Environmental Science and Pollution Research

https://doi.org/10.1007/s11356-019-04220-3

MULTI-STRESSORS IN FRESHWATER AND TRANSITIONAL ENVIRONMENTS: FROM LEGACY

POLLUTANTS TO EMERGING ONES

Occurrence, sources and effects of polycyclic aromatic hydrocarbons

in the Tunis lagoon, Tunisia: an integrated approach using multi-level
biological responses in Ruditapes decussatus
Houssem Chalghmi 1,2 & Jean-Paul Bourdineaud 1 & Ikram Chbani 3 & Zohra Haouas 4 & Saida Bouzid 3 &
Hassan Er-Raioui 3 & Dalila Saidane-Mosbahi 2

Received: 22 June 2018 / Accepted: 10 January 2019

# Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract
Coastal lagoons are critical ecosystems presenting a strategic economic importance, but they are subjected to potential anthropo-
genic impact. As part of the Tunis lagoon (Tunisia) biomonitoring study, levels, composition pattern and sources of polycyclic
aromatic hydrocarbons (PAHs) in surface sediments along with their bioavailability in clam Ruditapes decussatus were investigated
in polluted (S2–S4) and reference (S1) sites. In order to investigate the contamination effects at different biological levels in clams, a
wide set of biomarkers, including gene expression changes, enzymatic activities disruption and histopathological alterations, was
analysed. Biomarkers were integrated in a biomarker index (IBR index) to allow a global assessment of the biological response.
Principal component analysis (PCA) was used for chemical and biological data integration to rank the sampling sites according to
their global environmental quality. Sediment PAHs levels ranged between 144.5 and 3887.0 ng g−1 dw in the Tunis lagoon sites
versus 92.6 ng g−1 dw in the reference site. The high PAH concentrations are due to anthropogenic activities around the lagoon.
PAH composition profiles and diagnostic isomer ratios analysis indicated that PAHs were of both pyrolitic and petrogenic origins.
Clams sampled from S2 and S3 exhibited the highest PAH contents with 2192.6 ng g−1 dw and 2371.4 ng g−1 dw, respectively.
Elevated levels of tissue PAHs were associated to an increase in biotransformation and antioxidant activities, and lipid peroxidation
levels along with an overexpression of different genes encoding for general stress response, mitochondrial metabolism and
antioxidant defence, in addition to the emergence of severe and diverse histopathological alterations in the clams’ digestive glands.
IBR index was suitable for sampling sites ranking (S1 = 0 < S4 = 0.4 < S3 = 1.15 < S2 = 1.27) based on the level of PAH-induced
stress in clams. PCA approach produced two components (PC1, 83.8% and PC2, 12.2%) that describe 96% of the variance in the
data and thus highlighted the importance of integrating contaminants in sediments, their bioaccumulation and a battery of bio-
markers of different dimensions for the assessment of global health status of coastal and lagoon areas.

Keywords Tunis lagoon . Biomonitoring . Ruditapes decussatus . Polycyclic aromatic hydrocarbons . Multivariate analysis .
Multi-level approach . Integrated biomarker response

Responsible editor: Cinta Porte

Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s11356-019-04220-3) contains supplementary
material, which is available to authorized users.

* Houssem Chalghmi 3
Laboratory of Environment, Oceanology and Natural Resources,
[email protected] Faculty of Sciences and Technology, University of Abdelmalek
Essaâdi, B.P. 416 Tangier, Morocco
1
UMR CNRS 5805 EPOC, University of Bordeaux, Arcachon Marine
Station, Place du Dr Peyneau, 33120 Arcachon, France
2
Laboratory of Analysis Treatment and Valorization of Environmental
4
Pollutants and Products, Faculty of Pharmacy, University of Laboratory of Histology Cytology and Genetics, Faculty of Medicine,
Monastir, 5000 Monastir, Tunisia University of Monastir, 5019 Monastir, Tunisia
 Environ Sci Pollut Res

Introduction including molecular, biochemical and histopathological bio-

markers that when integrated, can clarify issues of contami-
Coastal lagoons are particular ecosystems presenting a region- nant bioavailability, bioaccumulation and ecological effects
al economic importance because they have high biological while enabling a better understanding of the effects of
productivity and provide harbour and navigation area xenobiotics.
(Pérez-Ruzafa et al. 2011). Despite their economic impor- The challenge when using multi-biomarker approach is to
tance, these ecosystems are highly threatened and modified evaluate and interpret the global response of biomarker set
by human impact (Chalghmi et al. 2016a; Capolupo et al. instead of interpreting each of them independently. The global
2017). Located in northern Tunisia, Tunis lagoon is a assessment of biomarker battery facilitates their interpretation
Mediterranean coastal lagoon presenting high strategic eco- in monitoring programs (Devin et al. 2017). The integrated
nomic importance. This lagoon is situated between the capital biomarker response (IBR) was proposed by Beliaeff and
city of Tunis and the Gulf of Tunis and consists of North Burgeot (2002) to integrate individual biomarker responses.
lagoon, South lagoon and a navigation canal connecting the The IBR approach combines all measured biomarkers into
Mediterranean Sea to the Rades (freight traffic) and Goulette one numerical value to rank the level of pollutant-induced
(fishing and passenger traffic) harbours. The neighbouring stress in the organisms and provides a graphical synthesis of
Tunis city is characterised by a variety of activities related to the different biomarker responses.
tourism, agriculture, urban agglomerations and petrochemical This study was performed as part of a more extensive sur-
industries. The lagoon drains a watershed of 40 km2, includ- vey into the use of a global approach combining chemical and
ing 15 km2 occupied by industrial areas. It thus receives ap- biological parameters to assess the health status of the Tunis
proximately 5500 m3 day−1 of industrial wastewater rich in lagoon. It provides the first database referring to PAH contam-
organic and metal contaminants (Mzoughi and Chouba 2011; ination and their biological effects in this ecosystem. The main
Chalghmi et al. 2016c). objectives of the present work were (i) to determine the dis-
Polycyclic aromatic hydrocarbons (PAHs) are among the tribution, composition and sources of PAHs in the sediments
most ubiquitous contaminants in marine environments of Tunis lagoon, (ii) to assess the bioavailability of PAHs
(Tuncel and Topal 2015; Zaghden et al. 2017; Cacciatore through their quantification in tissues of clam Ruditapes
et al. 2018). PAH contaminants are mainly derived from an- decussatus and (iii) to evaluate the biological effects of
thropogenic sources, such as petrogenic and pyrolitic sources PAHs on clams using a set of biomarkers reflecting specific
(Zrafi et al. 2013; Bi et al. 2018). Occurrence of anthropogenic responses at different levels of biological organisation (molec-
PAHs in the aquatic environments arises from oil spill ular, biochemical and tissular). In addition, biomarker re-
(petrogenic) and/or from incomplete combustion of fossil sponses were integrated into the IBR index to rank the level
fuels and biomass (pyrolitic) (Wurl and Obbard 2004; of PAH-induced stress in clams, and all the chemical and
Zaghden et al. 2017). These organic compounds are biological data were further processed by the principal com-
characterised by their high hydrophobicity and their strong ponent analysis (PCA) to rank sampling sites according to
affinity for organic matter, thus favouring their accumulation their global environmental quality.
in fine-grained sediments (Culotta et al. 2006). PAHs are con-
sidered very harmful for organisms due to their toxicity, car-
cinogenicity and mutagenicity (Shaw and Connell 1994; Lu Material and methods
et al. 2012; Kim et al. 2013).
Organisms facing environmental stress respond at all levels Sampling sites
of biological organisation. Molecular and biochemical re-
sponses are specific, sensitive, reproducible and easier to de- Sediments and clams were collected from three sampling sites
termine but difficult to relate to ecological changes. Whereas, in the Tunis lagoon according to the location of contamination
histological responses can be related to fitness of individuals, sources in the area (Fig. 1). The reference site is located in
which, in turn, allows further extrapolation to population/ Louza region (rural area) which was considered as not
community effects (Au 2004). Consequently, it is very useful modified/threatened site by anthropogenic contaminants
and recommended to integrate several biomarkers at different (Banni et al. 2009; Chalghmi et al. 2016a, b). Coordinates,
functional complexity in environmental monitoring programs contamination sources and water physicochemical parameters
to provide both an early warning of contamination effects and for each sampling site are reported in Table S1.
ecological relevance (Nicholson and Lam 2005). Unlike our About 100 R. decussatus clams (32.2 ± 1.8 mm) were
previous in situ studies which were based on assessing organic collected at the sampling sites (S1–S4) in April. The clams
and inorganic pollution effects on clams using only biochem- were immediately transported to the laboratory, measured and
ical biomarkers (Chalghmi et al. 2015, 2016b, c), in the pres- dissected. The clams were dissected for excision of the soft
ent work, we investigated the multi-level biological response tissue for PAH determination and the digestive gland for
Environ Sci Pollut Res

