Biomarkers Study in Rainbow Trout Exposed to Industrially Contaminated Groundwater

Abstract

The spill of liquid industrial waste from chemical and petrochemical industries in Mercier lagoons located 20 km south of Montreal, Quebec, caused a major groundwa- ter contamination by industrial contaminants. The aim of this study was to investigate the toxic effects of Mercier groundwater, following 4 and 14 days of exposure to graded concentra- tions from three wells at increasing distances 1.2, 2.7 and 5.4 km from the source of contam- ination. Rainbow trout were examined for sev- eral biomarkers of defense [ethoxyresorufin O-deethylase (EROD) and gluthatione S-trans- ferase (GST) activities] and those of tissue damage [lipid peroxidation (LPO) and DNA strand breaks]. The results showed that EROD activity was significantly enhanced in hepatic tissue at 1.2 and 5.4 km, whereas inhibition in activity was observed in group at 2.7 km. Therefore, GST activity was significantly increased at 3.1% concentration for the 2.7 km well. No change in LPO was observed. However, a significant induction of DNA strand breaks in liver was obtained at each distance. In conclusion, the data suggest that the release of these contaminants in groundwater leads to increased biotransformation for copla- nar aromatic hydrocarbons and DNA damage in groundwater.

Introduction

The dumping of 40,000 to 170,000 m³ of industrial hydrocarbon waste in an abandoned gravel pit (Mercier, Quebec, Canada) caused severe water supply problems in the region.[1] The main industrial waste consisted of used oil from chemical and petrochemical indus- tries.[2] In the gravel pit lagoon, the contami- nants permeate through soil down to the groundwater. Among the organic compounds found, volatiles organic compounds (VOCs) (

Table 1. Samples of water collected in campaign 2010 from 15 wells in order to study the quality of Mercier groundwater. Analysis showed that approximately half of wells (7 wells) have concentrations of volatiles organic compounds higher than detection limit [(the concentration values (µg/L)]. Vinyl chloride concentrations within a well ranged up to 0.5 ug/L and a maximum concentration of 85 ug/L was recorded in one well. Analyses were provided by the Centre d’Expertise en Analyse Environnementale of the province of Quebec, and determined according to their standard methods (CAEQ,2010).

) such as vinyl chloride (chloroethene), trans 1,2-dichloroethylene, 1,1-dichloroethane, 1,2-dichloroethane, cis 1,2-dichloroethene, benzene, chlorobenzene, m+p-xylènes, chloroethane, 1,3-dichloro- propane and phenol.[1,2,3] Their physicochemical properties and hydrogeological conditions of the site are responsible of the complexity of contamination problem.[4] Even if the lagoons are the main source of contamination, the industrial waste incinerator may be consid- ered as well.

Organic industrial wastes containing VOCs, polyaromatic hydrocarbons (PAHs), polychlori- nated biphenyls (PCBs) and metals are usual- ly found at hazardous waste sites.[5] They are known to produce a variety of adverse effects in organism including, immunotoxicity,[6,7] genotoxicity[8,9,10] and carcinogenesis.[11,12,13] Those contaminants are able to induce the biotrans- formation enzyme cytochrome P450 and glu- tathione S-transferase in fish[12,14,15,16] and induce lipid peroxidation (LPO) and DNA strand breaks in exposed animals.[9,17,18]

Although chemical analysis could find some of chemicals in these complex mixtures, it remains difficult to determine the cumulative effects in exposed organisms. The measure- ment of biomarkers represent a mean to deter- mine the toxicological outcome of exposure to complex industrial contaminants mixtures[19,20] and give early warning signals about environ- mental threats of contaminants.[21] Indeed, exposure to PAHs leads to the induction of cytochrome P4501A1 responsible for ethoxyre- sorufin-O-deethylase (EROD) activity in fish liver and kidney.[22,23,24] Gwinn et al.[9] showed that certain volatile organic compounds (e.g. vinyl chloride, dichloroethane) are metabolized via oxidation mediated by CYP 450 system to form electrophilic metabolites which may be detoxi- fied by glutathione S-transferase (GSTs).

Glutathione S-transferase is an important phase II enzyme that catalyses the conjugation of electrophilic compounds to GSH,[25] GST reacts with a wide spectrum of environmental pollutants.[26,27] It has been reported that the enzyme GST react differently to a variety of compounds.[18] These same authors, reported that GST was inhibited by benzene, whereas its activity was enhanced in animals exposed to 2,4-dichlorophenol.[28] Since this reaction consumes reduced glutathione, sustained or increased activity could lead to oxidative stress in cells.

