Introduction

Freshwater contains many xenobiotics derived from human activities such as pesticides, hydrocarbons, and trace metals. Pharmaceutical products (PPs) are a class of xenobiotics emerging as freshwater pollutants. Human consumption and veterinary activity represent the main source of PPs found in the environment (Jørgensen and Halling-Sørensen 2000). After therapeutic or diagnostic use, PPs and/or their metabolites are excreted into aquatic environments primarily from sewage treatments plants (STP) which cannot always completely eliminate them with a removal rate ranging from 0 to 97 % (Cardoso et al. 2014). They may therefore reach surface water at concentrations range in the nanogram per liter to microgram per liter (Vasieva 2011). These drugs can cause toxic effect even in low nanomolar concentrations; (Cardoso et al. 2014). It is fundamental to evaluate the ecotoxicological risk of these mixtures. Biomarkers involved in detoxification reactions are often used to evaluate the impact of organic pollutants on different aquatic animals. Among these, three biomarkers are particularly used to report an activation of the detoxification systems. These enzymes participating in the metabolic transformation of xenobiotics are classified into three phases, namely phases I, II, and II pathways. Cytochrome P450s (CYP) are one of the major phase I-type classes of biotransformation enzymes and are generally involved in the functionalization of exogenous and endogenous compounds thereby creating a more polar and water-soluble compound (Nelson et al. 1996; Rewitz et al. 2006). The cytochrome P450 4 (CYP4) family is one of the oldest P450 families, having evolved about 1.25 billion years ago, just after the formation of the steroid biosynthetic genes (Simpson 1997). The CYP4 family is divided into 72 subfamilies and the majority is found in invertebrates. In invertebrates, CYP4 has been identified in several phyla such as Arthropoda, Mollusca, Annelida, and Echinodermata (Snyder 1998). Several studies have demonstrated that they are involved in xenobiotics metabolism. Indeed, cyp4 expression is modified by different types of organic xenobiotics (pesticides, phenolic compounds, drugs). In Chironoma tentans, the cyp4 gene is induced by atrazin, biocide tributylin (TBTO) and reduced by nonylphenol (alkylphénols) and bisphenol A (Londoño et al. 2004; Martínez-Paz et al. 2012). In Annelida, xenobiotics such as B[a]P and phenobarbital induce the expression of cyp4 genes (Chen et al. 2012; Rewitz et al. 2004). In Mollusca, cyp4 cDNA have been cloned in seven marine species Perna viridis, Mytilus galloprovincialis, Venerupis philippinarum, Cyphoma gibbosum, and recently in Ruditapes philippinarium, Chlamys farreri, and Crassotrea angulata (Miao et al. 2011; Pan et al. 2011) but little is known about their functions. Induction of cytochrome P450 by xenobiotics exposure can also be studied at enzyme activity level. EROD is commonly used to evaluate CYP1A in vertebrates (Jung et al. 2011; Sureda et al. 2011). But to date, there is no evidence of cyp1A 1 gene expression in bivalves. Actually, only predicted sequences are available for cyp1a1 and a2 or cyp1-like cDNA in bivalves Crassotrea angulata, Mytilus californianus, and Crassotrea gigas (Zhang et al. 2012a; Zanette et al. 2010). Therefore, several studies investigate phase I xenobiotic metabolism in invertebrates (polychaetes, bivalves) exposed to organic chemicals in environment with EROD activity (Binelli et al. 2006; Falfushynska et al. 2012; Ramos-Gómez et al. 2011).

Glutathione S-transferase (GST) are phase II detoxification enzymes and catalyze the attachment of glutathione (GSH) to a variety of electrophilic species for the elimination of potentially toxic xenobiotics. GST are localized in cytosolic or microsomal fraction of cells and have been grouped in several classes. Among these classes, cytosolic pi-GST are known to be involved not only in drug metabolism but also in antioxidant system by inactivating lipoperoxidation product, lipid hydroperoxide, and their derivates. The pi-GST can also directly inactivate ROS with their SH group (Vasieva 2011). pi-GST has been identified in several bivalves. In Corbicula fluminea, pi-GST is expressed in digestive glands, excretory systems, and gills (Doyen et al. 2005). In bivalves, pi-GST gene expression or GST global activity has been used in several studies to evaluate the impact of environmental stresses (Bigot et al. 2010; Gonzalez-Rey and Bebianno 2011; Rola et al. 2012). An additional crucial step in detoxification is the excretion of xenobiotics and/or their metabolites by the ATP-binding cassette (ABC) transporters named multixenobiotic resistance (MXR) system. They are involved in phase III detoxification pathways and are often used as biomarkers to evaluate exposure effects with several classes of xenobiotics in bivalves.

