are respectively, low, medium and high exposures at the corresponding total PAH concentration.C is control group. For C. islandica, the %DNA in comet tail in the control group is shown respectively, to the day when the low, medium and high exposure groups were analysed.See text for explanation.

Exposure group RPAH (lg L�1) C. islandica M. edulis

C L M H C L M H

0,2 1,6 3,8 0,15 0,8 3,7

Biomarker

GST +++ n.s. n.s. n.s. n.s. +++nmol/min/mg protein 283 ± 128 475 ± 184 355.5 ± 187.5 425 ± 247.5 56 ± 22 71 ± 30 62 ± 21 84.5 ± 23

(0.0032) (0.0072)

Catalase +++ +++ +++ n.s. n.s. n.s.nmol/min/mg protein 15 ± 9 28 ± 13 35 ± 20 25 ± 14 15.5 ± 10 21.5 ± 19 12 ± 7 17 ± 7

(0.001) (0.001) (0.027)

TOSC +++ +++ n.s. n.s. n.s. n.s.unit/mg protein 1091 ± 196 1356 ± 305 1274 ± 231 1243 ± 258 460 ± 116 407 ± 85 459 ± 128 477 ± 117

(0.0021) (0.047)

NRRT n.s. +++ +++ n.s. n.s. n.s.min 139 ± 14 126 ± 28 114 ± 19 102 ± 47 126 ± 42 93 ± 43.5 132 ± 47 114 ± 48.5

(0.0047) (0.026)

Comet n.s. +++ +++ +++ +++ +++%DNA in tail 14 ± 3/8 ± 1/10 ± 1 14 ± 4 10 ± 3 19 ± 7 8 ± 1 11 ± 1 14 ± 3 13 ± 2

(0.039) (0.0015) (<0.0001) (<0.0001) (<0.0001)

IBR

Sum

PAH

s( µ

g/kg

)

0

2

4

6

8

10

C L M H

M. edulisC. islandica

0

4

8

12

L M H

M. edulisC. islandica

A

B

Fig. 4. (A) Body burden of the sum of PAHs analysed in animal tissue for M. edulisand C. islandica at the three exposure concentrations. (B) IBR calculated from thescores of the five biomarkers for M. edulis and C. islandica at the three exposureconcentrations.

T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx 7

ARTICLE IN PRESS

directed by the goal of the Biosea I JIP program i.e. to evaluate theresponses of biomarkers in conditions relevant for the North Seaand the Barents Sea. The PAH composition of the two oils wereclose although the Barents Sea oil contained relatively more ofthe larger PAHs. Nevertheless, in the exposed seawater of thetwo experiments, we found the composition and concentration ofPAHs quite similar. Likewise, chemical PAH analyses in tissue sam-ples revealed no major differences in the pattern of PAH bioaccu-

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

mulation between M. edulis and C. islandica. BAFs were likewisein the same order of value for the two species.

Bivalves are efficient filter feeders that can accumulate rela-tively large amount of pollutants through several routes of expo-sure. The loss-rate constants of PAHs from oil particles toseawater proceed to a first-order kinetic based on molecular sizeand the ratio of surface area to volume (Short, 2002; Short and Hei-ntz, 1997). The fractions of dissolved and oil-bound PAHs in theseawater were likely governed by the same kinetic in our experi-mental system. Hence, in the exposure tanks, the dissolved fractionwas most likely dominated by the less lipophilic PAH moleculeswhich entered the bivalves via the gills. The larger, more lipophilicPAHs were more likely bound to oil particles and also algae thatwere ingested through the gut. Lipophilicity of PAHs expressedby the octanol–water partition coefficient appeared to govern thebioaccumulation pattern in our data. As metabolic degradation isnot as extensive in bivalves as it is reported for fish (Baussantet al., 2001a; Jonsson et al., 2004), we found a linear relationshipbetween the accumulation of PAHs in tissue and log Kow values.This relationship has been reported in several other bioaccumula-tion studies with these organisms (Meador et al., 1995; Baumardet al., 1999; Baussant et al., 2001b).

4.2. Biological responses in the two species

Generally, in comparison to control individuals, the biologicalresponses selected for this study showed more statistical signifi-cant alterations in exposed C. islandica than in exposed M. edulis.However, the relationship between exposure levels, enzymatic re-sponses and cellular damage was apparently complex. Severalexplanations can be proposed to explain the patterns revealed inour study. The difference in basal metabolism in the two species,the time-course of events of biomarker, the dose-related biomarkerresponses and the toxicity mechanisms involved can all be argued.

TOSC and the GST value measured in control organisms of thetwo molluscs were respectively, 5-fold and 2-fold lower in M. edulisthan in C. islandica. Cold water species are reported to have higherlevels of antioxidant enzymes to protect them against the deleteri-ous effects of different ROS in their natural pro-oxidant environ-ment (Regoli et al., 2000). In C. islandica, Camus et al. (2002)

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

Table 7Scores and corresponding integrated biomarker response (IBR) value for each treatment groups and for both species.