MEDITERRANEAN SEA
TUNIS

LOUZA

TUNISIA
Gulf of
Tunis lagoon Tunis

1 km

La Goulee
harbour
S3
LOUZA (S1)
Industrial
S4
Rades area
SFAX
harbour
S2 Gulf of
25 km Tunis 250 m
Power plant

Fig. 1 Map of the study area and location of sampling sites in Tunis lagoon (S2–S4) and Louza (reference site, S1)

biomarker analysis. The tissues were stored at − 80 °C for the The PAH fraction was analysed by gas chromatography–
analysis. The soft bodies of clams were then pooled and mass spectrometry (GC–MS) in SIM mode, using a HP 6890
freeze-dried for PAH measurement. gas chromatograph coupled with a HP 5973 MSD mass spec-
Surface sediments were collected at the sampling sites dur- trometer (Agilent Technologies, Wilmington, DE, USA),
ing the same period. Before PAH analysis, the sediments were equipped with a DB-5 MS fused-silica capillary column coat-
homogenised, freeze-dried and passed through a stainless steel ed with 5% phenyl methyl siloxane (30 m × 0.25 mm × 0.25
sieve (100 μm). μm, J&W Scientific, Agilent Technologies, USA). Helium
was used as a carrier gas at a flow rate of 1.4 mL min−1. The
Quantification of PAHs injector was maintained at 250 °C. The initial column temper-
ature was held for 2 min at 60 °C, next ramped at 25 °C min−1
PAHs were extracted according to the method described by to 100 °C and then at 2 °C min−1 to a final temperature of 310
Er-Raioui et al. (2009) with some modifications. The extrac- °C, which was held for 10 min. The PAHs analysed in this
tion of hydrocarbon compounds was carried out in a Soxhlet study are the 16 priority compounds recommended by the U.
extractor. An amount of 20 g of sediments or 5 g of clam soft S. Environmental Protection Agency (EPA): naphthalene
tissues (n = 3) were extracted with chloroform for 12 h. (Nap), acenaphthylene (Acy), acenaphthene (Ace), fluorene
Deuterated internal standards (p-terphenyl-d14) were added (Flu), phenanthrene (Phe), anthracene (Ant), fluoranthene
before the extraction. Saponification process was added for (Flt), pyrene (Pyr), benzo(a)anthracene (BaA), chrysene
tissue samples’ extraction to allow hydrocarbons liberation (Chr), benzo(b)fluoranthene (BbF), benzo(k)fluoranthene
from lipids. The saponification was carried out with KOH/ (BkF), benzo(a)pyrene (BaP), dibenzo(a,h)anthracene
distilled water (0.7 N) for 2 h. Then, a liquid/liquid extraction (DaA), indeno(1,2,3-cd)pyrene (IcP), and
was performed with n-hexane three times. The extract is there- benzo(g,h,i)perylene (BgP).
after concentrated by a rotary evaporator. The extract was All data were subjected to quality and control procedures.
purified and separated into non-aromatic hydrocarbons Deuterated internal standard mixtures (p-terphenyl-d14) were
(NAH) and PAHs, using chromatography column on neutral introduced before Soxhlet extraction to assess the recoveries
alumina/silica for organisms and silica for sediments. The elu- of the analytical procedure (extraction, evaporation and puri-
tion was done by using 5 mL of n-hexane and 5 mL of n- fication processes). Recoveries above 80% were accepted.
hexane/chloroform (2:1, v/v) for NAH and PAH fractions, Calibration external standards (36991, Sigma-Aldrich) as well
respectively (Zrafi-Nouira et al. 2008). as procedural and solvent blanks were run with each set of
 Environ Sci Pollut Res

samples to check for contamination and for quantification. measured at 532 nm using thiobarbituric acid (TBA) reagent
Detection limits for individual PAHs varied from 0.01 to and 1,1,3,3-tetramethoxypropane (standard) according to the
0.5 ng g−1 dry weight (dw). Results were expressed in nano- method described by Buege and Aust (1978). Results were
grams per gram dry weight. expressed as nanomoles equivalent MDA per milligram
proteins.
Biochemical analysis
Transcriptional analysis
Digestive glands of 30 clams were pooled into ten pools for
each sampling site (n = 10) and were homogenised in phos- Five clams from each site were analysed individually for tran-
phate buffer (0.1 M, pH 7.5). The homogenate obtained was scriptional analysis (n = 5). Digestive gland tissue was hold in
centrifuged at 9000g for 25 min at 4 °C for the cytosolic a RNA preserving solution (RNA Later; Sigma-Aldrich) at 4
fractions (S9). The obtained supernatant (S9) was split into °C for overnight, before to be frozen at − 80 °C. Total RNA
two tubes (T1 and T2). The tube T1 was used to perform from digestive gland was extracted using Absolutely RNA
glutathione S-transferase (GST), catalase (CAT) and Miniprep Kit (Agilent technologies), reverse transcribed to
malondialdehyde (MDA) analysis. cDNA using AffinityScript cDNA Synthesis Kit (Agilent
The S9 fraction in tube T2 was centrifuged at 100000g for technologies) and amplified in a Stratagene Mx3000P
25 min at 4 °C, and the resulting pellet was resuspended in the thermocycler (Agilent technologies) for gene expression anal-
phosphate buffer to obtain the microsomal fractions (P100). ysis, according to the method detailed in our previous work
The P100 fraction was used for benzo(a)pyrene hydroxylase (Chalghmi et al. 2016a). The specific primer pairs used for the
(BPH) analysis. The quantities of proteins present in S9 and RT-qPCR amplifications are listed in Table S2. The relative
P100 fractions were determined according to the Bradford expression of target genes (RGE) was calculated according to
(1976) method using Coomassie Blue reagent (BioRad). Pfaffl (2001) method. Relative expression data were normal-
BPH activity was measured in the P100 fractions by the ized to the housekeeping gene ribosomal RNA 18S. A differ-
fluorometric method of Michel et al. (1993) adapted to a mi- ential expression factor (DEF) for each gene was calculated
croplate reader. Following incubation with B(a) P, the reaction according to the formula:
was stopped by the addition of 10% TritonX-100.
Fluorescence of the sample was obtained by difference in RGE ðcontaminated siteÞ
DEF ¼
fluorescence between the respective emission/excitation RGE ðreference siteÞ
wavelengths of 492/430 and 510/430 nm. The 3-OHB(a)P
was used as internal standard in all samples to control for
quenching. Results were expressed in picomoles per minute Histological analysis
per milligram proteins.
GST activity was analysed in S9 fractions according to the The digestive glands of ten clams were fixed in Bouin’s fixa-
method of Habig et al. (1974), using 10 μg of cytosolic pro- tive for 48 h prior to histological analysis (n = 10). The fixed
teins, 1 mM 1-chloro-2,4-dinitrobenzene (CDNB) (Sigma- digestive glands were then dehydrated in graded ethanol so-
Aldrich, St Louis, MO, USA) and 4 mM reduced glutathione lution and embedded in paraffin. Embedded tissues were cut
(GSH), in 100 mM sodium phosphate buffer (pH 7.5). GST into sections of 5 μm thickness by a rotary microtome (Leitz
activity was evaluated by kinetic measurement at 25 °C using WETZLAR 1512). The thin sections of the digestive gland
a spectro UV-VIS double beam PC scanning spectrophotom- tissues were stained by haematoxylin-eosin or Masson’s
eter UVD-2960 (λ = 340 nm). Results were expressed as Trichrome for observation by the light microscope. Sections
micromoles GS-CDNB produced per minute and per milli- were photographed with a microscope (LEICA DM 750)
gram proteins. equipped with a numerical camera (LEICA ICC50 HD). For
CAT activity was measured according to Claiborne (1985) each individual digestive gland, seven slides were examined
method in a reaction mixture containing 0.78 mL of phosphate with four sections per slide. Nine alterations were found in the
buffer (0.1 M, pH 7.5) and 0.2 mL of H2O2 (0.5 mM). After clams’ digestive glands.
30 s pre-incubation, the reaction was started by adding 0.02 A semi-quantitative evaluation of the histopathological
mL of S9 fraction. CAT activity was determined by kinetic lesions was performed according to Riba et al. (2004). The
measurement at 25 °C using a spectro UV-VIS double beam severity of alterations was ranked as follow: grades 0 (−),
PC scanning spectrophotometer UVD-2960 (λ = 240 nm). 0.5 (+/−), 1 (+), 2 (++) and 3 (+++). An index of damages
Results were expressed as micromoles hydrogen peroxide (ID) was established for the comparison of histopatholog-
transformed per minute and per milligram proteins. ical responses between sampling sites. An average value
Lipid peroxidation (LPO) was analysed in terms of thio- was obtained from the original semi-quantitative assess-
barbituric acid reactive species (TBARS). The reaction was ment of the lesions.
Environ Sci Pollut Res