Bivalve molluscs are relevant species to study the freshwater quality. Indeed they are sedentary and filter large amounts of water to meet their respiratory and nutritional needs. The clam Corbicula fluminea is a freshwater bivalve originating from Asia. This species is invasive in many rivers and lakes from North America and Europe and is often used as a biological model in ecotoxicology in freshwater pollution assays (Brandão et al. 2014; Chen et al. 2014). The aim of this study is to assess the molecular responses of Corbicula fluminea clams exposed to wastewater from a psychiatric hospital. In this effluent, many drugs and especially neuroleptics are likely to modify cyp4, pi-GST, and multidrug resistant 1 (mdr1) expression and/or CYP and GST activity. In this study, Corbicula fluminea cyp4 complete cDNA and mdr1 partial cDNA sequence were characterized. We also determined tissue-specific expression of mdr1 and cyp4 genes. Afterwards, cyp4, pi-GST, and mdr1 mRNA expression and cytochrome P450 and GST enzyme activity in response to wastewater containing many pharmaceuticals were measured by real-time PCR, EROD, and measure of GST activity.

Materials and methods

Wastewater sampling

The study site is a regionally psychiatric hospital with 250 used beds and 120 places in ambulatory services (Dordogne, France). In 2010, approximately 25 kg of psychotropic drugs were consumed in the hospital. The wastewater of the hospital is discharged to a wastewater treatment plant (WWTP), which treats up to 600 m3/day. This wastewater treatment plant is composed of a primary decanter, a bacterial filter, and a secondary decanter. This WWTP does not operate continuously. When the volume in the lifting area reaches its maximum, waters fall over by pouring into the primary decanter. So, wastewater effluents are, more or less, always with the same composition.

The wastewater samples are the result of equal mixtures of six discharges of effluent. Samples are collected twice a week during the experiment.

Analysis of pharmaceuticals

The multi-residue analytical procedure was adapted from Togola and Budzinski (2008) and Devier et al. (2013). Briefly, each effluent sample (200 mL) was percolated on Oasis® MCX (60 mg, 3 mL) cartridges. Then cartridges were dried under vacuum and elution consisted in a three-step combination of solvents: 3 mL of ethyl acetate, 3 mL of ethyl acetate/acetone (50/50 v/v), and 3 mL of methanol/dichloromethane (50/50 v/v) with 5 % of ammonium hydroxide. After complete evaporation to dryness under a gentle stream of nitrogen, samples were reconstituted with 300 μL ultra-pure water/acetonitrile (90/10 v/v).

Instrumental analysis was performed using a Waters Ultra Performance Liquid Chromatography (UPLC) ACQUITY coupled to Waters Quattro Premier XE triple quadrupole (Waters, Saint Quentin en Yvelines, France), equipped with a electrospray ionization (ESI). The chromatographic separation was achieved with a Kinetex C18 column (100 × 2.1 mm; 1.7 μm) in 17 min. Identification of each compound was ensured through the monitoring of two transitions, except for Norfluoxetine where fragmentation pattern gives a unique fragmentation (Table 1, supplementary data). The ion ratio was used as a confirmation criterion.

Instrumental blank and procedural blank samples were performed for each batch experiment in order to prevent any contamination.

In addition, in each batch extraction experiment, Vittel mineral water was spiked with native compounds (200 ng L−1) and extracted as a sample to calculate extraction recoveries. For the majority of the compounds analyzed, the values were greater than 50 %.