Species Exposure group Code CAT GST TOSC Comet NRRT IBR

M. edulis C Mk0 0.92 0 0.56 0 0 0L ML 2.42 1.18 2.31 1.07 2.09 8.38M MM 0 0.51 0.58 2.22 0 0.80H MH 1.24 2.31 0 1.94 1.04 3.38

C. islandica C Ck0 0 0 0 0.39 0 0L CL 1.71 2.30 2.39 0 1.39 6.61M CM 2.36 0.86 1.65 0.89 1.39 3.31H CH 1.18 1.69 1.37 2 2.44 8.81

8 T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx

ARTICLE IN PRESS

measured TOSC (towards peroxyl) value in control individuals of2500 unit/mg protein. In another species (Mya truncata), Camuset al. (2003) found values of 4010 unit/mg protein. In our studywith C. islandica TOSC was about 1100 unit/mg protein which iscomparable to the basal level found by Regoli et al. (2000) for thisspecies. For M. edulis, the basal TOSC level was 460 unit/mg protein,an order of value found for mussels in temperate regions (Frenzilliet al., 2004; Bocchetti and Regoli, 2006). A relatively high basal levelof antioxidant defence should help C. islandica to cope with moder-ate changes in ROS formation better than M. edulis. Our datashowed that enzymatic induction was observed in the low rangeof exposure in the scallop but no induction of the enzymatic re-sponses was observed in the mussel at the same range of exposure(except for GST at the high exposure). This appears to indicate thatdifferences in the basal metabolism of the two species do not ap-pear to explain solely our observations but different sensitivity tooil may exist.

In C. islandica, a ROS-mediated toxicity was apparently revealedfrom the biomarker patterns. A counteractive response towards theenhanced production of ROS is suggested by the induction of GST,catalase and TOSC in the low exposure range. At higher level, thescavenging capacity may be overwhelmed and the mollusc doesnot manage to cope effectively with ROS, hence the cellular dam-ages revealed at the medium and high exposure concentration. Ina field translocation experiment in the Genova harbour with cagedMytilus galloprovinicalis for four weeks, similar findings were re-ported by Regoli et al. (2004) and Frenzilli et al. (2004). Theseauthors measured a set of single antioxidants, TOSC and cellulardamages including DNA integrity and the impairment of lysosmalmembrane stability in haemocytes. Their observations showed abiphasic trend for single antioxidants and TOSC with an increaseduring the first two weeks of exposure in the harbour, then a de-crease followed by a severe depletion at the end of the deployment.Cell damages were progressively enhanced during their experi-ment. Similar observations were made by Camus et al. (2004) inthe Venice Lagoon although these authors observed a significantreduction of TOSC during the first two weeks, then a return to ini-tial level after four weeks.

In the case of prolonged exposure, the induced response of sev-eral parameters may decline with time even if the level of pollutantremains constant. These time-related changes may be related to anadaptation of organisms during exposure in order to restorehomeostasis (Wu et al., 2005) or an enzymatic inhibition at highexposure level or after prolonged exposure (Camus et al., 2002).

In M. edulis, the prevalence of DNA impairment by the cometassay at all exposure concentrations may also be related to aROS-mediated toxicity. However, in our study, only GST showeda significant increase at the high exposure level. This parameterindicated a response of phase II metabolism to detoxify foreigncompounds by conjugation with gluthathione. The apparent ab-sence of significant changes in TOSC and catalase at any of theexposure concentrations may be explained by a transient responseof antioxidants, with return to control levels at the time of

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

sampling, or enzymatic inhibition at higher exposure level. In addi-tion to DNA integrity changes, a reduced ability to neutralise ROSappears to correlate with cellular membrane impairment in bi-valves (Regoli et al., 2004). The lysosomal integrity in mussels is re-ported as a reversible biomarker after a recovery period in cleanseawater. In their study, Fang et al. (2008) found no further changeor adaptation of this biomarker over a period of 62 days to ben-zo[a]pyrene exposure and following a significant reduction after6 days. No significant alteration of the lysosomal membrane stabil-ity was demonstrated in M. edulis exposed for a month in ourstudy. Several environmental and biological factors can influencelysosomal membrane alterations. For example, tissue reorganiza-tion during the reproductive cycle of the individuals has beenshown to affect NRRT (Bocchetti and Regoli, 2006; Harding et al.,2004; Nasci et al., 1998). This may also have been the case in ourstudy. However, another possible explanation is that the degreeof oxidative stress was not high enough to induce a response inM. edulis. The alteration of DNA integrity could then be directly re-lated to strand break generation or replicative gaps without gener-ation of ROS. Here, we used a limited number of antioxidantresponse parameters. Possibly, the use of other antioxidant en-zyme activities like superoxide dismutase or gluthathione peroxi-dase might have helped us to better reveal how these organismsadapt to the pro-oxidant challenge in presence of oil.

