Genotoxic and CYP 1A enzyme effects consecutive to the food transfer of oil spill contaminants from mussels to mammals

Aquat. Living Resour. 17, 303–307 (2004)
Aquatic
c EDP Sciences, IFREMER, IRD 2004
Living
DOI: 10.1051/alr:2004028
www.edpsciences.org/alr
Resources
Genotoxic and CYP 1A enzyme effects consecutive to the food
transfer of oil spill contaminants from mussels to mammals
Sébastien Lemière1,a, Carole Cossu-Leguille1, Sylvie Chaty1, François Rodius1, Antonio Bispo2,
Marie-José Jourdain2, Marie-Claire Lanhers3, Daniel Burnel3 and Paule Vasseur1
1 ESE, CNRS FRE 2635, Université de Metz, UFR SciFA, Av. Delestraint, 57070 Metz, France
2 IRH Environnement, 11 bis rue G. Peri, BP 286, 54515 Vandœuvre-les-Nancy, France
3 INSERM U420, Faculté de Médecine, 9 Av. Forêt de Haye, 54505 Vandœuvre-les-Nancy, France
Received 16 December 2003; Accepted 14 May 2004
Abstract – The transfer of polycyclic aromatic hydrocarbons (PAHs) from bivalves, through the food chain to verte-
brates was of concern. Our research aimed at estimating potential e?ects for consumers resulting from the ingestion of
seafood contaminated by oil spill pollutants. After the “Erika” wreck, mussels (Mytilus sp.) were collected from sites
of the Atlantic coast impacted to various degrees by the oil slick and constituted contaminated food for rats during 2
and 4 weeks. Genotoxic damage were studied in rats by means of COMET assay carried out in liver, bone marrow and
peripheral blood. Biochemical and genomic e?ects such as the induction of CYP 1A1 and the expression of cytochrome
genes were measured in rat livers. The most sensitive biological parameter re?ecting the transfer of contaminants via
the food appeared to be DNA breaks studied by means of the COMET assay. Genotoxic damage, observed mainly in
the liver, were rather moderate and remained not persistent. This study underlined the bioavailability of pollutants in
fuel oil contaminated mussels for consumers, and the complexity of the contamination consecutive to the oil spill. The
occurrence of related PAH compounds in addition to non-substituted PAHs in fuel oils and mussels raised the question
of their implication in the registered e?ects.
Key words: Polycyclic aromatic hydrocarbons / Oil spill / COMET assay / DNA damage / EROD activity / CYP 1A /
Food transfer
Résumé – E?ets génotoxiques et au niveau du CYP 1A, consécutifs au transfert des contaminants du pétrole de
la marée noire par voie alimentaire, des moules aux mammifères. Nos préoccupations venaient du transfert des hy-
drocarbures aromatiques polycycliques (HAPs) des bivalves aux mammifères, par la voie alimentaire. Notre recherche
visait à estimer les e?ets potentiels pour des consommateurs résultant de l’ingestion de fruits de mer contaminés par
les polluants de marée noire. Suite au naufrage de l’« Erika », des moules (Mytilus sp.) ont été échantillonnées sur des
stations de la côte atlantique plus ou moins polluées par la marée noire. Des rats ont reçu pendant 2 et 4 semaines des
rations de cette nourriture contaminée. La génotoxicité a été étudiée au moyen du test des comètes au niveau du foie,
de la moelle osseuse et du sang périphérique des rats. Les e?ets biochimiques et génomiques tels que l’induction du
CYP 1A1 et l’expression des gènes du cytochrome, ont été recherchés au niveau du foie. Le paramètre biologique le
plus sensible, re?étant un transfert des contaminants par la voie alimentaire, apparaît être l’étude des cassures à l’ADN
par le test des comètes. Les dommages génotoxiques, observés principalement dans le foie, restent plutôt modérés et
non persistants. Cette étude a souligné la biodisponibilité des polluants dans les moules pour le consommateur, et la
complexité de la contamination engendrée par la marée noire. La présence de composés apparentés aux HAPs, en plus
des HAPs non-substitués dans les ?ouls lourds libérés et dans les moules, soulève la question de leur implication dans
les e?ets observés.
1 Introduction
(at 75 km of Belle-Ile and at 65 km of the Pointe of Penmarch).