The instrumental method was validated through estimation of linearity, repeatability, and sensitivity. Method detection limits (MDLs) were extrapolated from signal to noise ratio (S/N = 3) measured on enriched Vittel water samples extracted in triplicate. Thus, the MDLs of the compounds studied ranged from 0.1 to 1.1 ng L−1.

Organisms and wastewater exposure experiment

Corbicula fluminea clams (20 to 30 mm in length) were collected from a local lake (la Roucarié, Carmaux, Midi-Pyrénées, France) and acclimated in the laboratory for 1 week in tanks with dechlorated and aerated tap water with 2 cm of sterile sand substrate at 17 °C under natural photoperiod. While there, clams were fed twice a week with JBL Korall Fluid. For the exposure, clams (115 per tank) were then randomized in three tanks per experiment (control and exposed) with 40 L of dechlorated and aerated water (controls) or effluent (exposed) and 2 cm of sterile sand substrate. The sand had been sterilized by baking at 300 °C for 12 h. For tissue analysis and cloning experiments, clams were collected directly after the acclimatization step. The tank water was changed by half every 2 days with dechlorated and aerated potable water from the hospital for the control tanks or with wastewater for exposed organisms. The concentration of psychotropics in tanks (with and without mollusk) was determined after 2 days of exposure.

RNA isolation and reverse transcription

As regards the exposure experiment, five clams of each condition were sampled at 0, 1, 3, 7, 14, and 21 days according to Bigot et al. (2009). For tissue expression analysis, ten clams were sampled. For cDNA cloning, three clams were sampled. Gills, digestive gland, adductor muscles, and mantle were then immediately dissected. The gills and digestive gland of exposed and control clams were stored at −80 °C in RNA latter. The gills, digestive gland, muscles, and mantle used for tissue expression analysis were processed for RNA extraction immediately after dissection.

Total RNA was extracted using TRIzol® reagent (Invitrogen, USA) according to the manufacturer’s protocols. Contaminating genomic DNA was removed using DNA-freeTM kit (Applied Biosystem, USA) following the manufacturer’s instruction. RNA quantity and purity were evaluated using a nanodrop (NanoDrop 2000, Thermo, USA) and electrophoresis on 1.5 % agarose gel in TBE (pH 8.3 Tris-borate-EDTA). cDNA was synthesized from 1 μg of total RNA extracted from the digestive gland for cDNA cloning experiments and from 500 ng of total RNA for the real-time PCR experiment by M-MLV reverse transcriptase (Invitrogen, USA) at 37 °C for 50 min with random hexamers (Invitrogen, USA) following the protocol of the manufacturer.

cyp4 and mdr1 cDNA cloning

An internal fragment of cyp4 and mdr1 cDNA was amplified using degenerated primers (CYP4for, CYP4rev, Mdr1Fow and Mdr1rev, see Table 1). CYP4 primers were designed based on CYP4 amino acid sequences alignment of P. viridis (B1A7T0), Mytilus galloprovincialis (O76505), V. philippinarum (B9VAW8), Chlamys farreri (B8YQ60), Unio tumidus (Q6QWQ4), Mytella strigata (Q2PPJ4), Mactra veneriformis (D3J8D8), and Modiolus modiolus (C6EQB8) and Cyphoma gibbosum (B3FYF8). Mdr1 primers were designed based on MDR1 amino acid sequence alignment of Mus musculus (P06795), Homo sapiens (P08183), Caenorhabditis elegans (P34712), and Mytilus galloprovincialis (G4RJB1). The primers were obtained from Invitrogen (Fisher Bioblock, France). The PCR reaction was performed in a total volume of 25 μL using thermocycler (Multigene-gradient, Labnet) containing recombinant taq DNA polymerase (Invitrogen, USA), 2 mmol L−1 MgCl2, 2 μmol L−1 of each primer, and 2 μL of cDNA. Gene amplification was carried out as follow: 95 °C for 5 min; 40 cycles in three steps: 95 °C for 30 s, 55 °C for 1 min, and 72 °C for 1 min; 72 °C for 10 min. The amplification products were analyzed on 1.5 % agarose gel. The target gene bands were purified using agarose gel DNA purification kits (Invitrogen, USA). The fragments were cloned into pCR®II vector using TA-cloning® Kit Dual promoter (pCR®II) (Invitrogen, USA). Vectors containing cloned insert were transformed into One shot®INVαF' chemically competent Escherichia coli and incubated overnight at 37 °C. Positive clones were screened by blue/white selection, verified by EcoRI digestion, and then sequenced with M13 reverse and forward(-20) primers (Get-TQ, Purpan, Toulouse, France).