4.3. Multivariate analyses with IBR and PCA

Based on the five biomarkers measured in M. edulis and C. islan-dica, we have used the integrated biomarker response that com-bines all the responses into one index. The goal was tosummarize and grade in a simple manner the level of biological re-sponses, rather than consider each biomarker response separately.The IBR from Beliaeff and Burgeot (2002) is a simple method tovisualize the results of a suite of biomarkers. The method has beenapplied to field samples collected from hydrocarbon polluted areas(Bocquene et al., 2004; Bodin et al., 2004). Here, we employed it toscale the overall biomarker responses in the two species and at thedifferent oil exposure concentrations used in our experiments.Generally, the IBR values for exposed individuals of both specieswere more elevated than control individuals. This argues for bio-logical impairments at the concentration range tested for the NorthSea and the Barents Sea oil exposures. We showed that the IBR pat-tern was comparable and the relationship to oil concentration wasU-shaped for the two species. There was a relatively elevated IBR atthe low and the high exposure concentrations while this index waslower at the medium exposure concentration. IBR in M. edulis wasactually highest at the lowest exposure where only comet showeda statistical significant difference to the control. It is possible thateven though the statistic did not reveal any significant differencerelated to enzymatic responses, there were some changes in thedistribution of biomarker data in some individuals compared tothe control group. This was better revealed by the methodologyused to calculate IBR likely because it combines the responses of

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

-1.5

-1

-0.5

0

0.5

1

1.5

2

Prin

2

Mk0

MHMM

MLCk0

CH

CMCL

-3 -2 -1 0 1 2 3Prin1

-1.5

-1

-0.5

0

0.5

1

1.5

2

Ck0

-3 -2 -1 0 1 2 3

-1.5

-1

-0.5

0

0.5

1

Prin

3 Mk0

MH

MMML

Ck0

CHCM

CL

-2 -1.5 -1 -0.5 0 0.5 1 1.5 2Prin2

-1.5

-1

-0.5

0

0.5

1

-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

-0.5

-0.25

0

0.25

0.5

1

PC2

Com

Cat

ToscGst

Nrrt

-0.1 0 .1 .2 .3 .4 .5 .6

PC1

-0.5

-0.25

0

0.25

0.5

1Com

Cat

ToscGst

Nrrt

-0.1 0 .1 .2 .3 .4 .5 .6

-1

-0.5

0

0.5

1

PC3 Tosc

GstNrrt

Cat

Com

-0.5 -0.25 0 .25 .5 .75 1

PC2

-1

-0.5

0

0.5

1

ToscGst

Nrrt

Cat

Com

-0.5 -0.25 0 .25 .5 .75 1

Fig. 5. Principal component analysis performed on biomarker score data computed from the IBR calculation. The three first principal components (PC) are plotted againsteach other (left panel) and in correspondence with the variable loading plots (right panel). The abbreviations are as those reported in Table 7 (species-related exposuregroups) and Table 2 (biomarkers).

T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx 9

ARTICLE IN PRESS

the different biomakers instead of considering each biomarkerindependently.

In their study, Bocquene et al. (2004) used four biomarkers andcomputed the IBR to quantify the impact of the ‘‘Erika” oil spill onthe mussel M. edulis. Two of the biomarkers, GST and CAT activitieswere common to our studies. Acetylcholiesterase (AChE) and mal-ondialdehyde (MDA) concentrations were also measured in theirstudies. They found that IBR values were elevated during the firstyear after the spill and strongly related to MDA levels, a time-inte-grated marker of oxidation stress. Taking the Baltic Sea as a casestudy, Beliaeff and Burgeot (2002) computed the IBR values basedon AChE, CAT and GST measured in gills and digestive gland of M.edulis in March 1995, October 1995 and November 1996. IBR starplots showed good correspondence with total PAHs measured inmussel tissues in October 1995 while other sources of contami-nants were probably present in March 1995 and November 1996.Generally, Beliaeff and Burgeot concluded that a good visual con-cordance does not appear to exist between IBR and PAH plots. Like-wise, Damiens et al. (2007) combined five biomarkers (AChE, GST,CAT, TBARS and MT concentrations) and used the IBR to assess theenvironmental effect using transplanted M. galloprovincialis. Theyfound a good visual correlation between their IBR values and bothcopper and PCB concentrations in mussel tissues but not with PAHconcentrations. In the present study, in both bivalve species, therewas no good agreement between body burden of PAHs and IBRvalues.

Careful consideration of the IBR mathematical expressionshould be taken to avoid false interpretation and oversimplifica-tion. In their article, Broeg and Lehtonen (2006) underlined somespecific considerations concerning IBR. The value of the IBR isdependent on the type and number of biomarkers used for the cal-culation. These authors claimed that the set of biomarkers selected

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

should ideally be larger than five to reduce the ‘‘relative weight” ofeach biomarker in the final index value. Also, with more than fivebiomarkers, the order of the biomarkers within the data set doesnot play a significant role. Due to the multiplication of the stand-ardised biomarker values with each other, the IBR calculationmay produce a zero or a low value even if some biomarkers are ele-vated. In the present study, all control individuals from both spe-cies showed zero or low scores. In M. edulis at the mediumconcentration, comet score was the highest of all exposure groupsbut the score for the other biomarkers was zero or low. Hence, thefinal IBR value at the medium concentration was low.

The goal of the multivariate analysis by PCA was to reveal betterthe combination of parameters explaining the IBR results andeventually demonstrate different patterns in the two species. Aseparation of the two species based on the variable loadings inPC1 and PC2 was impossible. Only PC3 was able to separate thetwo species due to differences in the pattern of catalase activityfor the different exposure concentrations. Hence, from the multi-variate analysis, a similar combination of biomarkers and patternscharacterized both M. edulis and C. islandica in the oil concentrationrange tested. This observation may indicate that an analogous tox-icity mechanism existed in both species following the exposure tocrude oil. Possibly, in both species, a similar chain of biologicalevents following ROS enhancement took place but there were spe-cific differences in the time-course of biomarker responses.