20 000 tons of fuel would have been released in the marine
On December 12, 1999, the wreck of the oil tanker
environment. The oil slick spoiled more than 400 km of the
Erika caused the most important pollution by hydrocarbons
French Atlantic coasts. The fuel oil consisted in a type No 2
occurring in France since the oil spill of the Amoco-Cadiz in
heavy fuel, a residue of distillation of crude oil. This petroleum
1978. This Maltese tanker sunk along the coasts of Brittany
fraction contained toxic compounds including PAHs, many
heterocyclic hydrocarbons, sulphur
organic derivatives,
a Corresponding author:
sebastien.lemiere@umail.univ-metz.fr
solvents, toxic metals (molybdenum, vanadium, mercury).

---

304
S. Lemière et al.: Aquat. Living Resour. 17, 303–307 (2004)
The accident had dramatic consequences on the marine fauna
lyophylisation of samples and extraction under warmth and
and birds.
pressure.
As Invertebrates could bioaccumulate polycyclic aromatic
hydrocarbons (PAHs) (Meador et al. 1995; Solé et al. 1996;
Animal treatment:
Law et al. 2002), the transfer of these pollutants through the
Rats were fed with contaminated mussels during two and
food chain to vertebrates was of concern (Nunn et al. 1996).
four weeks. Two main studies were conducted. In the ?rst
Indeed, PAHs were present at no negligible levels in heavy
experiment, rats received one mussel ration of 13.5 g fresh
fuel oils, released in the marine environment after the Erika
weight every two days, and in the second experiment, once
oil spill on the Atlantic French coast.
a day. The experimental design (Fig. 1) corresponded to a reg-
Our research aimed at estimating potential e?ects for con-
ular and important consumption of contaminated food. Five
sumers resulting from the ingestion of seafood contaminated
levels of contamination were studied: around 85, 310, 480,
with the oil pollutants. The genotoxic, mutagenic and carcino-
570, 870 µg total PAHs (sum of the sixteen US-EPA) kg?1 of
genic properties of PAHs, and especially of the congeners with
food d.w. Control rats in both experiments received a standard
?ve and six aromatic rings, are known in vertebrates (IARC
food for laboratory animals. In all cases, the animals (six per
1987). PAHs are also inducers of cytochrome P450 1A1 (CYP
lot) were sacri?ced 3 days after the last meal of contaminated
1A1) in mammals (Nordqvist et al. 1979).
mussels. Mussels (B1) sampled in a non-impacted site on the
Therefore, we studied the e?ects of repeated ingestions
North Brittany coast (Les Saumonards, France), were used as
of oil-contaminated mussels on rats by investigating geno-
additional controls. For biochemical parameters, dissected tis-
toxic damage using single-cell gel electrophoresis (COMET)
sues were immersed in a 5 M guanidium isothiocyanate bu?er
assay (Singh et al. 1988) and by measuring the induction of
to avoid RNA degradation and frozen at ?196 ?C until RNA
ethoxyresoru?n-o-deethylase (EROD) activity (AFNOR 2001;
extraction. Other pieces of tissues were placed in a 50 mM
Burke and Mayer 1974), and the expression of CYP1A1
phosphate bu?er containing 0.15 mM KCl and 20% glycerol,
gene by means of Reverse Transcription – Polymerase Chain
and frozen at ?196 ?C. For the COMET assay, all the collected
Reaction (RT-PCR). One of our objectives was to estab-
tissues (liver and peripheral blood) were kept at ?196 ?C in
lish a threshold of contamination devoid of toxic e?ects for
liquid nitrogen except for intact femurs which were kept at
mammals exposed to PAHs through their food.