Table 1 Primers used for cyp4 and mdr1 mRNA isolation and expression analysis
Full size table

Amplification of cyp4 cDNA ends

Full-length cyp4 cDNA was obtained using FirstChoice® RLM-RACE Kit (Invitrogen, USA) according to the manufacturer’s instructions. Gene-specific primers (Table 1) were designed from the partial cyp4 cDNA sequence. For 5′ RACE, PCR was performed as follow: 3 min at 95 °C followed by 7 cycles of 95 °C for 30 s, 65 °C for 1 min minus 1 °C per cycle, 72 °C for 1 min, followed by 33 cycles of 95 °C for 30 s, 58 °C for 1 min and 72 °C for 1 min with an increase of 20 s per cycle, then an additional extension time for 10 min. For 3′ RACE, PCR was performed as follow: 5 min at 95 °C followed by 40 cycles of 95 °C for 30 s, 60 °C for 30 s and 72 °C for 1 min then an additional extension time for 10 min. Specific 1,200 pb 5′ PCR and 600 pb 3′ PCR products were then purified after gel migration with MinElute Gel Extraction Kit (Qiagen, Netherlands). The RACE products were cloned using the same TA-cloning® protocol described above and sequenced with M13 reverse and forward(-20) primers (Get-TQ Purpan, Toulouse, France).

Sequence analysis

Partial and complete sequences were analyzed using Mobyle pasteur site (http://mobyle.pasteur.fr/ ). The amino acid sequences were deduced using Transeq software. The homology searches of nucleotides and protein sequences were conducted with Blast program on NCBI and Uniprot (http://www.ncbi.nlm.nih.gov/, www.uniprot.org/).

Quantification of cyp4, pi-gst, mdr1 gene expression after wastewater exposure

For cyp4, mdr1, and pi-gst primer design, we used Primer3 software. Primers were designed to obtain about 200 pb cDNA-specific product. The primers were first tested by real-time PCR to verify their efficiency. The efficiency of all the primers used was around 100 % (96 % for pi-gst primers, 104 % for EF1alpha primers and 103 % for cyp4 primers, 98 % for mdr1 primers). The expression of cyp4, pi-gst, and mdr1 mRNAs in gills and digestive gland was quantified using a SYBR Green real-time PCR method. The real-time PCR was carried out in replicates on 96-well STEP-one + (Applied biosystem) and analyzed with StepOne + v2.2.2 software. The PCR reactions were performed in a final volume of 12 μL containing 6 μL of 2× EvaGreen® qPCR Real Time Master Mix Plus (Biorad), 2 μL of cDNA template, 0.6 μL of 10 μmol L−1 of each primer (Table 1), and 2.8 μL of ultrapure water. Elongation factor 1-alpha (Ef1alpha) was chosen as an internal gene reference to normalize the expression level between samples. Ef1alpha primers are listed in Table 1 and are the same as in Martins et al. (2011). All data were relatively calculated and expressed for Ef1alpha to compensate any difference in initial RNA quantity and in Reverse transcriptase efficiency. The relative gene expression fold change was calculated using the 2−ΔΔCt method (Livak and Schmittgen 2001).

Tissues expression analysis

We proceeded by following the same protocol described above.

Enzymatic activities

The gills, digestive gland, mantle, and muscles of the molluscs were dissected. Pools were made with the tissues of three organisms. Each pool of tissue was homogenized in 100 mM phosphate buffer; pH 7.4 containing 250 mM sucrose, 100 mM KCl, 1 mM EDTA, and protease inhibitors according to commercial advice (SIGMA FAST, S8820) using TissueLyser II (Qiagen) with 25 frequencies/s for 3 min. Eight pools were prepared and aliquots of each homogenate were sampled for total protein, GST, and EROD analysis. In the GST fraction, dithiothreitol was added (100 mM). Aliquots were frozen at −80 °C until analysis. For GST, EROD, and protein analyses, samples were defrosted and centrifuged at 15,000 × g for 30 min and at 4 °C. Pellets were discharged and supernatants were used in the enzymatic assays.