5. General recommendations and conclusive remarks

For the management of seawater effluent related to offshoreindustries, biomarkers that are subject to complex relationshipswith exposure are challenging to employ due to the possibility ofincorrect conclusions. In particular, it is critical to describe better

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

10 T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx

ARTICLE IN PRESS

the temporal changes of these biomarkers before eventually mak-ing use of them in environmental marine monitoring (Wu et al.,2005). Currently, in offshore biomonitoring programs, the cost ofin situ operations and in some situations (Arctic) the difficulty toaccess the sites, restrict the sampling to one event i.e. at the endof the deployment period (Hylland et al., 2008). Hence, in practice,the time-course of biomarker evolution during the deploymentperiod may be unknown. Consequently, the selection of biomark-ers for these programs should be carefully based on considerationsrelated to consistency of response to the dose and maintenance ofthe response with time to be useful for impact assessment anddecision-making (Hagger et al., 2008).

This study shows that, for biomarkers related to enzymatic re-sponses, it is important to know the temporal changes to correctlyinterpret them. A simple relationship between the magnitude ofthese responses and the tissue PAH concentration was not foundbased on one sampling event after 4 weeks exposure. From analo-gous experiences with TOSC for example, a sampling event withinthe first two weeks following the start of the exposure would havebeen preferable (Regoli et al., 2004; Camus et al., 2004). Neverthe-less, linking these types of toxic responses to the exposure is hard-er, possibly limiting their practical use for field monitoring.

Indubitably, the comet assay performed on haemocytes was themost consistent biological response for the two studied molluscspecies. Taban et al. (2004) also found a significant concentrationrelated increase in the percentage of DNA in comet tail in sea urch-in coelomocytes exposed at the medium and high exposure dose ofthe North Sea oil. This genotoxic marker appears to be a suitablemethod for chronic exposure related to offshore petroleum. Thesuitability of the comet assay for petroleum-related compoundshas also been reported in several other studies (Gorbi et al.,2008; Thomas et al., 2007; Laffon et al., 2006; Akcha et al., 2003).In addition, the responses from the comet assay have been linkedto effects on growth, development and reproduction in severalorganisms (Jha, 2008). This makes the method even more appropri-ate for risk assessment and long-term biological consequences athigher organization levels than the individual level. Yet, for a rou-tine application in biomonitoring studies e.g. offshore monitoringaround platforms and rigs, this assay still requires proper optimi-zation, inter-laboratory harmonization, basal level measurementswith different species as well as an evaluation of the sensitivityto exposure dose (Jha, 2008; Akcha et al., 2004). The use of othergenotoxic markers with time-integrated responses like the micro-nucleus test (Gorbi et al., 2008; Barsiene et al., 2006) and DNA ad-duct (Bocquene et al., 2004) appears also recommended in thecontext of offshore biomonitoring.

The neutral red retention time to assess cell membrane impair-ment in haemocytes is one of the most frequently used biomarkerswith bivalves for environmental monitoring. It is often described asa general ‘‘health” parameter of cells responding well in the labo-ratory as in the field to several sources of xenobiotics. Current dis-cussions within the Working Group on Biological Effects ofContaminants of ICES are proposing lysosomal membrane stabilityas the core biomarker in tier 1 of a 2-tier approach for biomonitor-ing programmes (ICES, 2008b). In M. edulis exposed to North Seaoil, NRRT did not show any statistical significance to the controlgroup. Structural changes in the digestive lysosomal system of bi-valves by the application of histo-cytochemical technique may bemore robust than the neutral red assay on haemocytes. Thesechanges appear to be expressed under pollution-induced stressand even under very complex field situations where seasonal fac-tors (temperature, salinity, food) and biological variations (repro-ductive status) may affect the structure of the digestivelysosomal system (Marigomez et al., 1996).

Our study clearly demonstrates that the selection of metricsshould not be restricted to only one level of biological organization

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

(enzymatic or cellular) but rather carefully made at several levelsfrom the sub-individual level to more ecologically-relevant levelsi.e. population or community. This would avoid the problem re-lated to false-negative responses and provide an ecologicallymeaningful assessment of aquatic ecosystem health. Also, in com-plement to the usual way of analysing biomarker responses (i.e.statistical significance to a reference group or set of data), theircombination into an overall index may provide an easier way tosummarize the environmental status and communicate to deci-sion-makers.

Biomarkers are too seldom associated with ecological relevantend point fitness parameters. We believe this is a weakness ofthe current use of biomarkers in many studies. Establishing linksbetween biomarkers and actual effects should provide more mean-ingful answers to the critical questions raised by decision-makersand help towards a better integration into regulatory context. Inour study with blue mussels, other long term effects on reproduc-tive status were appraised by histological observations of the go-nad and fitness parameters on early life stages obtained from theexposed parents. This work will be presented in a following paper.