The measurement of LPO levels in animal tissues has been recognized as biomarkers of oxidative damage towards unsaturated phos- pholipids.[18,20] LPO is reported to cause loss of cell function under oxidative stress.[29] Genotoxicity was determined by evaluating DNA strand breaks and has been proposed as effective biomarkers in assessing the impact of contaminants released into the aquatic environment.[30] Sasaki et al.[31] showed hepatic DNA damage in rodents exposed to 1,2-dichloroethane (EDC). Exposure to benzene leads to single strand breaks in erythrocytes in rodents.[10] However, the toxicity of EDC has yet to be examined in fish. In this context, the present study was carried out to investigate biomarkers enzyme effects of Mercier ground- water exposure on rainbow trout. A set of bio- markers of defense and tissue damage were measured in the liver of trout exposed to groundwater at various distances from the landfill.

Materials and methods

Fish

Juvenile rainbow trout (O. mykiss) (total N=216), weighing 23.14±9.3 g, were obtained from a local fish farm Les Arpents Verts (Sainte-Edwidge-de-Clifton, Quebec, Canada). They were kept for a minimum of 2 weeks in 300-L tanks at 15°C under a photoperiod of 12 h light: 12 h dark and constant aeration. They were fed daily at a rate of 2% body weight with food pellets.

Groundwater exposure experiments

The groundwater samples were collected in the summer of may 2010 from three wells at 1.2, 2.7 and 5.4 km from Mercier lagoons (Figure 1). These samples were then stored at 4°C in the dark until exposure. In six plastic vessels (31 cm diameter × 44 cm height, 20 L capacity) filled with 15 L of water sample using polyethylene plastic bags, six groups of 12 juve- nile rainbow trout were exposed to increasing concentrations of groundwater (3.1% and 50%) for 4 days and 14 days at 15°C. The con- trol group and the dilution water consisted of UV- and charcoal-treated tap water from the City of Montréal (Quebec, Canada). The expo- sure experiments were repeated twice and the water was changed twice a week. The fish were fed 3 times weekly with commercial fish pellets and the extra food was removed after 10 min. The feeding was stopped 24 h before the end of the exposure. Water chemistry (e.g., temperature, pH, oxygen, conductivity, nitrate- nitrite and ammonia) according to standard methods of the Centre d’Expertise en Analyse Environnementale of the province of Quebec[3] was within acceptable limits. The following organic VOC parameters were determined: vinyl chloride, trans1, 2-dichloroethylene, 1-1 dichloroethane, 1, 2-dichloroethane, cis1, 2- dichloroethene, benzene, and chlorobenzene (

Table 2. Summary of analytical results of groundwater sites at 1.2, 2.7 and 5.4 km sampled for 61 organic contaminants (data present- ed only compounds with concentration above limit detection).

Values of measurements during sampling in may 2010. (-) No detected.

).[32,33,34]

Biochemical analyses

Fish were humanely anaesthetized with tri- caine methanesulfonate 0.1% (MS222) (Sigma-Aldrich, ON, Canada) after 4 days and 14 days (n=6), in accordance with the recom- mendations of the animal care committee and length and body weight measured. Condition factor (CF) for each fish was calculated accord- ing to the following equation (White and Fletcher, 1985):

and the hepatosomatic index (HSI) was deter- mined by the following:

Fish samples were conserved at −80°C until biochemical analyses. Frozen livers were thawed and homogenized (20 s duration) using a Teflon pestle tissue grinder in an ice- cold homogenization buffer (10 mM Hepes- NaOH, pH 7.4 containing 140 mM NaCl, 1 mM dithiothreitol and 1 mg/mL aprotinin (a pro- tease inhibitor). Aliquots of each homogenate were taken for LPO, DNA strand break determi- nations and protein concentrations. The remainder of the homogenate was centrifuged at 15,000 g for 20 min at 4°C. The supernatant (S15) was collected to measure 7-ethoxyre- sorufin-O-deethylase activity, GST activity and proteins. These biomarkers were normalized with both homogenate and S15 protein con- centrations as determined by the method of Bradford[35] using standards of bovine serum albumin.

