Induction of the cytochrome P450 enzymes in fish, measured as ethoxyresorufin-O-deethylase (EROD) activity, has been extensively used as a biomarker in assessing exposures and responses of aquatic organisms to contaminants. This study focused on whether exposure to Troy (Alabama) wastewater treatment plant (TWWTP) effluent at the Walnut Creek mixing zone, induced transcription of mRNA for cytochrome P4501A1 enzyme production and increased EROD activity in channel catfish (Ictalurus punctatus). Water from Big Creek (Barbour County, AL), regarded as least impacted by pollutants, was used as a reference source for administration of a positive control chemical (PCB – Aroclor 1254). All water samples were transported from field sites to the Troy University laboratory for conducting the study. Reverse transcriptase-polymerase chain reaction (RT-PCR) indicated the presence of compounds capable of inducing transcription of CYP1A1 mRNA in catfish liver. Increased transcription of mRNA for cytochrome P4501A1 enzyme production, confirmed by a 3-fold induction of EROD activity, was found in catfish exposed to water from the TWWTP mixing zone on Walnut Creek compared to catfish exposed to Walnut Creek Upstream. Catfish exposed to water from Big Creek treated with PCBs were found to have only slightly higher enzyme activities than those exposed to water from Big Creek Control, but a 2.7-fold EROD level was found in catfish exposed to Big Creek Control compared to Walnut Creek Upstream. Determining the induction of cytochrome P450 and subsequent enzymatic activity in catfish and other fish species common to this region could be useful early molecular warnings of possible pollution effects, beyond those methods typically used to characterize water quality.
The responses of various xenobiotic metabolizing enzymes in fish are important biomarkers for monitoring exposure to unacceptable levels of environmental contaminants (Široká and Drastichová, 2004). Several studies have reported on the induction of various molecular biomarkers in fish living in polluted aquatic environments by metals, organics, and stress (Kloepper-Sams and Benton, 1994; Schlenk et al., 1997; Nadeau et al., 2001). Biomarkers respond to contaminants at concentrations that may be too low to be detected by current laboratory water quality analysis techniques (Široká and Drastichová, 2004). Dilution and dispersion of chemicals can produce concentrations that are not immediately lethal, but can affect the future survival of the organism exposed (Haasch et al., 1993). Even when quantitative and accurate analytical procedures are available, they may not be adequate for assessing the impacts of chemical mixtures (Arinc et al., 2000). Analysis may be difficult or impossible because the identity of the contaminant may be unknown. Furthermore, levels of contaminants in water may fluctuate and their availability for uptake by organisms may vary (Walker et al., 2006). The use of biomarkers reduces some of the limitations of traditional environmental risk assessment by directly measuring the toxic effects of the compound, a mixture, or product of chemical interactions, as it becomes available to the affected species (Klaassen and Watkins, 2001).
Induction of cytochrome P450 (CYP) enzyme activity, often measured as ethoxyresorufin-O-deethylase activity (EROD) (Burke and Mayer, 1974), has been used extensively as a biomarker of exposure following xenobiotic contamination of the aquatic environment (Burgeot et al., 1996; Flammarion et al., 1999). The CYP1 enzyme family (subfamily CYP1A1) has been shown to be an early molecular indicator for detecting exposures to contaminants, such as PAHs and complex industrial effluents, in aquatic systems (Campbell et al., 1996; Machala et al., 1997; Machala et al., 2000). The ability of certain types of aquatic pollutants to induce CYP in fish and also the potential for this property to be used as a monitoring tool, has advanced the need to examine the relationship between water quality and CYP1A1 activity in fish (Melancon et al., 1987).