In conclusion of this work, this study showed that: (1) C. islan-dica appears a good candidate species for biomonitoring studies ofpetroleum-related pollution in the Barents Sea; (2) for long termeffect assessment related to offshore effluents, the selection of bio-markers should take into consideration both dose- and time-re-lated events; (3) the use of cellular or other higher level longlasting biological responses can be recommended for chronic expo-sure while enzymatic measurements with transient responses aremore challenging to interpret and (4) integrative multivariate pre-sentation of biomarkers could be adopted to facilitate their com-munication to environmental managers and decision-makers.

Acknowledgements

The Biosea I JIP is a joint industrial programme financed by ENINorge AS and Total E&P Norge AS. We are grateful to extra supportprovided through projects financed by the Research Council of Nor-way under the program ‘‘PROOF”. Also, we would like to thank Em-ily Lyng, Harald Berland, Rolf Sundt, Lars Petter Myrhe and ØyvindF. Tvedten at IRIS for their precious help and support in the analyt-ical work load. The final writing of this paper has been financedthrough additional funds from Total E&P Norge AS.

References

Aarab, N., Minier, C., Lemaire, S., Unruh, E., Hansen, P.D., Larsen, B.K., Andersen, O.K.,Narbonne, J.F., 2004. Biochemical and histological responses in mussel (Mytilusedulis) exposed to North Sea oil and to a mixture of North Sea oil andalkylphenols. Marine Environmental Research 58, 437–441.

Akcha, F., Hubert, F.V., Pfhol-Leszkowicz, A., 2003. Potential value of the cometassay and DNA adduct measurement in dab (Limanda limanda) for assessmentof in situ exposure to genotoxic compounds. Mutation Research – GeneticToxicology and Environmental Mutagenesis 534, 21–32.

Akcha, F., Tanguy, A., Leday, G., Pelluhet, L., Budzinski, H., Chiffoleau, J.F., 2004.Measurement of DNA single-strand breaks in gill and hemolymph cells ofmussels, Mytilus sp., collected on the French Atlantic Coast. MarineEnvironmental Research 58, 753–756.

Allan, I.J., Vrana, B., Greenwood, R., Mills, G.A., Roig, B., Gonzalez, C., 2006. A‘‘toolbox” for biological and chemical monitoring requirements for theEuropean Union’s Water Framework Directive. Talanta 69, 302–322.

Barsiene, J., Lehtonen, K.K., Angela, K., Broeg, K., Vuorinen, P.J., Lang, T., Pempkowiak,J., Yvokien, J., Dedonyte, V., Rybakovas, A., Repecka, R., Vuontisjarvi, H., Kopecka,J., 2006. Biomarker responses in flounder (Platichthys flesus) and mussel (Mytilusedulis) in the Klaipeda-Butinge area (Baltic Sea). Marine Pollution Bulletin 53,422–436.

Baumard, P., Budzinski, H., Garrigues, P., Burgeot, T., Michel, X., Bellocq, J., 1999.Polycyclic aromatic hydrocarbon (PAH) burden of mussels (Mytilus sp.) indifferent marine environments in relation with sediment PAH contamination,and bioavailability. Marine Environmental Research 47, 415–439.

Baussant, T., Sanni, S., Jonsson, G., Skadsheim, A., Borseth, J.F., 2001a.Bioaccumulation of polycyclic aromatic compounds: 1. Bioconcentration intwo marine species and in semipermeable membrane devices during chronic

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx 11

ARTICLE IN PRESS

exposure to dispersed crude oil. Environmental Toxicology and Chemistry 20,1175–1184.

Baussant, T., Sanni, S., Skadsheim, A., Jonsson, G., Borseth, J.F., Gaudebert, B., 2001b.Bioaccumulation of polycyclic aromatic compounds: 2. Modeling,bioaccumulation in marine organisms chronically exposed to dispersed oil.Environmental Toxicology and Chemistry 20, 1185–1195.

Beliaeff, B., Burgeot, T., 2002. Integrated biomarker response: a useful tool forecological risk assessment. Environmental Toxicology and Chemistry 21, 1316–1322.

Bocchetti, R., Regoli, F., 2006. Seasonal variability of oxidative biomarkers,lysosomal parameters, metallothioneins and peroxisomal enzymes in theMediterranean mussel Mytilus galloprovincialis from Adriatic Sea.Chemosphere 65, 913–921.

Bocquene, G., Chantereau, S., Clerendeau, C., Beausir, E., Menard, D., Raffin, B.,Minier, C., Burgeot, T., Leszkowicz, A.P., Narbonne, J.F., 2004. Biological effects ofthe ‘‘Erika” oil spill on the common mussel (Mytilus edulis). Aquatic LivingResources 17, 309–316.

Bodin, N., Burgeot, T., Stanisiere, J.Y., Bocquene, G., Menard, D., Minier, C., Boutet, I.,Amat, A., Cherel, Y., Budzinski, H., 2004. Seasonal variations of a battery ofbiomarkers and physiological indices for the mussel Mytilus galloprovincialistransplanted into the northwest Mediterranean Sea. Comparative Biochemistryand Physiology C-Toxicology & Pharmacology 138, 411–427.

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein–dye binding.Analytical Biochemistry 72, 248–254.