Integrated biomarker response while the lower concentration was recorded in S4, which
could be attributed to discharges from the urban agglomera-
The IBR index was calculated as described in Beliaeff and tion, industries and harbours. Baumard et al. (1998a) ranked
Burgeot (2002) and modified by Broeg and Lehtonen (2006) the PAH contamination in sediments into several levels, in-
to assess clam health status. In the IBR calculation, bio- cluding low (0–100 ng g−1), moderate (100–1000 ng g−1),
markers were ranged according to their hierarchy of biological high (1000–5000 ng g−1) and very high (> 5000 ng g−1) con-
organisation from the subcellular to the tissular level (cox1, tamination. According to this classification, the Tunis lagoon
16SrRNA, sod, hsp70, mt, BPH, GST, CAT, MDA and ID), could be considered moderately (S2 and S4) to highly (S3)
following the method described in Serafim et al. (2012). The PAH-contaminated. Meanwhile, site S1 presented a low con-
IBR index was computed for each site as follows: individual tamination level, approving its selection as a reference site.
areas Ai connecting the ith and the (i + 1)th radius coordinates The PAH pollution in the Tunis lagoon could be the result of
of the star plot were obtained according to the following for- various anthropogenic activities present around/in the site
mula: such as urban activities (industrial and domestic effluents),
  power generation, harbour activities, shipping and urban traf-
1 2π
Ai ¼ −sin SiSi þ 1 fic through dry and wet deposition processes of atmospheric
2 n PAHs derived from the vehicle emissions.
Anthropogenic PAHs that have reached the aquatic envi-
where Si and Si + 1 represent the individual biomarker scores
ronment could have different origins (pyrolitic and
and their successive star plot radius coordinates, and n repre-
petrogenic). Analysis of PAH profiles and their diagnostic
sents the number of radii corresponding to the biomarkers
ratios was performed to discriminate between these sources.
used in this study. The final IBR index was obtained by divid-
The proportional distribution of PAHs based on the number of
ing the IBR value by the number of investigated biomarkers
aromatic rings in the sediments is presented in Fig. S1. All
and it was expressed as IBR/n (Broeg and Lehtonen 2006).
sampling sites present low amounts (1.8–20.8%) for low mo-
lecular weight (LMW) PAHs (2–3 rings) and high amounts
Statistical analysis (79.2–98.2%) for high molecular weight (HMW) PAHs (4–6
rings). Pyrolitic PAHs are characterised by HMW compounds,
The data was tested for normality and homogeneity of vari- and they are generated by incomplete combustion of organic
ance. Data statistical analysis was performed using one-way matter (e.g. coal and wood fossil fuels), while petrogenic
analysis of variance (ANOVA) and Duncan’s test for multiple PAHs are generally of LMW (Rajpara et al. 2017). Thus, the
range comparison; p < 0.05 was considered as significant. dominance of HMW PAHs in our sampling sites is indicative
Pearson’s correlation analysis was carried out to find relation- of contamination by pyrolitic sources. Naphthalene is a 2-ring
ships between tissue PAH concentrations and biomarkers. PAH that is abundant in crude oils and petroleum products
SigmaStat 3.5 (SYSTAT Software, Inc) was used for statistical with lighter fractions, and was considered as an indicator of
analysis. A principal component analysis was applied to the p e t r o g e n i c s o u r c e s o f e n v i r o n m e n t a l PA H s
mean values of environmental parameters, PAH concentra- (Bemanikharanagh et al. 2017). The presence of naphthalene
tions and biomarkers obtained from all sites to evaluate the in the surface sediments of sampling sites suggests a recent
relationships between variables and the relative influence of petroleum inputs. Comparing the three Tunis lagoon sites, S4
different parameters in the overall results, using XLSTAT® presented the higher amount of LMW PAHs suggesting a
v7.5.2. higher petroleum contamination due to the Rades harbour
activities, the oil leak and the remove of engine oil from boats.
The results showed a decrease in the amount of 5–6-ring
Results and discussion PAHs in the surface sediments as we move away from the
electric power plant located near the site S2 (S2 (64.3%) >
Distribution, composition pattern and sources S3 (48.8%) > S4 (38.7%)). This suggests that the electric
identification of PAHs in sediments power plant is the main source of the pyrolitic PAHs present
in the Tunis lagoon. On the other side, an increase of 4-ring
The Total PAH concentrations in surface sediments of the PAHs was observed by getting closer to S4 (S4 (53.8%) > S3
Tunis lagoon varied between 144.5 ± 10.3 and 3887 ± (49.4%) > S2 (32.9%)). This may be related to petroleum
105 ng g−1 versus a concentration of 92.6 ± 13.4 ng g−1 re- combustion, particularly the fuel combustion by boats. The
corded in the reference site (Table 1). Thus, lagoon sites pre- PAH profile in the reference site (S1) is characterised by high
sented higher PAHs concentrations than the reference site contribution of 4-ring PAHs (74.3%) followed by 2–3-ring
(S1). In the lagoon, the distribution pattern in sediment sam- PAHs (20.8%) and low contribution of 5–6-ring PAHs
ples was found to vary; S3 presented the higher concentration (4.9%). Such PAH profile may be explained by the presence
 Environ Sci Pollut Res

Table 1 Polycyclic aromatic

hydrocarbon concentrations (ng Sediments Clams
g−1 dw) in surface sediments and
tissues of clams collected from S1 S2 S3 S4 S1 S2 S3 S4
Tunis lagoon (S2–S4) and
reference (S1) sites Naphthalene 16.9 2.7 6.1 9.1 4.5 < dl < dl < dl
Acenaphthylene < dl 1.6 12.5 0.5 5.3 < dl < dl < dl
Acenaphthene < dl < dl < dl 0.4 3.5 121.6 93.2 53.7
Fluorene 0.7 1.7 0.1 < dl 7.6 19.5 70.2 24
Phenanthrene 1.4 1.7 43.3 0.4 3.7 < dl < dl < dl
Anthracene 0.2 2.6 7.4 0.5 7.4 < dl < dl < dl
Fluoranthene 18.3 26.5 707.3 22.2 269.2 739.1 334.0 381.2
Pyrene 20.8 4.4 391 21.2 < dl < dl < dl < dl
Benzo(a)anthracene 8.0 26.6 222.4 9.4 101.3 400.6 668.4 239.6
Chrysene 21.8 62.2 601.1 24.8 137.4 367.1 387.5 183.6
Benzo(b)fluoranthene < dl 135.9 1062.3 38.1 53 82.6 161.5 67.7
Benzo(k)fluoranthene < dl < dl < dl < dl 79.3 150.8 263.0 129.2
Benzo(a)pyrene 4.5 46.1 267.5 4.1 15.9 199.5 206.4 174.0
Dibenzo(a,h)anthracene < dl < dl < dl < dl 52.7 55.0 84.4 39.5
Indeno(1,2,3-cd)pyrene < dl 25.1 251.5 6.2 22.2 28.7 1.6 19.9
Benzo(g,h,i)perylene < dl 27.0 314.5 7.4 62.4 28.0 101.2 21.1
Σ 2–3 rings 19.2 10.4 69.4 11.0 32.0 141.1 163.4 77.7
Σ 4–6 rings 73.4 353.9 3817.6 133.5 793.2 2051.5 2208.0 1255.8
Σ PAHs 92.6a 364.3b 3887.0c 144.5d 825.2a 2192.6b 2371.4b 1333.5c

Results are expressed as mean of three values (n = 3). Different letters indicate significant difference (p < 0.05)
between sampling sites. < dl, below detection limit

of some artisan fishing boats near the sampling site (petroleum PAHs in all sampling sites with a dominance of pyrolitic
combustion and petroleum sources) and the inexistence of origins.
industries or intensive urban activities around the site. The recorded PAH levels were compared with the sedi-
In addition to the composition pattern analysis, molecular ment quality guidelines (SQGs) from OSPAR Background
indices using PAH isomer ratios have been widely used to and Environmental Assessment Criteria (BAC and EAC)
determine the origins of environmental PAHs (Zrafi et al. (OSPAR 2009) and the NOAA (National Oceanic and
2013; Zaghden et al. 2017; Rajpara et al. 2017) (Table S3). Atmospheric Administration) effects low and median range
In this study, six isomeric ratios were used (Fig. 2). The four (ERL and ERM) (Long et al. 1995) (Table S4). Total PAH
sampling sites (S1–S4) displayed Ant/Ant+Phe ratio > 0.1 and concentrations in all stations were below ERL. For individ-
Flt/Flt + Pyr ratio > 0.4, which highlighted pyrolitic sources of ual PAHs, Flt, Chr, IcP and BgP concentrations in site S3
PAHs. The Flt/Flt + Pyr ratios in Tunis lagoon sites (S2–S4) exceeded the ERL, indicating occasional harmful effects on
were above 0.5, suggesting a contribution of grass, coal and/or aquatic organisms with 30 to 50% of incidence (Long et al.
wood combustion. Whereas, Flt/Flt + Pyr ratio in the reference 1995). Moreover, Flt and Pyr concentrations in S3 were
site (S1) was 0.47, suggesting that the contamination was due above the EAC level, suggesting a potential for significant
to petroleum combustion (De La Torre-Roche et al. 2009) adverse effects to the environment (OSPAR 2009). The
(Fig. 2a). For S2, S3 and S4 samples, Phe/Ant ratio was below concentrations of all individual PAHs were below EAC
10 and Flt/Pyr ratio was above 1, indicative of contamination and ERL at S1, S2 and S4. The application of the traffic
by pyrolitic sources. S1 sample had Phe/Ant ratio < 10 light system developed by (OSPAR 2009) on the sediment
(pyrolitic origins) and Flt/Pyr ratio < 1 (petrogenic origins), data indicated that sites S1, S2 and S4 had an acceptable
which implied both petrogenic and pyrolitic sources of PAHs state without significant risk of adverse effects to the envi-
in sediments (Fig. 2b). For Tunis lagoon sites (S2–S4), IcP/IcP ronment. Whereas, site S3 presented an unacceptable status
+ BgP ratio ranged from 0.44 to 0.48 and BaA/BaA + Chr due to Chr, IcP and BgP levels, which could cause an en-
ratio ranged from 0.27 to 0.30, suggesting mixed sources of vironmental health concern (Fig. S2). Consequently, PAH
PAH contamination (Fig. 2c). Overall, the used diagnostic concentrations in surface sediments of the site S3 may be
ratios revealed mixed sources for sediment contamination by harmful for marine organisms.
Environ Sci Pollut Res