EROD activity

EROD activity was measured using the method described by Gagné and Blaise (1993) with ethoxyresorufin as substrate. The reaction was initiated by the addition of NADPH. The produced 7-hydroxyresorufin was recorded fluorometrically (λex = 520 nm; λem = 590 nm) for 120 min. Each pool was analyzed in triplicate and EROD activity was expressed as picomole per minute per milliliter (volumetric acidity) or picomole per minute per milligram proteins (specific activity).

Glutathione S-transferase activity

GST activity was measured using the method described by Habig et al. (1974) with 1-chloro-2,4-dinitro-benzene (CDNB) as substrate of spectrophotometrical measurement at 340 nm and 37 °C.

Protein content was measured according to Bradford (1976) using bovine serum albumin (BSA) as standard.

Statistics

For both enzyme activity and RNA quantification, the equality of variances and normality were tested (using Bartlett test and Shapiro-Wilk normality test, respectively). Differences among transcript levels in different organs and enzyme activity among control and treated groups relative to the controls and time exposure were evaluated using one-way analysis of variance (ANOVA), followed by Tukey’s test for multiple comparisons. Differences in RNA quantification among control and contaminated groups were evaluated using non-parametric Mann–Whitney test. The probability level for statistical significance was P < 0.05. All statistical analysis was performed using R software version 2.14 (www.r-project.org/).

Results

Cloning and sequence analysis of cyp4 cDNA

cyp4 cDNA was amplified using degenerated primers CYP4For and CYP4Rev (Table 1) which are based on highly conserved regions of mollusca CYP4 proteins. A complete cDNA sequence was then obtained using RACE PCR strategy. The full length of mRNA is 1,946 bp with an open reading frame of 516 amino acid (GenBank accession number: JQ678818) (Fig. 1). A blast search on uniprot revealed that this sequence was similar to those of Cytochrome p450 family 4 of other molluscs and other vertebrates or invertebrates with a higher homology with cytochrome p450 4 V2 subfamily of Crassostrea gigas (65 %, E value = 0) and cytochrome p450 family variant 2 (CYP4V10 gene) of Cyphoma gibbosum (53 %, E value = 0). An alpha helice was predicted at the N terminal end (probably the signal sequence of import in endoplasmic reticulum). Alignment with other mollusc CYP4 revealed many conserved motifs on Corbicula fluminea CYP4 (Fig. 2). A typical heme-binding motif takes place near the C-terminus FxxGxxxCxG (F450SAGARNCIG459) followed by a QK(F/Y)AxLE sequence specific to family 4 (Q460KFAIL465 for Corbicula fluminea CYP4 protein). A typical WxxxR motif is also present near the N-terminus (W125KIRR129) together with the consensus sequence GxxT located in the I-helix and the 13 invariant amino acid residues E312VDTFMFEGHDTT324 unique to family 4. Finally, the charge pair consensus motif ExxR (E377ALR380) is found within the K helix. The predicted 2D structure reveals the potential existence of 12 alpha-helix and 5 beta-sheet (data not shown). Phylogenetic analysis with 25 CYP4 complete amino acid sequences of molluscs, Drosophila melanogaster, Mus musculus, H. sapiens, and Danio rerio revealed that Corbicula fluminea CYP4 shows more similarities with CYP4V subtype of Cyphoma gibbosum and Crassostrea gigas but also with the CYP4V subtypes of vertebrate phylum (Fig. 3).