The concentrations of many different compounds in wastewater treatment plant effluent can be sufficiently high to be of environmental concern (Petrasek et al., 1983). The City of Troy (Alabama) Wastewater Treatment Plant (TWWTP) treats wastewater up to the secondary level and discharges its effluent into Walnut Creek, a third-order stream in the Choctawhatchee River basin, in accordance with Alabama Department of Environmental Management (ADEM) permit AL0032310, under the National Pollutant Discharge Elimination System (ADEM, 2007). Since secondary treatment is not sufficient to remove 100% of toxic compounds, the effluent must be tested for other parameters. Chemical analyses of TWWTP effluent have found increased conductivity, and several increased inorganics and nutrients below the discharge zone. Chronic toxicity testing of effluents using Ceriodaphnia dubia and Pimephales promelas, have both passed and failed at different times. Nonetheless, ADEM indicates that Walnut Creek at the TWWTP typically meets its water use classification for fish and wildlife (ADEM, 1997). While these analytical and toxicological studies are conducted periodically, no studies have examined the biomarker responses of native fish to TWWTP effluent.
The channel catfish (Ictalurus punctatus), which is an economically important species native to the southeastern United States (Mettee et al., 1996), can be used as a sentinel species for biomarker studies in aquatic systems of this region (Schlenk et al., 1997). Channel catfish have been used in previous biomarker studies, which resulted in the induction of hepatic CYP1A enzyme and EROD activity due to exposure to various PAHs (Hill et al., 1976; Rice and Schlenk, 1995; Rice and Roszell, 1998; Burton et al., 2002), as well as from other unknown chemicals at contaminated sites (Haasch et al., 1993).
The present study examined the response of channel catfish (I. punctatus) exposed to TWWTP effluent by measuring molecular biomarkers that have been shown to be early molecular sentinels of exposure to pollution. The objectives of this study were to determine whether the TWWTP effluent induced transcription of mRNA for the drug metabolizing enzyme cytochrome P450 and whether the effluent increased the enzymatic activity (EROD) in channel catfish.
Materials and methods
Design of study
The experiment involved exposing hatchery-reared channel catfish under four non-replicated exposure conditions: (1) Walnut Creek Upstream (water collected one mile upstream from the TWWTP at the Elm Street intersection, Troy, Alabama), (2) Walnut Creek Treatment (water collected at the TWWTP effluent mixing zone), (3) Big Creek Control (water collected at the creek's intersection with Alabama Highway 10, Barbour County), and (4) Big Creek Treatment (creek water that had been treated with 12 μ g l−1 of PCB, Aroclor 1254, a known inducer of EROD activity in channel catfish) (Hill et al., 1976). Previous studies by Bennett et al., (2004) identified Big Creek as least impacted by land use and pollution based on a high Invertebrate Community Index (ICI) score; therefore, Big Creek was proposed as a reference water source.
Water from each field site was transported to the laboratory at Troy University (Troy, Alabama) at the beginning of the study and again on day five. In the laboratory, water in exposure tanks was continuously recirculated, aerated and filtered through a biofiber (Aqua-Tech, Aquaria Inc., Moorpark, CA) designed to reduce ammonia and nitrate levels. Biofibers provide an environment for cultivation of aerobic bacteria such as nitrosomonads and nitrobacters that convert these fish waste products to less harmful substances. Since highly chlorinated PCB congeners need to undergo reductive dechlorination by anaerobic bacteria before any aerobic degradation can occur (Abramowicz, 1995), it is unlikely that there was any interaction between the Aroclor 1254 and biofibers. Activated carbon was removed from filters so as not to alter other water constituents. Additional water from each field site was independently maintained in separate polyethylene drums and recirculated until needed; no filtration was applied.
Chemicals and materials
PCB (Aroclor 1254) was purchased from Accustandard, Inc. (New Haven, CT). Potassium chloride (KCl), 7-ethoxyresorufin, resorufin sodium salt and nicotinamide adenine dinucleotide phosphate (NADPH) were purchased from Sigma Chemical Company (St. Louis, MO). Tris was purchased from Angus Buffers and Biochemicals (Niagara Falls, NY), and ethylenediamine tetraacetetic acid (EDTA), agarose, and Tween 80 from Fisher Scientific (Fairlane, NJ). Dithiothreitol (DTT) and protein assay kits were purchased from Pierce Biotechnology (Rockford, IL). Ninety-six well microplates (Falcon) were purchased from Ward's Natural Science (Rochester, NY). The RNA Isolation System and Access RT-PCR System were purchased from Promega (Madison, WI) and primers from Invitrogen Corporation (Carlsbad, CA). Water quality test kits were purchased from LaMotte Company (Chestertown, MD).