PAHs showed a predominance of 4- and 5-ring PAHs.

Similarly, Orbea et al. (2002) reported a predominance of
tetra- and penta-ring PAHs in mussels, oysters and crabs sam-
pled from the Bay of Biscay, France. Hence, PAH composi-
tion pattern in clams was similar to that found in sediments for
the Tunis lagoon sampling sites, with a dominance of HMW
PAHs (93–95% in clams versus 92–98% in sediments) in
comparison to LMW PAHs (5–7% in clams versus 2–8% in
sediments). Such tissue PAH profiles may be explained by the
fact that clams live in close contact with sediments (Baumard
et al. 1998b). The tissue PAH concentration in S3 was 1.1-fold
and 1.8-fold higher than in S2 and S4, respectively. However,
high gradient of sediments’ PAH concentrations was detected
across sampling sites. Sediments’ PAH level in site S3 was 11-
fold and 27-fold higher than in S2 and S4, respectively. This
may indicate that sediments were not the only source of PAH
contamination for clams, but that other sources contributed to
PAH accumulation in these organisms such as water-column
and food (Baumard et al. 1998b; Mirsadeghi et al. 2013).
Clam PAH concentrations were assessed against the assess-
ment criteria threshold values (BAC and EAC) determined for
bivalves (OSPAR 2009) (Table S4). Clam concentrations in
Flt and BaA exceeded the EAC in the four sampling sites and
the same for BkF concentration in site S3, suggesting that
acute and long-term adverse biological effects are likely to
occur in the four sites. Clam concentrations in Chr, BaP and
BgP exceeded the BAC in all sampling sites, and BkF and IcP
concentrations were above BAC in S1, S2 and S4. These
findings indicate that Chr, BaP, BkF, IcP and BgP contamina-
tion in the sampling sites should be monitored to detect any
potential environmental problem in the future. According to
the traffic light system proposed by OSPAR (2009), the health
status of the Tunis lagoon should be considered as unaccept-
able based on clams’ PAH data (Flt, BaA and BkF) (Fig. S2).
Fig. 2 Cross-plots of the diagnostic ratios for PAHs in sediments. a Ant/
Ant + Flt vs. Flt/Flt + Pyr. b Phe/Ant vs. Flt/Pyr. c IcP/IcP + BgP vs. BaA/ Biomarker responses in clams
BaA + Chr
Multi-biomarker approach was applied to assess the biological
response in R. decussatus to PAH contamination at different
PAH bioaccumulation in clams levels of biological organisation: molecular, cellular and
tissular.
The clam PAH contents ranged from 1333.5 ± 112.1 (S4) to In terms of biochemical response, BPH, GST and CAT
2371.4 ± 130.2 ng g−1 (S3) in the Tunis lagoon. These tissue activities and MDA content were measured in the digestive
PAH concentrations remain very high compared with the ref- gland of R. decussatus (Fig. 3). The PAH accumulation in
erence site S1 (825.2 ± 83.3 ng g−1) (Table 1). The highest clam soft tissues at the investigated sites appeared to influence
clam PAH concentration was recorded in S3, which corre- the expression level of all biochemical biomarkers. The cyto-
sponds to the site with the highest PAH concentration in sed- chrome P450 family, involved in Phase I, and GST, involved
iments. Likewise, the lowest clam PAH concentration was in Phase II of the biotransformation process, have been used as
recorded in S4, corresponding to the site with the lowest sed- biomarkers of organic contaminant detoxification (Bebianno
iment PAH level in the Tunis lagoon. The composition pro- and Barreira 2009; Kamel et al. 2014). In this study, signifi-
files of PAHs in clams were quite similar across sites, with cantly higher BPH and GST levels were detected in clams
high amounts of HMW PAHs (93–96%) versus low percent- from Tunis lagoon than in those from reference site. BPH
ages of LMW PAHs (4–7%). The accumulation pattern of and GST inductions in the digestive gland seem to be caused
 Environ Sci Pollut Res

a 250 b 600
b
BPH pmol/min/mg proteins

GST µmol/min/mg proteins

b 500 b
200 b b
c 400
150
a a
300
100
200

50
100

0 0
S1 S2 S3 S4 S1 S2 S3 S4
c 250 d
18
b
CAT µmol/min/mg proteins

b 16 c

MDA nmol/mg proteins

200 c c 14 d
12
150
10
a
100
8 a
6

50 4
2
0 0
S1 S2 S3 S4 S1 S2 S3 S4
Fig. 3 Responses of benzo(a)pyrene hydroxylase (a), glutathione S- reference sites. Results are expressed as mean ± SD; n = 10. Different
transferase (b) and catalase (c) activities, and lipid peroxidation levels superscripts indicate significant difference (p < 0.05) between sampling
(d) in digestive glands of clams sampled from the Tunis lagoon and sites according to the ANOVA, multiple comparisons and Duncan’s test

by the accumulated PAHs in clams (positively correlated, The LPO level has been evaluated by assessing MDA content
Table S5) and are probably related to their role in the PAH (Kamel et al. 2014; Bebianno et al. 2015; Chalghmi et al.
biotransformation in this tissue. Furthermore, the highest BPH 2016b). It has been reported that exposure to PAHs could
level was found in clams from sites S2 and S3, which accu- induce lipid peroxidation in bivalves (Giannapas et al. 2012;
mulated the highest PAH levels. These high BPH levels sug- Xiu et al. 2014; Xie et al. 2017). In the present study, the
gest an increase of the Phase I biotransformation process and increase of MDA contents in clams from Tunis lagoon seems
hence, a higher production of Phase I reactive metabolites to be related to the alteration of the balance between ROS
such as quinones, phenols and diols (Michel et al. 1995). production and antioxidant defences in favour to oxidants
Moreover, the increased GST activity in clams from the la- due to the Phase I biotransformation of PAHs. Indeed, the
goon indicates an increase in Phase II biotransformation pro- general pathway of toxicity induced by PAHs in aquatic or-
cess (Banni et al. 2010; Breitwieser et al. 2018). However, ganisms is mainly based on the production of oxyradicals
GST showed a very low capacity to discriminate between (Livingstone 2001). In polluted sites, the increase of CAT
Tunis lagoon sites (no significant differences were found). activity was accompanied by an increase in LPO level, which
This is maybe related to the complexity of contamination in indicates cellular lipid damage and suggests that the rise in
Tunis lagoon due to the presence of other contaminants such antioxidant capacity was not sufficient to prevent oxidative
as metals (Chalghmi et al. 2016b). damage.
The investigated oxidative stress biomarkers were site- A transcriptional analysis of various genes (cox1, 16S
dependent and presented the same variation trends. CAT ac- rRNA, sod, hsp70 and mt) encoding proteins involved in mi-
tivity and MDA content were higher in clams from Tunis tochondrial metabolism, antioxidant defence, stress response
lagoon with respect to the reference site, indicating an oxida- and detoxification system was performed in the digestive
tive stress state likely caused by the increased PAH body bur- gland tissue (Table 2). PAH concentration-dependent upregu-
dens. The highest CAT and MDA levels were detected in the lation of cox1 gene (positive correlation, Table S5) was ob-
sampling sites S2 and S3 in which high level of tissue PAHs served for the Tunis lagoon sites. Cox1 gene contributes to
was detected. The increased CAT activity in the clams from ATP generation during the mitochondrial respiration (Powell
the polluted sites indicates that the antioxidant enzyme system et al. 2017). Cox1 induction could reflect a higher energy
tries to maintain the cellular redox state (Capó et al. 2015). d e m a n d i n d i g e s t i v e g l a n d l i k e l y d u e t o PA H
Environ Sci Pollut Res