?20 ?C before bone marrow extraction. Cellular suspensions
of bone marrow were prepared by scraping and washing the
inside of the femur with phosphate-bu?ered saline solution
2 Material and methods
(Ca++ and Mg++ free, pH 7.4) (PBS), with 10% dimethylsul-
foxide (DMSO) added. For this tissue, suspensions were kept
Food preparation:
at ?196 ?C after progressive freezing (4 hours at ?20 ?C, one
After the Erika wreck, mussels (Mytilus sp.) were collected
night at ?80 ?C and ?196 ?C in liquid nitrogen).
from sites of the Atlantic coast impacted to various degrees by
EROD enzymatic activity measurement:
the oil slick. The samples were obtained from the IFREMER
(Institut Français de Recherche pour l’Exploitation de la Mer)
Each liver was homogenized in phosphate bu?er, with an
Centre in Nantes and kept at ?20 ?C. Lots of mussels were
antiprotease cocktail added, by means of a manual Potter. The
constituted in order to study the e?ects at di?erent levels of
hepatic microsomes were prepared by two di?erential centrifu-
PAH contamination, between 50 and 1000 µg total PAHs kg?1
gations (9000 g for 15 min, and 100 000 g for 60 min at 4 ?C).
dry weight (d.w.). Food batches were prepared by pooling
Microsomal EROD activity was measured according to the
mussels from impacted sites as follows: frozen mussels were
AFNOR Norm NF T 90-385 (2001) using a ?uorimetric mi-
opened, soft tissues were mixed, homogenized in a grinder and
croplate reader. Protein concentration determination was per-
conditioned in pools of 12?15 g (each constituting one meal).
formed according to Bradford (1976).
The 16 PAH congeners listed as priority pollutants by the
Expression of CYP 1A1 gene:
US-EPA for environmental biomonitoring (OFR 1982) were
Total RNAs were extracted from the livers using the
analyzed in mussels after preparation of the rations.
GenElute Mammalian Total RNA kit (Sigma) and reverse-
transcribed using the RevertAidTM M-MuLV reverse tran-
PAH analyses:
scriptase (MBI Fermentas). PCRs were performed in a
The quanti?cation of PAHs in food was performed as fol-
MJ Research PTC-100 thermal cycler (Global Medical Instru-
lows. Samples were dried at 40 ?C until constant mass. Extrac-
mentation Inc.). The PCR reactions were conducted on cDNA
tion of PAHs was conducted by ASE extractor (Accelerated
from control or treated animals using primers chosen in highly
Solvent Extractor 200, from DIONEX, Sunnyvale, USA), al-
conserved coding sequences of rat CYP1A1 gene. PCR am-
lowing a solid/liquid extraction at 100 ?C and under a pressure
pli?cation conditions were described in Chaty (2003). Am-
of 136 bars, with dichloromethane and hexane as extraction
plimers were separated by electrophoresis on 1.5% agarose
solvents (ratio 50:50; v:v). After solvent evaporation, extracts
gel in TAE bu?er (TRIS 40 mM, acetic acid 1 mM, EDTA
were dissolved in acetonitrile. PAHs were separated by re-
40 mM); the gels were stained with ethidium bromide and the
versed high performance liquid chromatography (HPLC) and
PCR products were visualized under UV light. The intensity of
detected by a spectro?uorimeter coupled with a UV/diode ar-
PCR products was quanti?ed with Molecular AnalystTM (ver-
ray detector. In addition, the methylated and related PAHs,
sion 1.1.1., Biorad). The statistical analyses for EROD activity
such as thiophenes, were measured in mussels by gas chro-
measurement and CYP1A1 expression consisted in two-tailed
matography coupled to mass spectrometry (GC-MS) after
Mann-Whitney U tests.

---

S. Lemière et al.: Aquat. Living Resour. 17, 303–307 (2004)
305
Fig. 1. Experimental design.
Alkaline single cell gel electrophoresis (COMET) assay:
DNA damage were studied by means of the COMET assay
as described by Singh et al. (1988) with minor modi?cations.
Brie?y, the assays were performed under inactinic light at 4 ?C
and in triplicate for each sample. The cellular suspensions
were included between two layers of agarose on microscope
slides and submitted to lysis for 2 h at 4 ?C. Thereafter, slides
were placed in an electrophoresis tanker, covered with the elec-
trophoresis bu?er (300 mM NaOH, 1 mM Na2EDTA) to allow
DNA unwinding and then electrophoresed (300 mA, 21 min,
20 V). The slides were batched twice in the neutralizing bu?er
(0.4 M Tris, pH 7.5) before the dehydratation step in abso-
lute ethanol for 10 min. Nuclear DNA was stained with 25 µl
Fig. 2. Individual average nuclei of hepatocytes visualized by the
ethidium bromide (20 µg ml?1) and observed using an epi?uo-
COMET assay in livers of rats fed during 15 days with either standard
rescence microscope (BX60 Olympus) connected to an image
food (control) or mussels sampled in contaminated areas (second ex-
analysis system (Komet 3:1, Kinetic Imaging Ltd). The per-
periment, 6 animals per treatment); Averaged percentages of DNA in
centage of DNA in the tail was chosen to express DNA damage
the tail of nuclei with standard deviation.