7-Ethoxyresorufin-O-deethylase

7-ethoxyresorufin-O-deethylase activity was measured according to Gagné and Blaise[36] method. The reaction mixture contained 50 µL of S15 and 160 µL of 50 µM 7-ethoxyresorufin in 10 mM Tris-HCL, pH 7.4 and Tween 20. The reaction was started by the addition of 10 µL NADPH 1 mM. The mixture was incubated at room temperature for 30 and 60 min where flu- orescence of resorufin was measured using 535 nm (excitation) and 635 nm (emission) filters. Calibration was achieved by comparing the rate of fluorescence change in the samples with fluorescence of resorufin standards. Enzyme activity (EROD) was expressed in nmol of resorufin/min/mg of proteins.

Lipid peroxidation

Lipid peroxidation was measured according to the thiobarbituric acid method (Wills, 1987). A volume of 150 µL of the homogenate were mixed with 300 µL of 10% trichloroacetic acid solution containing 1 mM FeSO4 and 150 µL of 0.67% thiobarbituric acid. The mixture was allowed to stand in a water bath (70°C) for 10 min. Standards solutions were prepared with 0.001% of tetramethoxypropane for calibration. The thiobarbituric acid reactants (TBARS) were measured by fluorescence at 540 nm (excitation) and 590 nm (emission). Results were expressed in nmol of malonaldehyde equivalents per mg of proteins.

DNA strand breaks

DNA strand breaks were determined using the alkaline precipitation.[37] Briefly, 25 µL of the homogenate was added to 200 µL of SDS solu- tion 2% containing 40 mM NaOH, 10 mM Trisma base and 10 mM EDTA. An equal of vol- ume of 0.12 M KCl was added to the mixture which was allowed to stand in water bath 60°C for 10 min, and then cooled at 4°C for 30 min; to precipitate SDS associated nucleoproteins and genomic DNA. The mixture was then cen- trifuged at 8000 g for 5 min. DNA present in the supernatant were measured by mixing 50 µL of the supernatant with 150 µL of Hoechst dye at a concentration of 1 µg/mL in 0.4M NaCl, 4 mM sodium cholate and 0.1 M tris-acetate pH 8.5. Fluorescence was then assessed using 360 nm (excitation) and 460 nm (emission) filters. DNA quantification was measured with stan- dard solutions of Salmon sperm DNA. Results were expressed as µg DNA/mg of proteins.

Glutathione S-transferase

Glutathione S-transferase activity was measured using 1-chloro-2-4-dinitrobenzene (CDNB) as the co-substrat. A volume of 50 µL of S15 was mixed with 1 mM GSH and 1 mM CDNB substrate in 50 mM Hepes-NaOH, pH 7.4, containing 100 mM NaCl. The mixture was incubated at 15°C for 0, 5, 15, 25 and 45 min. Activity was expressed by the increase in absorbance at 340 nm per min per milligram of protein.

Statistical analyses

The data were expressed as the mean with the standard error. In each experience, differ- ences between control and groups of fish exposed to Mercier groundwater were evaluat-ed by one-way analysis of variance (ANOVA) followed by post-hoc test (P≤0.05). The calcula- tions were performed using Statistica for Windows (Version 7.0, StatSoft Inc., 1995). Correlation was performed using Pearson test. Discriminant function analysis was performed to determine the well distance characteristics. Significance was set at P 0.05.

Results

Chemical analysis

Among the 61 volatile organic compounds measured, vinyl chloride (4 µg/L), trans-1,2- dichloroethylene (2 µg/L), 1,1-dichlorethane (0.4 µg/L), 1,2-dichloroethane (2 µg/L), cis-1,2-dichloroethene (0.37 µg/L) and benzene (0.29 µg/L) were found at concentrations above the detection limit at 2.7 km (Table 2).

Morphological parameters

Neither trout weights, lengths nor CF and HSI showed any significant differences in sampling sites after 4 and 14 days of exposure time (

Table 3. Morphometric data of trout exposed to groundwater at 1.2 km, 2.7 km and 5.4 km.

Values reported as mean±standard error.

). However, significant correla- tions were found between weights and CF (r= 0.91, P 0.05), between lengths and CF (r= 0.96, P 0.05) in groups at 5.4 km.