Table 2 Relative expression and expression factors of selected genes in digestive gland of clams sampled from Tunis lagoon and reference sites

Mitochondrial metabolism Oxidative stress Protein reparation Detoxification system

response and protection
Sampling site cox1 16S rRNA sod hsp70 mt

S1 RGE 0.08 ± 0.02a 1.46 ± 0.28a (2.2 ± 0.4) 0.10−4a (0.30 ± 0.09) 0.10−4a (0.10 ± 0.03) 0.10−6a
S2 RGE 1.07 ± 0.1b 7.89 ± 1.14b (16.8 ± 5.4) 0.10−4b (10.7 ± 2.3) 0.10−4b (0.77 ± 0.07) 0.10−6bc
DEF 13 5 8 35 8
S3 RGE 1.15 ± 0.14b 5.88 ± 1.68b (18.4 ± 6.4) 0.10−4b (9.7 ± 1.1) 0.10−4b (1.16 ± 0.28) 0.10−6b
DEF 14 4 9 32 11
S4 RGE 0.58 ± 0.14c 5.21 ± 1.04b (7.3 ± 0.8) 0.10−4b (5.1 ± 0.7) 0.10−4c (0.50 ± 0.05) 0.10−6c
DEF 7 4 3 17 5

Gene expression was normalized against 18S rRNA (mean ± SEM, n = 5). RGE, relative gene expression; DEF, differential expression factor. Differential
expression factors for sites S2, S3 and S4 were calculated in relation to site S1. Different letters indicate significant difference (p < 0.05) between
sampling sites

biotransformation and detoxification mechanisms (Al gene was detected in clams sampled from the Tunis lagoon.
Kaddissi et al. 2014; Chalghmi et al. 2016a). On the other Expression of mt gene in these clams could be induced indi-
hand, cox1 induction may indicate an alteration of electron rectly by PAHs through oxidative stress incitation and/or di-
transport chain and hence could be a compensatory mecha- rectly by metal detoxification process.
nism to restore the reduced activity of mitochondria and to In parallel with biochemical and transcriptional changes,
limit cellular oxidative damage caused by PAHs through an histopathological alterations were observed in the digestive
effective oxygen consumption (Cambier et al. 2009). An over- gland of clams (Fig. S3). The intensity of each lesion and
expression of 16S rRNA gene was recorded for the three pol- the index of damage (ID) were determined (Table 3).
luted sites (S2–S4), indicating an increased de novo synthesis Overall, clams sampled from reference site revealed normal
of mitochondria. The 16S rRNA induction might correspond structure of the digestive gland with preserved cellular struc-
to the compensatory response explained above and suggests tures. The epithelia of digestive tubules were regular with an
that cells synthesized more mitochondria to substitute those intact basal lamina. Digestive tubules are composed of diges-
that were deficient (Chalghmi et al. 2016a). Overexpression of tive (eosinophilic cells, EC) and secretory (apical basophilic
sod gene involved in antioxidant defence was detected along cells, ABC) cells. The tubule lumen was narrow or almost
with upregulations of mitochondrial genes in Tunis lagoon occluded (Fig. S3A, B and C). In some individuals, very
clams suggesting an increased oxidative stress, which could low prevalence of histopathological lesions was observed in
be related to the increase in PAH metabolisation that lead to the digestive gland, in both severity and dissemination, with
increased ROS production (Livingstone 2001). Results also respect to haemocytic infiltration (HI) and lumen enlargement
revealed an increased hsp70 mRNA levels in association with
a rise in PAH soft body burdens. Hsp70 may be induced after
Table 3 A semi-quantitative evaluation of the histopathological lesions
protein alteration directly by the non-detoxified contaminants
in the digestive gland of clams sampled from Tunis lagoon and reference
or indirectly by the generated ROS (Gupta et al. 2010; sites
Mrdaković et al. 2016). As mentioned above, we have already
reported a likely increased generation of PAH metabolites S1 S2 S3 S4
(revealed by high BPH activity) and ROS (revealed by high Haemocytic infiltration (HI) +/− +++ +++ +++
CAT activity) in clams accumulating high PAH concentra-
Lumen enlargement (LE) +/− +++ +++ ++
tions, which can damage proteins and other cellular compo-
Fragmentation of the apical cytoplasm (FAC) − +++ +++ ++
nents. Thus, overexpression of hsp70 gene occurred to prevent
Narrowing of the epithelial cell layer (NEC) − ++ ++ +
and repair these damages. Our previous studies showed a met-
Cell debris (CD) − +++ +++ +
al pollution in the Tunis lagoon (Chalghmi et al. 2016a, b),
Rupture of the basal lamina (RBL) − ++ + +
and hence, transcriptional response of mt gene was used here
Cell necrosis (CN) − +++ ++ ++
as general stress biomarker for PAH pollution and specific
Digestive tubule necrosis (TN) − ++ + +
biomarker for metal contamination in the lagoon. Indeed, the
Fibrosis (Fb) − + +/− −
mt gene is involved in the detoxification of toxic metals
Index of damages 0.11 2.44 2.05 1.44
(Amiard et al. 2006), cell protection against oxidative stress
(Bebianno et al. 2005) and eventually in the regulation of gene Incidence of lesions: (−) absent, (+/−) sometimes, (+) frequent, (++) very
transcription (Roesijadi et al. 1998). An upregulation of mt frequent, (+++) always present
 Environ Sci Pollut Res

(LE) (Table 3). However, the most recurrent alterations ob- and reference sites (Fig. 4). In general, IBR index shows a
served in clams sampled from the Tunis lagoon sites consisted large range of variation across sampling sites (Fig. 4a).
of digestive gland inflammation revealed by a variable degree Increased IBR values were observed in PAH-contaminated
of HI in the connective tissue (Fig. S3E, F and I). Haemocytic sites, indicating an environmental stress. Based on the IBR
infiltration in bivalves was considered as a part of the inflam- index, sites S2 (IBR = 1.27) and S3 (IBR = 1.15), closer to
matory response and was associated to severe histopatholog- the petrochemical industries, were identified as the most
ical alterations generated by pollutants (Sheir and Handy stressed sites. The sampling site S4 (IBR = 0.40), closer to
2010; Costa et al. 2013; Chalghmi et al. 2016a). Moreover, the Rades harbour, seemed to be more impacted than the ref-
clams from Tunis lagoon presented foci of regressing tubules erence site and less than S3 and S4. The IBR values found in
(tubule atrophy) with a pre-necrotic aspect, characterised by a the four sites are in good agreement with the PAH levels
fragmentation of the apical cytoplasm (FAC) leading to reduc- detected in clams. Considering each site individually, almost
tion in epithelial cell layer (NEC), lumen enlargement (LE) all of biomarkers showed a strong contribution to the IBR
and occasional loss of cells into the lumen (CD) (Fig. S3D, E, index in S2 and S3, except mt gene expression in S2, implying
G, H, K, L, M and O). Further severe lesions, such as the a global stressful state in organism at molecular, cellular and
rupture of the basal lamina (RBL) and cell and tubule necrosis tissular levels due to the bioaccumulation of high PAH levels,
(CN and TN), were also observed in these clams (Fig. S3D, E, mainly the HMW compounds (Fig. 4b). IBR index of site S4
F, G, H, K, M, N and O). The highest severe lesions were showed an apparent difference of biological response from
observed in clams sampled from S2 and S3 (accumulating those measured for S2 and S3. It was mostly affected by the
the highest PAH concentrations) and consisted of fibrosis antioxidant enzyme (CAT), biotransformation enzyme (GST)
(Fb) formation (Fig. S3F). Similar lesions were reported in and mitochondrial gene (16SrRNA). Whereas, cox1, sod,
the digestive glands of bivalves exposed to PAHs and metals hsp70, mt and BPH biomarkers presented a low contribution
(Costa et al. 2013; Cuevas et al. 2015; dos Reis et al. 2015). to the IBR value. These results seem to be related to the low
The fragmentation and spillage of epithelial cells into the lu- PAH body burden in S4 and reveal less severe effects in or-
men of tubules that entail exposure of the basal membrane ganisms sampled from this site in comparison to S2 and S3.
could be the consequence of the vacuolization increase in
digestive cells (Usheva et al. 2006). The occurrence of abnor- Principal component analysis
mal vacuoles in digestive cells has been in turn associated to
an important increase in volume of contaminant-affected ly- PCA was conducted to obtain an overview of the spatial dis-
sosomes (Moore 1988; Lowe and Fossato 2000). tribution of the physicochemical, chemical and biological data
Krishnakumar et al. (1990) reported that breakdown and dila- and to evaluate the relative implication of these data in differ-
tation of tubules may be related to the autolytic processes as a ences among sampling sites. The plot of scores shows the
consequence of lysosomal destabilisation. The occurrence of position of investigated sites (S1–S4) in the ordination plane
tubule atrophy may be associated to PAH metabolisation by of the first two principal components (PC1 and PC2) and the
cytochrome P450 system, which could induce ROS produc- projection of different variables on the PC plane (Fig. 5). PC1
tion and hence leads to apoptosis (Lushchak 2011; dos Reis explained 83.8% of the variance and was positively loaded by
et al. 2015). The determined ID indicates that digestive glands transcriptomic, biochemical and histopathological parameters,
of clams from site S2 were the most altered followed by S3 PAH levels in clams and seawater salinity. PC1 clearly dis-
clams and finally S4 clams. In this study, the histopathological criminated the Tunis lagoon sites S2 and S3 (on the right
lesions seem to represent a progressive loss of the digestive quadrant) closely associated with the highest tissue PAH
gland functions and could eventually lead to dysfunctional levels and highest biological responses in organisms, from
organ. Therefore, histopathological damage observed in the the reference site S1 (on the left quadrant) marked by the
clam’s digestive gland could decrease the individual fitness lowest PAH concentrations and effects in clams. Site S4 was
by disrupting the vital biological processes (e.g. food absorp- in the centre of PC1 and was not particularly distinguished due
tion and digestion, detoxification process) and thus compro- to moderate levels of PAH contamination and biomarker re-
mise the growth and reproduction of organism. Thereby, the sponses. PC2 accounted for 12.2% of the variability of data
histopathological responses in bivalves have a high ecological and was negatively loaded by PAH concentrations in sedi-
relevance (Au 2004). ments. According to PC2 axis, the site S3 (lower quadrant)
associated with high PAH level in sediments was clearly dis-
Integrated biomarker response index criminated from the sites S2 and S4 (upper quadrant) which
presented a moderate PAH contamination in sediments.
The IBR was used as a general stress index combining multi- Taking into account the IBR index, PCA results also con-
level biological responses to diagnose and compare the overall firm that sites S2 and S3 were the most stressed among the
physiological state of clams sampled from the Tunis lagoon four sampling sites. However, the integration of chemical and
Environ Sci Pollut Res