and non-parametric statistical tests (the Kruskal-Wallis test,
the Median test and the Chi-square test) were used to compare
the results.
contaminated mussels, even heavily contaminated mussels
(870 µg total PAHs kg?1 of mussels dry weight).
2. Genotoxic e?ects were registered in the liver and the bone
3 Results
marrow, 3 days after the last ingestion (Fig. 2). The geno-
Those studies with repeated meals of contaminated food
toxicity was more persistent than the biochemical e?ects.
The intensity of DNA damage increased with mussel con-
on 15 and 30 days, showed that (Table 1):
tamination level but remained moderate. The threshold
1. The biochemical e?ects (induction and the expression of
of contamination of the mussels from which a genotox-
the gene of the CYP 1A1) were transient: a kinetic study
icity was observed, was 310 µg total PAHs kg?1 mussels
showed that the induction was maximum in the 12 hours
(dry weight), below the threshold recommended by
following the ingestion and was no more detectable af-
AFSSA (500 µg total PAHs kg?1 mussels (dry weight))
ter 3 days. In the 2 and 4 weeks experiments, no induc-
(AFSSA 2000). Yet, the genotoxicity did not increase over
tion was observed in rats 3 days after the last ingestion,
time, beyond 15 days. Indeed, e?ects after 30 days of ex-
although rats had been fed during 15 and 30 days with
posure with fuel oil contaminated food were equivalent to

---

306
S. Lemière et al.: Aquat. Living Resour. 17, 303–307 (2004)
Table 1. Synthesis of the biological e?ects (EROD activity (liver), levels of CYP 1A1 messenger RNA (by RT-PCR) (liver) and DNA damage
(COMET assay) (hepatocytes, bone marrow cells and peripheral blood nucleated cells) studied in rats fed with mussels sampled in areas im-
pacted to various degrees by the oil slick, during 15 or 30 days: relative scale of increasing e?ects (abbreviations: d.w. dry weight; + signi?cant
e?ect).
those measured after 15 days. This con?rmed the fast elim-
other derivatives, and their implications in registered e?ects re-
ination of PAHs in mammals, and the capacities of verte-
mained to de?ne, as well as the e?ects of nickel and vanadium
brates to repair DNA. In the present case, DNA repairing
likely to contribute to genotoxicity.
systems appeared unsaturated after 30 days, as showed by
Other PAH derivatives, such as methyl chrysene and
the return to normality in all cases within 14 days. Re-
thiophenes found in mussels, could also be responsible for
covery of DNA integrity was even registered after oral
the observed e?ects. The methylated PAH sum could reach
exposure with mussels containing 570 µg kg?1 d.w. (total
1900 µg kg?1 d.w. in the mussels A3, and should be considered
PAHs).
for their genotoxic e?ects. Few studies had been conducted to
study the genotoxic potentials of these compounds. Their gen-
eral toxicity was poorly known although their co-occurrence
with PAHs in the environment may be suspected. It would be
4 Discussion
advised to study their toxic potentialities and considered them
in regulations. Borosky (1999) underlined that a methyl sub-
Marine mussels had often been used to follow the impact of
stitution could lead to an increase in carcinogenicity for PAHs.
oil slick pollutions (Solé et al. 1996; Glegg et al. 1999; Harvey
In addition, 5-methyl chrysene was classi?ed as a carcinogen
et al. 1999; Thomas et al. 1999a,b; Fernley et al. 2000), but had
2B by IARC whereas chrysene was only in the class 3.
never been studied as a vector of contamination to mammals.
The thiophene compounds might possess mutagen proper-
One interest of our work was to study samples representa-
ties similar to benzo(a)pyrene (Sinsheimer et al. 1992). Studies
tive of the environmental pollution by PAHs and not arti?cially
of Poon et al. (1997, 1998), showed histological abnormalities
polluted samples.
and hepatoxic e?ects in the rat consecutive to a contaminated
This work raised questions about the biological conse-
diet with benzothiophene.
quences of DNA damage, which may modify the expression
Gilroy (2000) underlined the di?culty of the health risk
of crucial genes.
evaluation from contaminated seafood following an oil spill.