Broeg, K., Lehtonen, K.K., 2006. Indices for the assessment of environmentalpollution of the Baltic Sea coasts: integrated assessment of a multi-biomarkerapproach. Marine Pollution Bulletin 53, 508–522.

Cajaraville, M.P., Bebianno, M.J., Blasco, J., Porte, C., Sarasquete, C., Viarengo, A.,2000. The use of biomarkers to assess the impact of pollution in coastalenvironments of the Iberian Peninsula: a practical approach. Science of theTotal Environment 247, 295–311.

Camus, L., Jones, M.B., Borseth, J.F., Grosvik, B.E., Regoli, F., Depledge, M.H., 2002.Total oxyradical scavenging capacity and cell membrane stability ofhaemocytes of the Arctic scallop, Chlamys islandicus, following benzo(a)pyreneexposure. Marine Environmental Research 54, 425–430.

Camus, L., Birkely, S.R., Jones, M.B., Borseth, J.F., Grosvik, B.E., Gulliksen, B., Lonne,O.J., Regoli, F., Depledge, M.H., 2003. Biomarker responses and PAH uptake inMya truncata following exposure to oil-contaminated sediment in an Arcticfjord (Svalbard). Science of the Total Environment 308, 221–234.

Camus, L., Pampanin, D.M., Volpato, E., Delaney, E., Sanni, S., Nasci, C., 2004. Totaloxyradical scavenging capacity responses in Mytilus galloprovincialistransplanted into the Venice lagoon (Italy) to measure the biological impactof anthropogenic activities. Marine Pollution Bulletin 49, 801–808.

Claiborne, A., 1985. Catalase activity. In: Greenwald, R.A. (Ed.), CRC Handbook ofMethods for Oxygen Radical Research. CRC Press, Inc., Florida, pp. 283–284.

Damiens, G., Gnassia-Barelli, M., Loques, F., Romeo, M., Salbert, V., 2007. Integratedbiomarker response index as a useful tool for environmental assessmentevaluated using transplanted mussels. Chemosphere 66, 574–583.

Devier, M.H., Augagneur, S., Budzinski, H., Le Menach, K., Mora, P., Narbonne, J.F.,Garrigues, P., 2005. One-year monitoring survey of organic compounds (PAHs,PCBs, TBT), heavy metals and biomarkers in blue mussels from the ArcachonBay, France. Journal of Environmental Monitoring 7, 224–240.

Digiulio, R.T., Washburn, P.C., Wenning, R.J., Winston, G.W., Jewell, C.S., 1989.Biochemical responses in aquatic animals – a review of determinants ofoxidative stress. Environmental Toxicology and Chemistry 8, 1103–1123.

Douglas, G.S., Prince, R.C., Butler, E.L., Steinhauer, W.G., 1994. The use of internalchemical indicators in petroleum and refined products to evaluate the extent ofbiodegradation. In: Hinchee, R.E., Alleman, B.C., Hoeppel, R.E., Miller, R.N. (Eds.),Hydrocarbon Bioremediation. Boca Raton, US, pp. 219–236.

Fang, J.K.H., Wu, R.S.S., Zheng, G.J., Lam, P.K.S., Shin, P.K.S., 2008. Induction,adaptation and recovery of lysosomal integrity in green-lipped mussel Pernaviridis. Marine Pollution Bulletin 57, 467–472.

Forbes, V.E., Palmqvist, A., Bach, L., 2006. The use and misuse of biomarkers inecotoxicology. Environmental Toxicology and Chemistry 25, 272–280.

Frenzilli, G., Bocchetti, R., Pagliarecci, M., Nigro, M., Annarumma, F., Scarcelli, V.,Fattorini, D., Regoli, F., 2004. Time-course evaluation of ROS-mediated toxicityin mussels, Mytilus galloprovincialis, during a field translocation experiment.Marine Environmental Research 58, 609–613.

Galloway, T.S., Brown, R.J., Browne, M.A., Dissanayake, A., Lowe, D., Jones, M.B.,Depledge, M.H., 2004. A multibiomarker approach to environmentalassessment. Environmental Science & Technology 38, 1723–1731.

Gorbi, S., Lamberti, C.V., Notti, A., Benedetti, M., Fattorini, D., Moltedo, G., Regoli, F.,2008. An ecotoxicological protocol with caged mussels, Mytilus galloprovincialis,for monitoring the impact of an offshore platform in the Adriatic sea. MarineEnvironmental Research 65, 34–49.

Hagger, J.A., Jones, M.B., Lowe, D., Paul Leonard, D.R., Owen, R., Galloway, T.S., 2008.Application of biomarkers for improving risk assessments of chemicals underthe Water Framework Directive: a case study. Marine Pollution Bulletin 56,1111–1118.

Harding, J.M., Couturier, C., Parsons, G.J., Ross, N.W., 2004. Evaluation of the neutralred retention assay as a stress response indicator in cultivated mussels (Mytilusspp.) in relation to seasonal and environmental conditions. Journal of ShellfishResearch 23, 745–751.