Treatment of animals
Channel catfish fingerlings, 6.5 to 9.5 cm long and 5.5 gm mean weight, were obtained from Easterling Fish Hatchery (Clio, Alabama) and acclimated in recirculated, aerated, and filtered water from either Walnut Creek, upstream of the TWWTP, or from Big Creek for seven days prior to commencement of the study. Catfish were maintained on a 12-hour daylight:12-hour night photoperiod throughout the study and were fed ad libitum daily at a rate about equal to 2% of their body weight with catfish fingerling grower (Cloverbrand® Quality).
Following acclimation in Walnut Creek Upstream water, twenty catfish were transferred to a 176-litre polypropylene tank containing water from Walnut Creek Upstream, and twenty catfish were transferred to a 176-litre tank containing water from Walnut Creek Treatment. Following acclimation in Big Creek water, twenty catfish were transferred to a 176-litre tank containing water from Big Creek Control, and twenty catfish were transferred to a 176-litre tank containing water from Big Creek to which 12 μ g l−l PCB emulsified in Tween 80 had been added.
Three catfish were collected at days 0, 1, 3, 6, and 13, with day 0 considered to be the beginning of the exposure period and day 1 to be twenty-four hours after the exposure period began. By day 6, all catfish in Walnut Creek Upstream had either been sampled or died. The remaining catfish from the three other tanks were sampled at day 13. Catfish were sacrificed and livers were excised and weighed, rinsed of blood in ice cold 0.15 M KCl and blotted dry with disposable tissue wipes (Burke and Mayer, 1974). Livers were cut into halves (½ was reserved for RT-PCR and the other ½ for protein analysis and EROD activity assays) and stored at −80 °C.
Liver samples from the four exposure groups were examined for EROD activity by fluorimetric assay. Cytochrome P450 enzyme production and EROD activity may be suppressed by metals, stress, and defense mechanisms (Široká and Drastichová, 2004); however transcription of the CYP1A1 gene is less susceptible to endogenous (such as steroids) and exogenous factors (Campbell and Devlin, 1996). Detection of CYP1A1 mRNA provides supportive information on enzyme induction (Förlin et al., 1994). Therefore, RT-PCR was used to determine if compounds were present under exposure conditions that were capable of inducing transcription of CYP1A1 in catfish.
Ammonia and nitrate analyses were performed using a HACH DR/4000® spectrophotometer every other day throughout the study. At least 95% of the water from each tank was changed on days 3 and 6 in order to keep ammonia below toxic concentrations (0.5–1 mg l−1) for catfish (Tucker, 1991). PCB was added to restore the concentration to 12 μ g l−1 in the Big Creek Treatment tank (positive control) when water was replaced.
Microsome preparation and enzyme assays
Microsomes for each liver sample were prepared using the method described by Chan (2005) and stored at -80 °C until assayed. Protein concentrations of microsomal preparations were determined using a BCA™Protein Assay kit with bovine serum albumin as the standard (Stephensen et al., 2003).
Ethoxyresorufin-O-deethylase activity in liver samples was measured as described by Rice and Roszell (1998) using a Cytofluor™ 2350 plate reader. Fluorescence was measured at an excitation wavelength of 530 nm and emission wavelength of 590 nm at two sensitivities (3 and 4).
RNA isolation and RT-PCR
Total RNA was isolated from 25–30 mg liver tissue samples according to the guidelines in Promega's SV Total RNA Isolation System (Promega, SV Total RNA, 2006). Total RNA concentrations and A260/A280 ratios (between 1.85 and 2) were measured using an Eppendorf Biophotometer at 260 nm and 280 nm. Purified RNA was stored at −80 °C until needed for further analyses.