Fig. 1
figure 1

Nucleotide and deduced amino acid sequences of Corbicula fluminea Clam cyp4 (accession number: JQ678818)

Full size image
Fig. 2
figure 2

Multiple alignments of deduced amino acid sequences of Corbicula fluminea clam with other known mollusca CYP4: V. philippinarum (B9VAW8), Chlamys farreri CYP4 (B8YQ60), Cyphoma gibbosum (B3FYF4). First box: WxxxR: (C-helix). Second box: characteristic of the family 4 cytochrome p450, not shared by other CYP genes. (GxxTT part of helix-I). Third box: ExxR (K-helix, heme stabilization) The four box represent the heme binding site: FxxGxxxCxG

Full size image
Fig. 3
figure 3

Neighbor joining phylogenetic tree of Corbicula fluminea deduced amino acid sequence CYP4 and other CYP4 sequences

Full size image

Cloning and sequence analysis of partial mdr1 cDNA

We cloned a 333-bp mdr1 cDNA fragment encoding a predicted ORF of 111 amino-acid. A blast on uniprot databank revealed that this sequence was similar mdr1 of many bivalves, V. philippinarum, Crassostrea gigas, Mytilus galloprovencialis, around 80 %. Figure 4 shows alignment of Corbicula fluminea MDR1 with other bivalve MDR1.

Fig. 4
figure 4

Multiple alignments of MDR1 deduced amino acid sequences of Corbicula fluminea clam with other known MDR1: Mus musculus (P06795), Mytilus galloprovencialis (G4RJB1), H. sapiens (P08183), and Caenorhabditis elegans (P34712)

Full size image

Tissue distribution of cyp4 and mdr1 mRNA

We evaluated the tissue distribution of cyp4 and mdr1 mRNA in clams. We tested the level of expression in gills, digestive gland, adductor muscles, and mantle. The results are shown in Fig. 5. As expected, the expression level of cyp4 was significantly higher in digestive gland (fourfold, p value <0.001) than in other tissues. No significant difference was observed between the levels of expression in mantle, adductor muscles, and gills, despite a slightly higher expression in muscles. For mdr1, we observed the higher expression level in gills (twofold, p value <0.001). No significant difference was observed between expression level in mantle and adductor muscles.

Fig. 5
figure 5

Expression changes of cyp4 (a) and mdr1 (b) genes in digestive gland (DG), mantle (Ma), and adductor muscles (Mu) compared to gills. Levels of expression are measured by EvaGreen quantitative real-time PCR, cyp4 and mdr1 expression is measured relative to EF1alpha expression. Values are represented as means ± SE of relative expression ratio between digestive gland, mantle, or adductor muscles and gills. Significant differences from all the other tissues are indicated with an asterisk (ANOVA followed by Tukey’s test, p value <0.001)

Full size image

Tissular EROD activity

The results of tissular EROD activity are presented in Fig. 6. Cytosolic EROD activity was assessed in mantle, gills, muscles, and digestive gland. Mantle exhibited the highest value followed, in the order, by the digestive gland, muscles, and gills. By using a ANOVA post hoc Tukey HSD test, significant differences were found between any pairs of tissues except between digestive gland and muscles (p < 0.05).

Fig. 6
figure 6

Tissular distribution (gills, mantle, digestive gland, and muscles) of EROD activity in Corbicula fluminea. Asterisk and hashtag: EROD activity was significantly different in the tissue (ANOVA followed by Tukey’s test, p value <0.05)

Full size image

Pharmaceutical concentrations in effluents after mollusk exposure

The psychotropic concentrations in the effluent used for the experiment (at T0, immediately after introduction to the tanks, and after 72 h (with or without Corbicula fluminea)) are presented in Table 2.

Table 2 Psychotropic concentrations in the effluent used for the experiment—at T0, after 24 h, and 72 h with or without Corbicula fluminea
Full size table

The results obtained after 24 and 72 h showed significant decreases in the concentrations of all compounds when compared to those at T0. However, we observed that after 72 h of exposure and before the change of water, psychotropics were still concentrated in effluent. Mollusks did not totally accumulate or did not degrade pharmaceuticals in 72 h and were always in touch with pharmaceuticals.