Moreover, the repair process itself may increase the proba-
Their work highlighted the necessity of standardized protocols
bility of secondary mutations (Christmann et al. 2003; Sarasin
for this concern. In our study, the most sensitive biological pa-
2003), the latter being undetectable with the COMET assay.
rameter re?ecting the PAH transfer via the food appeared to
Finally, the questions of the genotoxic properties of methy-
be DNA breaks studied with the COMET assay. The CYP1A1
lated PAHs, PAHs related compounds such as thiophenes, and
induction studied by means of EROD activity appeared to be

---

S. Lemière et al.: Aquat. Living Resour. 17, 303–307 (2004)
307
more sensitive than the measurement of gene expression by
Glegg G.A., Hickman L., Rowland S.J., 1999, Contamination of
RT-PCR, but responses of these parameters were transitory.
limpets (Patella vulgata) following the “Sea Empress” oil spill.
Mar. Pollut. Bull. 38, 119-125.
Harvey J.S., Lyons B.P., Page T.S., Stewart C., Parry J.M., 1999, An
5 Conclusion
assessment of the genotoxic impact of the “Sea Empress” oil spill
by the measurement of DNA adduct levels in selected invertebrate
The DNA breaks studied with the COMET assay appeared
and vertebrate species. Mutat. Res. 441, 103-114.
the most sensitive biological parameter to study the impact
IARC, 1987, Monogaph on the evaluation of carcinogenic risks of
of food PAH contamination in mammals. Genotoxic damage
chemicals to humans. Overall evaluations of carcinogenicity:
were observed mainly in the liver. DNA injuries were rather
An updating of IARC Monographs, Vol. 1 to 42. International
moderate and not persistent. The threshold of mussel contami-
Agency for Research on Cancer, Lyon, France.
nation below which no genotoxicity was recorded, was 310 µg
Law R.J., Kelly C.A., Baker K.L., Langford K.H., Bartlett T., 2002,
total PAHs kg?1 mussels (dry weight). The measurement of the
Polycyclic aromatic hydrocarbons in sediments, mussels and
EROD activity seemed more sensitive than the study of the
crustacea around a former gasworks site in Shoreham-by-Sea,
genic expression to investigate the CYP 1A1 induction.
UK. Mar. Pollut. Bull. 44, 903-911.
This study underlined the bioavailability of pollutants in
Meador J.P., Stein J.E., Reichert W.L., Varanasi U., 1995,
fuel oil contaminated mussels for consumers, and the com-
Bioaccumulation of polycyclic aromatic hydrocarbons by marine
organisms. Rev. Environ. Contam. Toxicol. 143, 79-165.
plexity of the contamination of the invertebrates consecutive to
Nordqvist M., Thakker D.R., Levin W., Yagi H., Ryan D.E., Thomas
the Erika oil slick. The occurrence of substituted PAHs and re-
P.E., Cooney A.H., Jerina D.M., 1979, The highly tumori-
lated compounds such as benzothiophenes in addition to non-
genic 3,4-dihydrodiol is a principal metabolite formed from
substituted PAHs in fuel oils and mussels raised the question
dibenzo(a,h) anthracene by liver enzymes. Mol. Pharmacol. 16,
of their implication in the registered e?ects.
643-655.
Nunn J.W., Livingstone D.R., Chipman J.K., 1996, E?ect of ge-
Acknowledgements. This research was supported by the French
netic toxicants in aquatic organisms. In: Taylor E.W. (Eds),
Ministry of Ecology and Sustainable Development. We thank
Toxicology and Aquatic Pollution: Physiological, Molecular
IFREMER teams for their collaboration in providing us mussels.
and Cellular Approaches 57, Cambridge University Press,
Cambridge, pp. 225–251.
O?ce of the Federal Register (OFR), 1982, Appendix A: Priority
Pollutants. Fed. Reg. 47, 52309.