a IBR/n
S1
1.5

0.5

S4 0 S2

S3

b S2 S3 S4
cox1 cox1 cox1
3 3 2
ID 16SrRNA ID 16SrRNA ID 1.5 16SrRNA
2 2
1
MDA 1 sod MDA 1 sod MDA 0.5 sod
0 0 0
CAT hsp 70 CAT hsp 70 CAT hsp 70

GST mt GST mt GST mt

BPH BPH BPH
Fig. 4 Integrated biomarker response in clam R. decussatus. a IBR index (IBR/n) in clams from the four sampling sites. b Biomarkers star plots for the
PAH-contaminated Tunis lagoon sites (S2–S4)

Fig. 5 Principal component 1.0

analysis (PCA) integrating data
from environmental parameters, 0.8
PAH concentrations and S4
biomarker responses in clam R.
0.6 Temperature
decussatus from Tunis lagoon and
reference sites
0.4 GST
S2 16SrRNA
PC2 (12.20 %)

0.2 CAT
Salinity
MDA
ID
0.0
hsp70
cox1
-0.2 S1 PAHs clam
pH sod
BPH
-0.4 mt

-0.6
PAHs sediment
-0.8 S3

-1.0
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
PC1 (83.88 %)
 Environ Sci Pollut Res

biological parameters in PCA provided more global overview References

of the environmental quality in sampling sites and hence iden-
tified the site S3 as the most PAH-impacted site, even more Al Kaddissi S, Legeay A, Elia AC et al (2014) Mitochondrial gene ex-
pression, antioxidant responses, and histopathology after cadmium
than site S2. Overall, our results indicate that clams at the
exposure. Environ Toxicol 29:893–907. https://doi.org/10.1002/tox.
Tunis lagoon sites faced moderate to high PAH exposure 21817
and responded by activating their defence system through Amiard J-C, Amiard-Triquet C, Barka S et al (2006) Metallothioneins in
molecular and biochemical process. These organisms present- aquatic invertebrates: their role in metal detoxification and their use
as biomarkers. Aquat Toxicol 76:160–202. https://doi.org/10.1016/j.
ed an increased lipid peroxidation level and severe histopath-
aquatox.2005.08.015
ological lesions in digestive gland. These harmful effects Au DWT (2004) The application of histo-cytopathological biomarkers in
could be related to the PAH toxicity and overloading of the marine pollution monitoring: a review. Mar Pollut Bull 48:817–834.
defence system. https://doi.org/10.1016/j.marpolbul.2004.02.032
Banni M, Bouraoui Z, Ghedira J, Clearandeau C, Jebali J, Boussetta H
(2009) Seasonal variation of oxidative stress biomarkers in clams
Ruditapes decussatus sampled from Tunisian coastal areas. Environ
Monit Assess 155:119–128. https://doi.org/10.1007/s10661-008-
Conclusions 0422-3
Banni M, Negri A, Dagnino A, Jebali J, Ameur S, Boussetta H (2010)
This study revealed that Tunis lagoon was moderately to Acute effects of benzo[a]pyrene on digestive gland enzymatic bio-
highly PAH-contaminated. The PAH levels in sediments of markers and DNA damage on mussel Mytilus galloprovincialis.
Ecotoxicol Environ Saf 73:842–848. https://doi.org/10.1016/j.
site S3 exceeded the sediment quality criteria (ERL), indi-
ecoenv.2009.12.032
cating a high ecological risk for benthic organisms. Baumard P, Budzinski H, Garrigues P, Sorbe JC, Burgeot T, Bellocq J
Diagnostic ratios showed mixed origins for PAHs in sedi- (1998a) Concentrations of PAHs (polycyclic aromatic hydrocar-
ments (petrogenic and pyrolitic) with predominance of the bons) in various marine organisms in relation to those in sediments
and to trophic level. Mar Pollut Bull 36:951–960. https://doi.org/10.
pyrolitic sources. Likewise, clams from the Tunis lagoon
1016/S0025-326X(98)00088-5
sites exhibited high PAH body burdens which exceeded the Baumard P, Budzinski H, Garrigues P (1998b) Polycyclic aromatic hy-
assessment criteria (EAC) for some individual PAHs. drocarbons in sediments and mussels of the western Mediterranean
According to our results, sediments represent a major sea. Environ Toxicol Chem 17:765–776. https://doi.org/10.1002/
source of PAH bioaccumulation in clams. The present work etc.5620170501
Bebianno MJ, Barreira LA (2009) Polycyclic aromatic hydrocarbons
revealed also a biological response disruption at concentrations and biomarker responses in the clam Ruditapes
transcriptomic, biochemical and tissular levels in clams decussatus transplanted in the Ria Formosa lagoon. Ecotoxicol
sampled from Tunis lagoon, which was related to the tissue Environ Saf 72:1849–1860. https://doi.org/10.1016/j.ecoenv.2009.
PAH concentrations. The integration of chemical and bio- 03.016
Bebianno MJ, Company R, Serafim A et al (2005) Antioxidant systems
logical parameters allowed discrimination among sampling and lipid peroxidation in Bathymodiolus azoricus from Mid-
sites according to their stressful status. Indeed, S3 was Atlantic Ridge hydrothermal vent fields. Aquat Toxicol 75:354–
identified as the most stressed site followed by S2, while 373. https://doi.org/10.1016/j.aquatox.2005.08.013
S4 was less impacted despite its proximity to Rades har- Bebianno MJ, Pereira CG, Rey F, Cravo A, Duarte D, D'Errico G, Regoli
F (2015) Integrated approach to assess ecosystem health in harbor
bour. The present study generated a large amount of useful
areas. Sci Total Environ 514:92–107. https://doi.org/10.1016/j.
data for other environmental monitoring programs along scitotenv.2015.01.050
the Mediterranean coasts and provided an effective and Beliaeff B, Burgeot T (2002) Integrated biomarker response: a useful tool
powerful environmental tool integrating sediment chemis- for ecological risk assessment. Environ Toxicol Chem 21:1316–
try, contaminant bioaccumulation and biomarkers of differ- 1322. https://doi.org/10.1002/etc.5620210629
Bemanikharanagh A, Bakhtiari AR, Mohammadi J, Taghizadeh-
ent dimensions (transcriptomic, biochemical and histopath- Mehrjardi R (2017) Characterization and ecological risk of polycy-
ological) for more accurate assessment of health status of clic aromatic hydrocarbons (PAHs) and n-alkanes in sediments of
coastal areas, including lagoons. Shadegan international wetland, the Persian Gulf. Mar Pollut Bull
124:155–170. https://doi.org/10.1016/j.marpolbul.2017.07.015
Acknowledgements This study was supported by the University of Bi C, Wang X, Jia J, Chen Z (2018) Spatial variation and sources of
Bordeaux, France, the Ministry of Scientific Research and Technology, polycyclic aromatic hydrocarbons influenced by intensive land use
the University of Monastir, Tunisia, and the French Institute of Tunisia. in an urbanized river network of East China. Sci Total Environ 627:
671–680. https://doi.org/10.1016/j.scitotenv.2018.01.272
Bradford MM (1976) A rapid and sensitive method for the quantitation of
Compliance with ethical standards microgram quantities of protein utilizing the principle of protein-dye
binding. Anal Biochem 72:248–254
Conflict of interest The authors declare that they have no conflict of Breitwieser M, Thomas-Guyon H, Huet V, Sagerup K, Geraudie P (2018)
interest. Spatial and temporal impacts of the Skjervøy harbour diesel spill on
native population of blue mussels: a sub-Arctic case study.
Publisher’s Note Springer Nature remains neutral with regard to juris- Ecotoxicol Environ Saf 153:168–174. https://doi.org/10.1016/j.
dictional claims in published maps and institutional affiliations. ecoenv.2018.01.033
Environ Sci Pollut Res