Quantification of cyp4, pi-gst, and mdr1 mRNA expression in gill and digestive gland after wastewater exposure

We studied the expression pattern of cyp4, pi-gst, and mdr1 mRNA in clams tissue exposed to wastewater from a psychiatric hospital’s sewage treatment plant. In gills and digestive gland, we observed no significant difference in cyp4 expression between exposed and unexposed clams (Fig. 7). Pi-gst gene expression was significantly decreased after 14 days (−0.64, ±0.07, p value <0.05) of exposure in gills and showed no significant variation in digestive gland. Mdr1 gene expression was significantly overexpressed in gills during the whole experiment (×4.86 ± 1.2; × 13.13 ± 3.7; × 7.76 ± 1.75; × 4.96 ± 0.53; × 3 ± 0.3; p value <0.01). In the first times of exposure, individual responses were variables but always significantly increased. At 14 and 21 days of exposure, these individual variations were lower and overexpression maintained in exposure context. Mdr1 expression was also significantly upregulated in digestive gland at 3 days of exposure (×4.4, ±1.7 p value <0.05).

Fig. 7
figure 7

Expression changes of mdr1 (a) pi-gst (b), and cyp4 (c) in digestive gland and gills of unexposed clams at 1, 3, 7, 14, and 21 days of experiment. Levels of expression are measured by EvaGreen quantitative real-time PCR and mdr1, pi-gst, and cyp4 expression is measured relative to EF1alpha expression. Values are represented as means ± SE of relative expression ratio between control sample at 3, 7, 14, and 21 and control sample at 1 day. Significant differences from control in the same time and tissue are indicated with an asterisk (Mann–Whitney non-parametric test, p value <0.05)

Full size image

EROD and GST activity after wastewater exposure

We evaluated EROD activity in the digestive gland of exposed and unexposed clams. The results are presented in Fig. 8. After 24 h of exposure, EROD activity in Corbicula fluminea increased significantly in exposed clams when compared with control. No significant difference was observed between the exposed clams and controls for the other duration of exposure. GST activity was followed in gills and digestive gland of exposed and control Corbicula fluminea. GSTs activity in gills of Corbicula fluminea was significantly enhanced when compared with control clams after 14 days of exposure. In the digestive gland, GSTs activity decreased significantly compared to the controls after 3 and 7 days of exposure and increased after 21 days of exposure.

Fig. 8
figure 8

Mean variation (±SD) of GST (a) and cytochrome p450 (b) activities in digestive gland and gills of Corbicula fluminea exposed to effluent for 1, 3, 7, 14, and 21 days. Significant difference between control and exposed clams is indicated with an asterisk (ANOVA followed by Tukey’s test, p value <0.05)

Full size image

Discussion

We identified full-length cDNA of a cytochrome p450 family 4 member in Corbicula fluminea. Sequence analysis reveals that Corbicula fluminea CYP4 protein exhibits a high degree of conservation of consensus motifs specific to other cytochrome p450 family 4 in particular the heme motif binding site near the C-terminus and the conserved 13 amino acid residue E312VDTFMFEGHDTT324 of the I-helix. The typical E319 residue is known to bind covalently the heme group in certain member of CYP4 family through an ester linkage (Baer et al. 2005). This covalently link with the heme group is unique in family 4. Indeed the other cytochrome p450 bind their heme with non-covalent link. Cyp4 gene is expressed in all tissues analyzed but it expression level is significantly higher in digestive gland, which is consistent with results obtained in Ruditapes philippinarum (Pan et al. 2011). The digestive gland is known to be the major place for detoxification pathways in molluscs (Zanette et al. 2013). We can hypothesize that CYP4 is involved in these processes in Corbicula fluminea digestive gland. However, EROD activity is higher in the mantle than in the digestive gland. So, EROD is not only correlated to the activity of CYP4 in molluscs. Cytochromes p450 are involved in the metabolism of steroids in molluscs (Fernandes et al. 2011). However, sterols are distributed mainly in the mantle and digestive gland of the bivalve mollusc Scapharca inaequivalvis. In February, when the experiment was conducted, the sterol quantity was higher in the mantle than in the digestive gland of S. inaequivalvis (Piretti and Pagliuca 1989). EROD may reflect the activity of several cytochromes P450 involved in this metabolism.