Hylland, K., Tollefsen, K.E., Ruus, A., Jonsson, G., Sundt, R.C., Sanni, S., Utvik, T.I.R.,Johnsen, S., Nilssen, I., Pinturier, L., Balk, L., Barsiene, J., Marigomez, I., Feist, S.W.,

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

Borseth, 2008. Water column monitoring near oil installations in the North Sea2001–2004. Marine Pollution Bulletin 56, 414–429.

ICES, 2008a. Report of the Fourth ICES/OSPAR Workshop on Integrated Monitoringof Contaminants and their Effects in Coastal and Open Sea Areas (WKIMON IV),5–7 February 2008. ICES, Copenhagen, Denmark, ICES CM 2008/ACOM:49, 82pp.

ICES, 2008b. Report of the Working Group on Biological Effects of Contaminants(WGBEC), 31 March – 4 April. Sète, France, ICES CM 2008/MHC:07, 86 pp.

Jha, A.N., 2008. Ecotoxicological applications and significance of the comet assay.Mutagenesis 23, 207–221.

Jonsson, G., Bechmann, R.K., Bamber, S.D., Baussant, T., 2004. Bioconcentration,biotransformation, and elimination of polycyclic aromatic hydrocarbons insheepshead minnows (Cyprinodon variegatus) exposed to contaminatedseawater. Environmental Toxicology and Chemistry 23, 1538–1548.

Laffon, B., Rabade, T., Pasaro, E., Mendez, J., 2006. Monitoring of the impact ofPrestige oil spill on Mytilus galloprovincialis from Galician coast. EnvironmentInternational 32, 342–348.

Lee, R.F., Steinert, S., 2003. Use of the single cell gel electrophoresis/comet assay fordetecting DNA damage in aquatic (marine and freshwater) animals. MutationResearch-Reviews in Mutation Research 544, 43–64.

Livingstone, D.R., 2003. Oxidative stress in aquatic organisms in relation topollution and aquaculture. Revue De Medecine Veterinaire 154, 427–430.

Livingstone, D.R., Chipman, J.K., Lowe, D.M., Minier, C., Mitchelmore, C.L., Moore,M.N., Peters, L.D., Pipe, R.K., 2000. Development of biomarkers to detect theeffects of organic pollution on aquatic invertebrates: recent molecular,genotoxic. Cellular and immunological studies on the common mussel(Mytilus edulis L.) and other mytilids. International Journal of Environmentand Pollution 13, 56–91.

Lowe, D.M., Pipe, R.K., 1994. Contaminant induced lysosomal membrane damage inmarine mussel digestive cells – an in-vitro study. Aquatic Toxicology 30, 357–365.

Mamaca, E., Bechmann, R.K., Torgrimsen, S., Aas, E., Bjornstad, A., Baussant, T., LeFloch, S., 2005. The neutral red lysosomal retention assay and Comet assay onhaemolymph cells from mussels (Mytilus edulis) and fish (Symphodus melops)exposed to styrene. Aquatic Toxicology 75, 191–201.

Marigomez, I., Orbea, A., Olabarrieta, I., Etxeberria, M., Cajaraville, M.P., 1996.Structural changes in the digestive lysosomal system of sentinel mussels asbiomarkers of environmental stress in mussel-watch programmes. ComparativeBiochemistry and Physiology C –Pharmacology Toxicology & Endocrinology113, 291–297.

Meador, J.P., Stein, J.E., Reichert, W.L., Varanasi, U., 1995. Bioaccumulation ofpolycyclic aromatic hydrocarbons by marine organisms. In: Ware, G.W. (Eds.),Reviews of Environmental Contamination and Toxicology, vol. 143. Springer-Verlag, New York, Inc., pp. 79–166.

Mitchelmore, C.L., Chipman, J.K., 1998. DNA strand breakage in aquatic organismsand the potential value of the comet assay in environmental monitoring.Mutation Research – Fundamental and Molecular Mechanisms of Mutagenesis399, 135–147.

Moore, M.N., Willows, R.I., 1998. A model for cellular uptake and intracellularbehaviour of particulate-bound micropollutants. Marine EnvironmentalResearch 46, 509–514.

Moore, M.N., Depledge, M.H., Readman, J.W., Leonard, D.R.P., 2004. An integratedbiomarker-based strategy for ecotoxicological evaluation of risk inenvironmental management. Mutation Research – Fundamental andMolecular Mechanisms of Mutagenesis 552, 247–268.

Moreira, S., Guilhermino, L., 2005. The use of Mytilus galloprovincialisacetylcholinesterase and glutathione S-transferases activities as biomarkers ofenvironmental contamination along the northwest Portuguese coast.Environmental Monitoring and Assessment 105, 309–325.

Nasci, C., Da Ros, L., Campesan, G., Fossato, V.U., 1998. Assessment of the impact ofchemical pollutants on mussel, Mytilus galloprovincialis, from the VeniceLagoon, Italy. Marine Environmental Research 46, 279–282.

Orbea, A., Ortiz-Zarragoitia, M., Sole, M., Porte, C., Cajaraville, M.P., 2002.Antioxidant enzymes and peroxisome proliferation in relation to contaminantbody burdens of PAHs and PCBs in bivalve molluscs, crabs and fish from theUrdaibai and Plentzia estuaries (Bay of Biscay). Aquatic Toxicology 58, 75–98.