Broeg K, Lehtonen KK (2006) Indices for the assessment of environmen- Devin S, Buffet PE, Châtel A, Perrein-Ettajani H, Valsami-Jones E,
tal pollution of the Baltic Sea coasts: integrated assessment of a Mouneyrac C (2017) The integrated biomarker response: a suitable
multi-biomarker approach. Mar Pollut Bull 53:508–522. https:// tool to evaluate toxicity of metal-based nanoparticles. Nanotoxicology
doi.org/10.1016/j.marpolbul.2006.02.004 11:1–6. https://doi.org/10.1080/17435390.2016.1269374
Buege JA, Aust SD (1978) Microsomal lipid peroxidation. Methods dos Reis IMM, Mattos JJ, Garcez RC, Zacchi FL, Miguelão T, Flores-
Enzymol 52:302–310 Nunes F, Toledo-Silva G, Sasaki ST, Taniguchi S, Bícego MC,
Cacciatore F, Bernarello V, Boscolo Brusà R, Sesta G, Franceschini G, Cargnin-Ferreira E, Bainy ACD (2015) Histological responses and
Maggi C, Gabellini M, Lamberti CV (2018) PAH (polycyclic aromat- localization of the cytochrome P450 (CYP2AU1) in Crassostrea
ic hydrocarbon) bioaccumulation and PAHs/shell weight index in brasiliana exposed to phenanthrene. Aquat Toxicol 169:79–89.
Ruditapes philippinarum (Adams & Reeve, 1850) from the Vallona https://doi.org/10.1016/j.aquatox.2015.10.011
lagoon (northern Adriatic Sea, NE Italy). Ecotoxicol Environ Saf 148: Er-Raioui H, Bouzid S, Marhraoui M, Saliot A (2009) Hydrocarbon pollu-
787–798. https://doi.org/10.1016/j.ecoenv.2017.11.050 tion of the Mediterranean coastline of Morocco. Ocean Coast Manag
Cambier S, Bénard G, Mesmer-Dudons N, Gonzalez P, Rossignol R, 52:124–129. https://doi.org/10.1016/j.ocecoaman.2008.10.006
Brèthes D, Bourdineaud JP (2009) At environmental doses, dietary Giannapas M, Karnis L, Dailianis S (2012) Generation of free radicals in
methylmercury inhibits mitochondrial energy metabolism in skeletal haemocytes of mussels after exposure to low molecular weight PAH
muscles of the zebra fish (Danio rerio). Int J Biochem Cell Biol 41: components: immune activation, oxidative and genotoxic effects.
791–799. https://doi.org/10.1016/j.biocel.2008.08.008 Comp Biochem Physiol Toxicol Pharmacol CBP 155:182–189.
Capó X, Tejada S, Box A, Deudero S, Sureda A (2015) Oxidative status https://doi.org/10.1016/j.cbpc.2011.08.001
assessment of the endemic bivalve Pinna nobilis affected by the oil Gupta SC, Sharma A, Mishra M, Mishra RK, Chowdhuri DK (2010)
spill from the sinking of the Don Pedro. Mar Environ Res 110:19– Heat shock proteins in toxicology: how close and how far? Life
24. https://doi.org/10.1016/j.marenvres.2015.07.013 Sci 86:377–384. https://doi.org/10.1016/j.lfs.2009.12.015
Capolupo M, Franzellitti S, Kiwan A, Valbonesi P, Dinelli E, Pignotti E, Habig WH, Pabst MJ, Jakoby WB (1974) Glutathione S-transferases the
Birke M, Fabbri E (2017) A comprehensive evaluation of the envi- first enzymatic step in mercapturic acid formation. J Biol Chem 249:
ronmental quality of a coastal lagoon (Ravenna, Italy): integrating 7130–7139
chemical and physiological analyses in mussels as a biomonitoring Kamel N, Burgeot T, Banni M, Chalghaf M, Devin S, Minier C, Boussetta
strategy. Sci Total Environ 598:146–159. https://doi.org/10.1016/j. H (2014) Effects of increasing temperatures on biomarker responses
scitotenv.2017.04.119 and accumulation of hazardous substances in rope mussels (Mytilus
Chalghmi H, Zrafi I, Saidane-Mosbahi D (2015) Metabolism and oxida- galloprovincialis) from Bizerte lagoon. Environ Sci Pollut Res 21:
tive stress biomarkers in clam Ruditapes decussatus to assess pollu- 6108–6123. https://doi.org/10.1007/s11356-014-2540-5
tion in industrial area of Mediterranean lagoon. Int J Res Chem Kim K-H, Jahan SA, Kabir E, Brown RJC (2013) A review of airborne
Environ 5:42–49 polycyclic aromatic hydrocarbons (PAHs) and their human health effects.
Chalghmi H, Bourdineaud J-P, Haouas Z, Gourves PY, Zrafi I, Saidane- Environ Int 60:71–80. https://doi.org/10.1016/j.envint.2013.07.019
Mosbahi D (2016a) Transcriptomic, biochemical, and histopatho- Krishnakumar PK, Asokan PK, Pillai VK (1990) Physiological and cel-
logical responses of the clam Ruditapes decussatus from a metal- lular responses to copper and mercury in the green mussel Perna
contaminated Tunis lagoon. Arch Environ Contam Toxicol 70:241– viridis (Linnaeus). Aquat Toxicol 18:163–173. https://doi.org/10.
256. https://doi.org/10.1007/s00244-015-0185-0 1016/0166-445X(90)90024-J
Chalghmi H, Zrafi I, Gourves P-Y, Bourdineaud JP, Saidane-Mosbahi D Livingstone DR (2001) Contaminant-stimulated reactive oxygen species
(2016b) Combined effects of metal contamination and abiotic pa- production and oxidative damage in aquatic organisms. Mar Pollut
rameters on biomarker responses in clam Ruditapes decussatus gills: Bull 42:656–666. https://doi.org/10.1016/S0025-326X(01)00060-1
an integrated approach in biomonitoring of Tunis lagoon. Environ Long ER, Macdonald DD, Smith SL, Calder FD (1995) Incidence of
Sci Process Impacts 18:895–907. https://doi.org/10.1039/ adverse biological effects within ranges of chemical concentrations
c6em00139d in marine and estuarine sediments. Environ Manag 19:81–97.
Chalghmi H, Zrafi I, Saidane-Mosbahi D (2016c) Chronic effects of https://doi.org/10.1007/BF02472006
petroleum hydrocarbons in Tunis-navigation channel on phase I Lowe DM, Fossato VU (2000) The influence of environmental contami-
and II biotransformation enzymes in bivalve species. Int J Res nants on lysosomal activity in the digestive cells of mussels (Mytilus
Chem Environ 6:28–33 galloprovincialis) from the Venice Lagoon. Aquat Toxicol 48:75–85.
Claiborne A (1985) Catalase activity. In: Greenwald RA (ed) CRC hand- https://doi.org/10.1016/S0166-445X(99)00054-5
book of methods for oxygen radical research. CRC Press Inc., Boca Lu M, Zeng D-C, Liao Y, Tong B (2012) Distribution and characterization
Raton, pp 283–284 of organochlorine pesticides and polycyclic aromatic hydrocarbons in
Costa PM, Carreira S, Costa MH, Caeiro S (2013) Development of his- surface sediment from Poyang Lake, China. Sci Total Environ 433:
topathological indices in a commercial marine bivalve (Ruditapes 491–497. https://doi.org/10.1016/j.scitotenv.2012.06.108
decussatus) to determine environmental quality. Aquat Toxicol 126: Lushchak VI (2011) Environmentally induced oxidative stress in aquatic
442–454. https://doi.org/10.1016/j.aquatox.2012.08.013 animals. Aquat Toxicol 101:13–30. https://doi.org/10.1016/j.
Cuevas N, Zorita I, Costa PM, Franco J, Larreta J (2015) Development of aquatox.2010.10.006
histopathological indices in the digestive gland and gonad of mus- Michel XR, Suteau P, Robertson LW, Narbonne J-F (1993) Effects of
sels: integration with contamination levels and effects of confound- benzo(a)pyrene, 3,3′,4,4′-tetrachlorobiphenyl and 2,2′,4,4′,5,5′-
ing factors. Aquat Toxicol 162:152–164. https://doi.org/10.1016/j. hexachlorobiphenyl on the xenobiotic-metabolizing enzymes in
aquatox.2015.03.011 the mussel (Mytilus galloprovincialis). Aquat Toxicol 27:335–344.
Culotta L, De Stefano C, Gianguzza A et al (2006) The PAH composition of https://doi.org/10.1016/0166-445X(93)90062-6
surface sediments from Stagnone coastal lagoon, Marsala (Italy). Mar Michel XR, Beasse C, Narbonne J-F (1995) In vivo metabolism of
Chem 99:117–127. https://doi.org/10.1016/j.marchem.2005.05.010 benzo(a)pyrene in the mussel Mytilus galloprovincialis. Arch
De La Torre-Roche RJ, Lee W-Y, Campos-Díaz SI (2009) Soil-borne Environ Contam Toxicol 28:215–222. https://doi.org/10.1007/
polycyclic aromatic hydrocarbons in El Paso, Texas: analysis of a BF00217619
potential problem in the United States/Mexico border region. J Mirsadeghi SA, Zakaria MP, Yap CK, Gobas F (2013) Evaluation of the
Hazard Mater 163:946–958. https://doi.org/10.1016/j.jhazmat. potential bioaccumulation ability of the blood cockle (Anadara
2008.07.089 granosa L.) for assessment of environmental matrices of mudflats.
 Environ Sci Pollut Res