Cytochrome P4501A1 mRNA was detected by RT-PCR using Promega's Access RT-PCR System (Promega, Access RT-PCR, 2006). The primers used in this study, CYP1A1-1 [5′ -TTCACCATC/TCCICACA/TGCAC-3′] and CYP1A1-3 [5-′ CCAG/AGAAGAGGAAGACCTC-3′], were designed by Campbell and Devlin (1996) to anneal to highly conserved regions of published CYP1A1 sequences. CYP1A1-1 and CYP1A1-3 flank a 600 bp region in the CYP1A1 gene. Since the CYP1A1-1 primer spans an intron that will not amplify for the CYP1A1 gene, expectations were that only a 350 to 400 bp fragment of the channel catfish gene would be amplified by RT-PCR. The concentration of total RNA used in the RT reaction for each sample was 0.5 μ g in a total reaction volume of 50 μ l. CYP1A1 mRNA present in samples was reverse transcribed to cDNA and amplified. Each PCR product was run on 1.5% agarose gel at 150 V. Gels were stained with ethidium bromide and photographic images were obtained using a UV photodocumentation system (Epi Chem II Darkroom, UVP).
Single band identification and semi-quantification using the LabWorks Image Acquisition and Analysis software package (Version 126.96.36.199) was completed for each RT-PCR product following agarose gel electrophoresis. Since, the sequence information indicated that the 400 bp band is likely the CYP1A homologue (Campbell and Devlin, 1996), the area of interest (AOI) was selected from the region of the gel that correlated with the predicted product size. When necessary, manual correction was performed to insure the correct band was quantified. The absolute integrated optical density of each band was computed and the values were standardized relative to the internal standard (400 bp molecular weight marker from the gel containing each specific sample).
Hepatic EROD activity
Hepatic EROD activity was measured as nmols of resorufin per minute and referenced to the amount of protein in liver samples (Table 1). Hepatic EROD activity was higher in fish exposed to Walnut Creek Treatment than fish exposed to Walnut Creek Upstream (Figure 1). After day 1 exposure, the mean EROD activity was more than 3-fold higher in Walnut Creek Treatment than Walnut Creek Upstream. The mean EROD for Walnut Creek Treatment remained 2-fold higher than Walnut Creek Upstream on day 3 and ½-fold higher on day 6. Since all fish from Walnut Creek Upstream had been sampled or died by day 6, no samples were available for comparison for day 13. Only slight increases in daily and overall EROD activity occurred in catfish exposed to Big Creek Treatment over Big Creek Control (Figure 2). Hepatic EROD activity was 2.7 times higher in Big Creek Control compared to Walnut Creek Upstream at day 0 of the study and remained higher throughout the study (Figure 3).
The presence of a RT-PCR product at 400 bp in electrophoresis gels (Table 1) suggested that detectable levels of CYP1A1 mRNA were present. Optical densities of RT-PCR product at 400 bp were reported relative to the internal standard and indicate only relative values (Table 1). The range of background levels on the negative control lane on analyzed gels were from 0.23%–9.4%; samples from gels with background levels exceeding 3% were considered invalid and excluded from analysis.
Relative optical densities indicated a general correspondence between EROD and mRNA levels. Hepatic EROD for Walnut Creek Treatment was more than 3-fold higher than Walnut Creek Upstream on day 1 and 2-fold higher on day 3. Similarly, the relative optical density of mRNA for Walnut Creek Treatment was more than 2-fold that of Walnut Creek Upstream on days 1 and 3. Both EROD values and mRNA relative optical densities were at their peaks in Walnut Creek Treatment on day 1. On day 0, the relative optical densities of mRNA in Big Creek Control and Walnut Creek were 6.5 and 3.9 respectively, which corresponds to the more than 2-fold Big Creek Control EROD (11.14 nmol min−1 mg−1 protein) over Walnut Creek Upstream (4.08 nmol min−1 mg−1 protein). By days 6 and 13, EROD values in all samples had decreased; likewise, mRNA optical densities were greatly reduced or not discernible.