After clam exposure to wastewater, no variation of cyp4 gene expression was observed in gills or digestive gland. In U. tumidus, no change in cyp4 gene expression was observed after exposure to diethylhexylphthalate (DEHP) known to be an inducer of cyp4 in vertebrate (Chaty et al. 2004). In Ruditapes philippinarum, cyp4 gene shows a significant increase only after 10 days of exposure with 0.2 μg L−1 of BaP and no variation with 0.01 g/L of BaP (Pan et al. 2011). In other bivalves as Chalmys farreri, exposure to BaP at the same doses induced a high decrease in cyp4 gene expression at 3 and 10 days of exposure. Hence, the response of cyp4 gene expression in bivalves appears to be very variable. Interestingly, when we analyzed the cytochrome p450 global activity, we showed that at 1 day of exposure this activity was significantly increased in the digestive gland of exposed clams. At all the other times, EROD activity showed no significant variation when compared to control. Indeed, cytochrome p450 response is very early and transitory. Usually EROD activity is modified during all exposure time with organic pollutants (Curtis et al. 2011; Martín-Díaz et al. 2009) but some authors also showed a transitory response of EROD activity in case of exposure with different xenobiotics (PaB, PCP, pharmaceutics compounds, caffeine) in several organisms (bivalves, fishes) (Binelli et al. 2006; Li et al. 2012; Monari et al. 2009; Thibaut and Porte 2008).

In gills, pi-gst is significantly downregulated after 14 days of exposure therefore no significant variation is observed in digestive gland. We have also observed changes in GST activity from 3 days of exposure. The GSTs activity decreases at 3 and 7 days of exposure and increases at 21 days in digestive glands and at 14 days in gills. So the wastewater exposure induced changes in enzyme activities in both gills and digestive glands. The effluent from the wastewater treatment plant can be an inhibitor of GST. Gagné et al. showed this type of effect in mussels exposed to wastewater effluent for 7 weeks (Gagné et al. 2007). The metabolic activity of cytochrome p450 may cause the production of metabolites inhibiting GST. It has been shown that the metabolites of exogenous estradiol inhibited human GST (Tian et al. 2012). Moreover, the metabolism of paracetamol or carbamazepine using cytochrome p450 produces quinone (Ju and Uetrecht 1999) that may inhibit GST (van Ommen et al. 1991).

cyp4 and pi-gst gene expression cannot be linked to enzyme activity. The difference between variation of gene expression and enzyme activity may be due to the gap in gene expression time and protein activity time and to the multiple points of regulation that may occur between translational level and enzyme activity level. In the case of cyp4 gene expression, it will be necessary to test an anterior time as in Anwar-Mohamed and El-Kadi (2008).

Another possibility to explain these differences is that the protocols used to evaluate enzyme activity measure the global cytochrome p450 or GST families activity. This can explain this absence of correlation between the two types of analysis. The changes observed in GST or cytochrome p450 activity could be due to another class of GST such as GST-O, GST-S, or GST-M or to another family of cytochrome p450 such as cytochrome p450 family 3, 2, or 1-like that are also expressed in gills or digestive gland of other bivalves (Umasuthan et al. 2012; Zanette et al. 2013; Zhang et al. 2012b). For example, in Ruditapes philippinarum, Cyp3a65 and Cyp3u2 are upregulated in gills after 3 days of exposure to 1,000 μg L−1 of ibuprofen (Milan et al. 2013).

Mdr1 was expressed in all examined tissues but higher levels were found in gills. Mdr1 gene shows also a strong overexpression in gills of exposed clams compared to control. MXR are members of the phase III system and are involved in excretion of xenobiotics or conjugated metabolites (Luedeking and Koehler 2004). In aquatic mollusks, gills are the first tissue in contact with xenobiotics. Several authors showed the presence of MXR in the gills of bivalve molluscs (Galgani et al. 1996; Keppler and Ringwood 2001; Minier et al. 2000). Achard et al. (2004) related the induction of MXR in gills of Corbicula fluminea exposed to cadmium. These results revealed that MXR plays an important role in detoxification pathway after wastewater exposure and that mdr1 overexpression is an indicator of pollution exposure.

In conclusion, wastewater exposure modified expression and/or activity of phase I, II, and III proteins in both organs (gills and digestive gland). Except for mdr1 that is permanently overexpressed in gills, others biomarkers are very transitory increased or decreased either at transcriptional or enzymatic activity levels.