References
Poon R., Davis H., Lecavalier P., Liteplo R., Yagminas A., Chu I.,
Bihun C.J., 1997, E?ects of benzothiophene on male rats fol-
AFNOR, 2001, Qualité de l’eau. Méthode de mesure de l’activité en-
lowing short-term oral exposure. J. Toxicol. Environ. Health 50,
zymatique EROD chez le poisson d’eaux douces et marines. NF
53-65.
T 90-385, Association Française de Normalisation, Paris, France.
Poon R., Davis H., Park G., Lecavalier P., Yagminas A., Elwin J.,
AFSSA, 2000, Avis du groupe d’experts réunis par l’Agence
Chu I., Valli V.E., 1998, Subchronic toxicity of benzothiophene
française de sécurité sanitaire des aliments sur les critères de
on rats following dietary exposure. J. Toxicol. Environ. Health A
toxicité alimentaire présentés par la pollution engendrée par le
55, 133-150.
naufrage de l’Erika. Agence Française de Sécurité Sanitaire des
Sarasin A., 2003, An overview of the mechanisms of mutagenesis and
Aliments, Maisons-Alfort, France.
carcinogenesis. Mutat. Res. 544, 99-106.
Bradford M.M., 1976, A rapid and sensitive method for the quantita-
Singh N.P., McCoy M.T., Tice R.R., Schneider E.L., 1988, A sim-
tion of microgram quantities of protein utilizing the principle of
ple technique for quantitation of low levels of DNA damage in
protein-dye-binding. Anal. Biochem. 72, 248-254.
individual cells. Exp. Cell Res. 175, 184-191.
Borosky G.L., 1999, Theoretical study related to the carcinogenic
Sinsheimer J.E., Hooberman B.H., Das S.K., Savla P.M., Ashe A.J.,
activity of polycyclic aromatic hydrocarbon derivatives. J. Org.
1992, Genotoxicity of chryseno[4,5-bcd] thiophene and its sul-
Chem. 64, 7738-7744.
fone derivative. Environ. Mol. Mutagen. 19, 259-264.
Burke M.D., Mayer R.T., 1974, Ethoxyresoru?n: direct ?uorimet-
Solé M., Porte C., Biosca X., Mitchelmore C.L., Chipman J.K.,
ric assay of a microsomal O-dealkylate which is preferentially
Livingstone D.R., Albaigés J., 1996, E?ects of the “Aegean
inducible by 3-methylcholanthrene. Drug Metab. Dispos. 2,
Sea” oil spill on biotransformation enzymes, oxidative stress and
583-588.
DNA-adducts in digestive gland of the mussel (Mytilus edulis L.).
Chaty S., 2003, E?ets d’inducteurs-types sur l’expression du cy-
Comp. Biochem. Physiol. C, Pharmacol. Toxicol. Endocrinol.
tochrome P450 chez les bivalves après exposition directe et chez
113, 257-265.
les mammifères contaminés par la voie alimentaire. Thèse de
Thomas R.E., Brodersen C., Carls M.G., Babcock M., Rice S.D.,
doctorat, ESE, Toxicologie, Université de Metz.
1999a, Lack of physiological responses to hydrocarbon accu-
Christmann M., Tomicic M.J., Roos W.P., Kaina B., 2003,
mulation by Mytilus trossulus after 3–4 years chronic exposure
Mechanisms of human DNA repair: an update. Toxicology 193,
to spilled “Exxon Valdez” crude oil in Prince William Sound.
3-34.
Comp. Biochem. Physiol. C, Pharmacol. Toxicol. Endocrinol.
Fernley P.W., Moore M.N., Lowe D.M., Donkin P., Evans S., 2000,
122, 153-163.
Impact of the “Sea Empress” oil spill on lysosomal stability in
Thomas R.E., Harris P.M., Rice S.D., 1999b, Survival in air of Mytilus
mussel blood cells. Mar. Environ. Res. 50, 451-455.
trossulus following long-term exposure to spilled “Exxon Valdez”
Gilroy D.J., 2000, Derivation of shell?sh harvest reopening criteria
crude oil in Prince William sound. Comp. Biochem. Physiol. C,
following the “New Carissa” oil spill in Coos Bay, Oregon. J.
Pharmacol. Toxicol. Endocrinol. 122, 147-152.
Toxicol. Environ. Health A 60, 317-329.

---