Biomarker responses

Phase I and II biotransformation activities

EROD activity was significantly increased in fish exposed to groundwater at 1.2 km (50%) and 5.4 km (3.1%) following 4 and 14 days exposure respectively reaching 1.8-fold induc- tions. However, a significant inhibition in EROD activity was observed in fish exposed to groundwater at 2.7 km for all times of exposure at (3.1%) and (50%) concentrations (Figure 2). GST activity in the liver of fish exposed at 3.1% groundwater concentration from the site located at 2.7 km raised significantly after 4 days, reaching 2.1-fold relative to control (Figure 3). However, at (50%) groundwater concentration, GST activity returned to 1.3-fold with respect to control suggesting saturation. Indeed, measurement of GST activities follow- ing 14 days exposure to groundwater at (3.1%) and (50%) revealed no significant effects although the mean activity reached 1.2 and 1.6-fold respectively compared to control. The samples collected from groundwater at 1.2 km and 5.4 km did not trigger the increase in GST activity compared to control.

Biomarkers of tissue damage

Lipid peroxidation expressed by TBARS levels in liver of trout were no significantly different compared to control trout (Figure 4). This result was observed for all concentrations tested. Positive correlation was obtained between LPO and GST (r=0.88, P 0.05) in group at 5.4 km (

Table 4. Biomarkers correlation in rainbow trout exposed for 4 and 14 days to Mercier groundwater at 1.2, 2.7 and 5.4 km.

GST, gluthatione S-transferase; EROD, ethoxyresorufin O-deethylase. *Only significant correlations are shown (P 0.05).

). Exposure of trout to groundwater at 1.2 km caused a significant elevation of DNA strand breaks reaching 1.2-fold after 4 days exposure at (50%) concentration and (2-fold) after 14 days exposure at (3.1%) water concentration (Figure 5). At 5.4 km distance from the lagoons, a much stronger induction (reaching 3.5 fold induction at 50%) compared to control was observed after 14 days. However, DNA Strand breaks were somewhat lower (1.0 and 1.3 fold) in liver of fish exposed to 3.1% and 50% groundwater respec- tively at 2.7 km for 4 days. A slight raise was observed after 14 days exposure (1.57 and 2.02 fold) compared to control. In addition, significant correlation was found in groups at 2.7 km between DNA strand breaks and LPO levels (r=0.99), P 0.05) (Table 3). However, negative correlation was between DNA strand breaks and EROD activity (r= 0.91), P 0.05) (Table 4).

Discriminant function analysis

In the aim to describe the biochemical effects of chronic exposure (14 days) to contaminated groundwater on the liver of rainbow trout, dis- criminant function analyses were performed (Figure 6). The main biomarkers were identi- fied on X and Y-axis. Discrimination function analysis of the biochemical responses revealed that all three wells at 2.7 km, 5.4 km and 1.2 km and the controls were correctly classified: 81%, 83%, 62% and 100% correctness respectively. EROD was the main biomarker that discrimi- nated the sites.

Discussion

The aim of this study was to investigate the biochemical effects of exposure to groundwa-ter at increasing distance from an industrial dumping site (Mercier lagoons) in rainbow trout and determine the behavior of the con- tamination plume in groundwater.

The concentrations of volatile organic com- pounds (e.g., benzene, dichloroethane and vinyl chloride) measured in samples at 2.74 km were 0.29 µgL-[1], 2 µgL-[1] and 4 µgL-[1] respective- ly, which indicates that the plume is more con- centrated at this distance (Table 2). The pres- ence of VOCs in the groundwater could lead to serious health threats. The chronic exposure of organisms to low levels of VOCs can caused carcinogenesis,[38] i.e. vinyl chloride and ben- zene are widely recognized human carcino- gens.[8,39] However, DNA damage was more important at the closest and farthest sampling distance (i.e. 1.2 and 5.4 km). Hence, the com- plexity of the fate and transport of chemical mixtures in groundwater makes it difficult to suggest causative chemicals responsible for the observed effects.

The CF and HSI of the trout were homoge- neous in all exposure groups which removes size-related influences in the biomarker responses. It also suggests that the observed response were not immediately threatening at the fish morphological level. Fish from polluted environments generally show an increase in the HSI, exposition to PAHs or others sub- stances which involved biotransformation leads normally to an elevation of this index.[40] However, no significant increase in HSI was observed after 14 days of exposure time. May be the exposure period was not long enough to see HIS variations.