Some wastewater treatment plant effluents may cause adverse toxicological effects on fish populations (Schmidt et al., 1999; Giesy et al., 2003). However, the potential for these adverse effects may go undetected without the use of early warning systems.
The main objective of our study was to examine the impact on EROD induction following exposure to TWWTP effluent. Results showed 3-fold induction of EROD activity in catfish exposed to water from Walnut Creek at the TWWTP effluent mixing zone (Walnut Creek Treatment) compared to those exposed to Walnut Creek Upstream. Similar results (3.5-fold induction) were obtained in the study performed in chub (Leuciscus cephalus) exposed to WWTP effluent (Kosmala et al., 1998). Our conclusion that TWWTP effluent increases EROD activity in catfish was confirmed by the semi-quantitative assessment of increased CYP1A mRNA in catfish exposed to TWWTP effluent.
Variation in EROD activity and RT-PCR product within some groups was high, but similar results have been reported by other researchers (Kloepper-Sams and Benton, 1994; Kosmala et al., 1998). Some studies pool liver tissues prior to analyses to reduce variations (Jimenez and Burtis, 1989; Rice and Roszell, 1998). Despite these internal variations, EROD activity was consistently higher in channel catfish exposed to Walnut Creek Treatment than those exposed to Walnut Creek Upstream, thus identifying the TWWTP as a source of compounds that were capable of inducing cytochrome P450.
Big Creek was selected as a reference source of relatively unpolluted natural water to be used for administration of PCB as a positive control chemical. Only slight increases of EROD activity were found in catfish exposed to Big Creek water treated with PCB over Big Creek Control. Possible explanations for these findings include the presence of compounds in Big Creek that act as PCB antagonists or inactivators, or the presence of other PCB congeners. Investigations have demonstrated that di-ortho-substitution of polychlorinated biphenyls by halogenated hydrocarbons, when present in combination with CYP1A1 agonists (TCDD, non-ortho-substituted PCB, etc.) inhibited induction by these agonists (Suh et al., 2003). Previous studies have reported inhibition or suppression of EROD activity in channel catfish when high doses of tributyltin (TBT) and low doses of PCB were administered concurrently, while administration of PCB alone showed significant induction of EROD activity (Rice and Roszell, 1998). Inhibition of CYP1A response was also observed in plaice (Pleuronectes platessa) when exposed to high levels of PCB mixtures (Boon et al., 1992). Possible metal or pesticide contaminants in Big Creek may have prevented translation of CYP1A1 mRNA into a functional enzyme as observed in flounder (Platichthys flesus), where CYP1A activity was lower in cadmium-benzo[a]pyrene administered flounder than when benzo[a]pyrene was administered alone (Beyer et al., 1997). On the other hand, day 0 results for enzyme activity at Big Creek Control were 2.7-fold greater when compared to Walnut Creek Upstream, suggesting that contaminants may be present at Big Creek that were capable of inducing enzyme activity, but may not have had an adverse impact on macroinvertebrate assemblages. The results of our study suggested that the Big Creek water source may not be as unpolluted as indicated by Bennett et al. (2004).
While our study indicates that there are chemicals present that can affect induction of cytochrome P450 enzymes, more research is needed to establish normal ranges of these enzymes, as well as levels that are indicative of non-recuperative toxic effects. Our study suggests that tests, such as chemical analyses and chronic toxicity studies in Daphnia and fathead minnows that are typically used to characterize the quality and ultimate use of water bodies in this region may not be adequate or sufficiently early predictors of the responses of other native aquatic organisms to wastewater effluents. Consideration should be given to adding biomarkers such as EROD activity and CYP1A1 mRNA to the battery of tests used for predicting and assessing the impact of discharges from TWWTP to Walnut Creek. Further, these biomarkers should also be added to assessments of the impacts of other activities, such as domestic and agricultural runoff, to Big Creek and other water bodies in the region including those thought to be of good quality with no apparent contaminant source.
This study was supported by a faculty development grant from Troy University (Troy, Alabama).