Perez-Cadahia, B., Laffon, B., Pasaro, E., Mendez, J., 2004. Evaluation of PAHbioaccumulation and DNA damage in mussels (Mytilus galloprovincialis)exposed to spilled Prestige crude oil. Comparative Biochemistry andPhysiology C – Toxicology & Pharmacology 138, 453–460.

Porte, C., Sole, M., Borghi, V., Martinez, M., Chamorro, J., Torreblanca, A., Ortiz, M.,Orbea, A., Soto, M., Cajaraville, M.P., 2001. Chemical, biochemical and cellularresponses in the digestive gland of the mussel Mytilus galloprovincialis from theSpanish Mediterranean coast. Biomarkers 6, 335–350.

Regoli, F., Winston, G.W., 1999. Quantification of total oxidant scavenging capacityof antioxidants for peroxynitrite, peroxyl radicals, and hydroxyl radicals.Toxicology and Applied Pharmacology 156, 96–105.

Regoli, F., Winston, G.W., Mastrangelo, V., Principato, G., Bompadre, S., 1998. Totaloxyradical scavenging capacity in mussel Mytilus sp. As a new index ofbiological resistance to oxidative stress. Chemosphere 37, 2773–2783.

Regoli, F., Nigro, M., Chiantore, M., Gorbi, S., Winston, G., 2000. Total oxidantscavenging capacity of Antarctic, Arctic, and Mediterranean scallops. ItalianJournal of Zoology 67, 85–94.

Regoli, F., Gorbi, S., Frenzilli, G., Nigro, M., Corsi, I., Focardi, S., Winston, G.W., 2002.Oxidative stress in ecotoxicology: from the analysis of individual antioxidantsto a more integrated approach. Marine Environmental Research 54, 419–423.

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007

12 T. Baussant et al. / Marine Pollution Bulletin xxx (2009) xxx–xxx

ARTICLE IN PRESS

Regoli, F., Winston, G.W., Gorbi, S., Frenzilli, G., Nigro, M., Corsi, I., Focardi, S., 2003.Integrating enzymatic responses to organic chemical exposure with totaloxyradical absorbing capacity and DNA damage in the European eel Anguillaanguilla. Environmental Toxicology and Chemistry 22, 2120–2129.

Regoli, F., Frenzilli, G., Bocchetti, R., Annarumma, F., Scarcelli, V., Fattorini, D., Nigro,M., 2004. Time-course variations of oxyradical metabolism, DNA integrity andlysosomal stability in mussels, Mytilus galloprovincialis, during a fieldtranslocation experiment. Aquatic Toxicology 68, 167–178.

Sanni, S., Oysaed, K.B., Hoivangli, V., Gaudebert, B., 1998. A continuous flow system(CPS) for chronic exposure of aquatic organisms. Marine EnvironmentalResearch 46, 97–101.

Short, J.W., 2002. Oil identification based on a goodness-of-fit metric applied tohydrocarbon analysis results. Environmental Forensics 3, 349–355.

Short, J.W., Heintz, R.A., 1997. Identification of Exxon Valdez oil in sediments andtissues from Prince William Sound and the northwestern Gulf of Alaska based ona PAH weathering model. Environmental Science & Technology 31, 2375–2384.

Singh, N.P., Mccoy, M.T., Tice, R.R., et al., 1998. A simple technique for quantificationof low levels of DNA damage in individual cells. Experimental Cell Research 175,184–191.

Please cite this article in press as: Baussant, T., et al. Enzymatic and cellular respwith chronic levels of North Sea and Barents Sea dispersed oil. Mar. Pollut. Bu

Taban, I.C., Bechmann, R.K., Torgrimsen, S., Baussant, T., Sanni, S., 2004. Detection ofDNA damage in mussels and sea urchins exposed to crude oil using cometassay. Marine Environmental Research 58, 701–705.

Thomas, R.E., Lindeberg, M., Harris, P.M., Rice, S.D., 2007. Induction of DNA strandbreaks in the mussel (Mytilus trossulus) and clam (Protothaca staminea)following chronic field exposure to polycyclic aromatic hydrocarbons fromthe Exxon Valdez spill. Marine Pollution Bulletin 54, 726–732.

Vrana, B., Mills, G.A., Allan, I.J., Dominiak, E., Svensson, K., Knutsson, J., Morrison, G.,Greenwood, R., 2005. Passive sampling techniques for monitoring pollutants inwater. Trac-Trends in Analytical Chemistry 24, 845–868.

Winston, G.W., Moore, M.N., Kirchin, M.A., Soverchia, C., 1996. Production ofreactive oxygen species by hemocytes from the marine mussel, Mytilus edulis:lysosomal localization and effect of xenobiotics. Comparative Biochemistry andPhysiology C – Pharmacology Toxicology & Endocrinology 113, 221–229.

Wu, R.S.S., Siu, W.H.L., Shin, P.K.S., 2005. Induction, adaptation and recovery ofbiological responses: implications for environmental monitoring. MarinePollution Bulletin 51, 623–634.

onses in relation to body burden of PAHs in bivalve molluscs: A case studyll. (2009), doi:10.1016/j.marpolbul.2009.08.007