The data obtained in this study would sug- gest the presence of chemicals able to activate the Ah receptor. Indeed, EROD activity of groundwater exposure, were significantly enhanced over fish from the control at 1.2 km (50%) and 5.4 km (3.1%), but was reduced at 2.7 km for all times of exposure at (3.1%) and (50%) concentrations. The reduction of EROD activity in fish at site 2.7 km could be due to hepatotoxic damage which may inhibit the liver cells production of this enzyme or we hypothesized the presence of the blocker (antagonist) for AhR which reduce the enzyme activity mediated normally by AhR in trout. Whyte et al.[41] suggested that toxicity caused by xenobiotics accumulation and their metabo- lites might inhibit enzyme production leading to an inhibition of EROD activity. Biotransfor - mation of vinyl chloride, dichloroethane and benzene, the chemicals identified in ground- water samples at 2.7 km, are under cytochrome P450 2E1,[42] as well as through glutathione con- jugation.[43,44] Trans and cis 1,2-dichlorethylene inhibit their own metabolism in viv o by inacti-vation of the metabolizing enzyme presumably the CYP 450 isoform CYP 2E1.[16] The induction of GST activity as for CYP450 is regulated, in part by the Ah receptor.[25] Indeed, an increase in hepatic GST activity has been reported in several studies after fish exposure to PAHs, PCBs and certain VOCs.[17,45,46] Various sub- strates involved in GST activity are recognized by the binding site of hydrophobic substrates.[27] Nevertheless, the enzyme GST has been reported to respond differently to different sub- stances. For example, Otitoloju et al.[18] reported that enzyme was inhibited by benzene, while Qian et al.[47] reported increased GST activity in liver of Crucian carp (Carassius auratus) injected with chlorobenzene. In our study, a significant increased displayed in the phase II biotransformation in trout exposed to 3.1% groundwater at 2.7 km after 4 days exposure. The GST response to groundwater after 4 days exposure in all groups shows a bell-shaped trend with an initial increase in activity, but significantly only at 2.7 km. Accordingly, the low enzyme activities after 14 days in fish from 12 km, 2.7 km and 5.4 km could be associated with deficiency to compensate for oxidative stress, possibly due to high levels of pollutant exposure.[48] Since LPO was not significantly affected and was not related to GST activity, we cannot support the argument that decreased GST activity was associated to oxidative stress. Oxidative stress can be mediated by numer- ous organic contaminants, including, halo- genated hydrocarbons, PAHs, and dioxins.[18] In the current study, LPO levels in trout exposed to groundwater at all distance increased slight- ly even if non-significantly. Although LPO was positively correlated to GST, the change in GST activity seemingly did not influence LPO under analysis of covariance.

DNA damage was measured by the levels of soluble strand breaks and can be used as tools to monitor genotoxicity in organisms.[49] At 4 days exposure, an initial increase in DNA strand breaks was observed in trout exposed to 50% groundwater at 1.2 km. Although, after 14 days of exposure to groundwater at 5.4 km, the amount of DNA were more elevated (reaching 3.4-fold) with respect to control. In fact, Devaux et al.[50] observed that the chub (Lenciscus cephalus) caught in Rhone River close to an industrial area presented DNA dam- age in erythrocytes. Genotoxic effect seems to be reflected by metabolites effects of chemicals on DNA, including their capabilities to lead to DNA strand. According to Gwinn et al.[9] which reported that EDC, with its strong electrophilic affinity to DNA appears to induce DNA dam- age. Vinyl chloride oxidation produces an epoxide intermediate which reacts with a vari- ety of cellular nucleophiles including DNA and glutathione (GSH).[16] Mattes et al.[8] reported likewise that DNA damage of vinyl chloride is due to the formation of the extremely reactive epoxide chlorooxirane during the oxidative metabolism of VC by CYP 450 in the liver. Unfortunately, we did not measure CYP2E1 activity in this study. However, DNA damage is not correlated with EROD (r= 0.91, P 0.05), this suggests that VOCs are associated with genotoxicity.

Conclusions

The data revealed that exposure to contami- nated groundwater stimulated both phases I and II biotransformation activities as evi- denced by EROD and GST activities. The water samples were also genotoxic which was seem- ingly not related to oxidative stress given the lack of LPO changes. This study provides some insights on the toxicity of groundwater con- taminated by industrial waste to rainbow trout.