Sci Total Environ 454–455:584–597. https://doi.org/10.1016/j. thionein conjugate pair. J Biol Chem 273:17425–17432. https://doi.
scitotenv.2013.03.001 org/10.1074/jbc.273.28.17425
Moore MN (1988) Cytochemical responses of the lysosomal system and Serafim A, Company R, Lopes B, Fonseca VF, França S, Vasconcelos RP,
NADPH-ferrihemoprotein reductase in molluscan digestive cells to Bebianno MJ, Cabral HN (2012) Application of an integrated bio-
environmental and experimental exposure to xenobiotics. Mar Ecol marker response index (IBR) to assess temporal variation of envi-
Prog Ser 46:81–89 ronmental quality in two Portuguese aquatic systems. Ecol Indic 19:
Mrdaković M, Ilijin L, Vlahović M, Matić D, Gavrilović A, Mrkonja A, 215–225. https://doi.org/10.1016/j.ecolind.2011.08.009
Perić-Mataruga V (2016) Acetylcholinesterase (AChE) and heat Shaw GR, Connell DW (1994) Prediction and monitoring of the carcino-
shock proteins (Hsp70) of gypsy moth (Lymantria dispar L.) larvae genicity of polycyclic aromatic compounds (PACs). In: Ware G.W.
in response to long-term fluoranthene exposure. Chemosphere 159: (eds), Reviews of environmental contamination and toxicology, vol
565–569. https://doi.org/10.1016/j.chemosphere.2016.06.059 135. Springer, New York, NY, pp 1–62. https://doi.org/10.1007/978-
Mzoughi N, Chouba L (2011) Distribution and partitioning of aliphatic 1-4612-2634-5_1
hydrocarbons and polycyclic aromatic hydrocarbons between water, Sheir SK, Handy RD (2010) Tissue injury and cellular immune responses
suspended particulate matter, and sediment in harbours of the west to cadmium chloride exposure in the common mussel Mytilus
coastal of the Gulf of Tunis (Tunisia). J Environ Monit 13:689–698. edulis: modulation by lipopolysaccharide. Arch Environ Contam
https://doi.org/10.1039/C0EM00616E Toxicol 59:602–613. https://doi.org/10.1007/s00244-010-9502-9
Nicholson S, Lam PKS (2005) Pollution monitoring in Southeast Asia Tuncel SG, Topal T (2015) Polycyclic aromatic hydrocarbons (PAHs) in
using biomarkers in the mytilid mussel Perna viridis (Mytilidae: sea sediments of the Turkish Mediterranean coast, composition and
Bivalvia). Environ Int 31:121–132. https://doi.org/10.1016/j. sources. Environ Sci Pollut Res 22:4213–4221. https://doi.org/10.
envint.2004.05.007 1007/s11356-014-3621-1
Orbea A, Ortiz-Zarragoitia M, Solé M, Porte C, Cajaraville MP (2002) Usheva LN, Vaschenko MA, Durkina VB (2006) Histopathology of the
Antioxidant enzymes and peroxisome proliferation in relation to digestive gland of the bivalve mollusk Crenomytilus grayanus
contaminant body burdens of PAHs and PCBs in bivalve molluscs, (Dunker, 1853) from southwestern Peter the Great Bay, Sea of
crabs and fish from the Urdaibai and Plentzia estuaries (Bay of Japan. Russ J Mar Biol 32:166–172. https://doi.org/10.1134/
Biscay). Aquat Toxicol 58:75–98. https://doi.org/10.1016/S0166- S1063074006030047
445X(01)00226-0 Wurl O, Obbard JP (2004) A review of pollutants in the sea-surface
OSPAR (2009) Background Document on CEMPAssessment Criteria for microlayer (SML): a unique habitat for marine organisms. Mar Pollut
QSR 2010. OSPAR Commission 2009. Monitoring and Assessment Bull 48:1016–1030. https://doi.org/10.1016/j.marpolbul.2004.03.016
Series, ISBN 978-1-907390-08-1, 25. Xie J, Zhao C, Han Q, Zhou H, Li Q, Diao X (2017) Effects of pyrene
Pérez-Ruzafa A, Marcos C, Pérez-Ruzafa IM (2011) Mediterranean exposure on immune response and oxidative stress in the pearl oys-
coastal lagoons in an ecosystem and aquatic resources management ter, Pinctada martensii. Fish Shellfish Immunol 63:237–244. https://
context. Phys Chem Earth Parts ABC 36:160–166. https://doi.org/ doi.org/10.1016/j.fsi.2017.02.032
10.1016/j.pce.2010.04.013 Xiu M, Pan L, Jin Q (2014) Bioaccumulation and oxidative damage in
Pfaffl MW (2001) A new mathematical model for relative quantification juvenile scallop Chlamys farreri exposed to benzo[a]pyrene,
in real-time RT–PCR. Nucleic Acids Res 29:e45–e445 benzo[b]fluoranthene and chrysene. Ecotoxicol Environ Saf 107:
Powell RD, Goodenow DA, Mixer HV, Mckillop IH, Evans SL (2017) 103–110. https://doi.org/10.1016/j.ecoenv.2014.05.016
Cytochrome c limits oxidative stress and decreases acidosis in a rat Zaghden H, Tedetti M, Sayadi S, Serbaji MM, Elleuch B, Saliot A (2017)
model of hemorrhagic shock and reperfusion injury. J Trauma Acute Origin and distribution of hydrocarbons and organic matter in the
Care Surg 82:35–41. https://doi.org/10.1097/TA.0000000000001257 surficial sediments of the Sfax-Kerkennah channel (Tunisia, south-
Rajpara RK, Dudhagara DR, Bhatt JK, Gosai HB, Dave BP (2017) ern Mediterranean Sea). Mar Pollut Bull 117:414–428. https://doi.
Polycyclic aromatic hydrocarbons (PAHs) at the Gulf of Kutch, org/10.1016/j.marpolbul.2017.02.007
Gujarat, India: occurrence, source apportionment, and toxicity of Zrafi I, Hizem L, Chalghmi H, Ghrabi A, Rouabhia M, Saidane-Mosbahi
PAHs as an emerging issue. Mar Pollut Bull 119:231–238. https:// D (2013) Aliphatic and aromatic biomarkers for petroleum hydro-
doi.org/10.1016/j.marpolbul.2017.04.039 carbon investigation in marine sediment. J Pet Sci Res 2:145–155.
Riba I, González de Canales M, Forja JM, DelValls TA (2004) Sediment https://doi.org/10.14355/jpsr.2013.0204.01
quality in the Guadalquivir estuary: sublethal effects associated with Zrafi-Nouira I, Khedir-Ghenim Z, Zrafi F, Bahri R, Cheraeif I, Rouabhia
the Aznalcóllar mining spill. Mar Pollut Bull 48:153–163. https:// M, Saidane-Mosbahi D (2008) Hydrocarbon pollution in the sedi-
doi.org/10.1016/S0025-326X(03)00392-8 ment from the Jarzouna-Bizerte coastal area of Tunisia
Roesijadi G, Bogumil R, Vasák M, Kägi JHR (1998) Modulation of DNA (Mediterranean Sea). Bull Environ Contam Toxicol 80:566–572.
binding of a tramtrack zinc finger peptide by the metallothionein- https://doi.org/10.1007/s00128-008-9421-x