Induction of CYP1A mRNA in Carp (Cyprinus carpio) from the ...

6 downloads 0 Views 366KB Size Report
Abstract. The Kalamazoo River Superfund site in Michigan is contaminated with polychlorinated biphenyls (PCBs), which were heavily discharged into the river ...
Arch. Environ. Contam. Toxicol. 50, 14–22 (2006) DOI: 10.1007/s00244-004-0171-4

Induction of CYP1A mRNA in Carp (Cyprinus carpio) from the Kalamazoo River Polychlorinated Biphenyl-Contaminated Superfund Site and in a Laboratory Study M. A. Fisher,1,2 C. Mehne,1,4 J. C. Means,1,3 C. F. Ide1,2 1 2 3 4

Environmental Institute, Western Michigan University, 1903 West Michigan Avenue, Kalamazoo, Michigan 49008, USA Department of Biological Sciences, Western Michigan University, 1903 West Michigan Avenue, Kalamazoo, Michigan 49008, USA Department of Chemistry, Western Michigan University, 1903 West Michigan Avenue, Kalamazoo, Michigan 49008, USA Animal Clinic, 413 West Mosel, Kalamazoo, Michigan 49007, USA

Received: 18 October 2004 /Accepted: 14 May 2005

Abstract. The Kalamazoo River Superfund site in Michigan is contaminated with polychlorinated biphenyls (PCBs), which were heavily discharged into the river from several paper companies as part of the deinking process in the 1950s through 1970s. We characterized biomarkers of chronic PCB exposure in a resident fish population using real-time reverse transcriptase–polymerase chain reaction to examine mRNA expression levels of multiple genes in carp (Cyprinus carpio) liver from PCB contaminated and reference sites in the Kalamazoo River. We also measured these same genes in juvenile carp exposed to dietary PCBs for 4 months. Kalamazoo River carp had significantly increased levels of cytochrome P450 1A (CYP1A) mRNA as did carp fed PCBs in the laboratory. No significant mRNA upregulation occurred in the specific oxidative stress genes (gamma-glutamylcysteine synthetase and magnesium superoxide dismutase) and metabolic genes (phosphoenolpyruvate carboxykinase and nucleolin) examined. These data are consistent with the idea that carp from the Kalamazoo River Superfund Site are responding to PCB exposure via upregulation of CYP1A independent of activation of the oxidative stress response genes normally thought to be co-regulated with CYP1A.

The Kalamazoo River Superfund site (Michigan) is impacted by environmental contamination from a variety of industrial, municipal, and agricultural sources. Notably, polychlorinated biphenyls (PCBs), commercially synthesized compounds with hepatotoxic, immunotoxic, and neurotoxic activities, were heavily discharged into the river from several paper companies as part of the deinking process in the 1950s through 1970s. The United States Environmental Protection Agency (USEPA) estimated that >1.5 billion pounds of PCBs were manufactured in the United States before their ban in 1977; Kalamazoo River

Correspondence to: M. A. Fisher; email: [email protected]

sediments contain approximately 230,000 pounds (Blasland et al. 2000). Because of PCB contamination, an 80-mile stretch of the Kalamazoo River, from Morrow Dam to Lake Michigan, has been designated a Superfund Site (Fig. 1). Currently, the Michigan Department of Community Health bans consumption of sport fish from within this site. Despite known PCB contamination, fish responses to PCBs onsite remain undefined. We used real-time reverse transcriptase–polymerase chain reaction (RT-PCR) to measure transcriptional expression of multiple genes as potential biomarkers of PCB exposure and physiologic effects in the common carp Cyprinus carpio. Carp are found throughout the Kalamazoo River and feed benthically in heavily PCB-contaminated sediment. Carp are the most contaminated fish in the river, with PCB values as high as 22 mg/kg (combined 1997 and 1999 biennial trend monitoring fish PCB data from Blasland et al. 2000). We therefore predicted that carp would show alterations in gene expression levels associated with exposure to PCBs. To determine the relationship between PCB exposure and gene expression in carp, we raised juvenile hatchery-bred carp in the laboratory, and exposed them through diet to the PCB mixture Aroclor 1242 (A1242) the mixture most commonly found in the Kalamazoo River (Blasland et al. 2000). We also compared gene expression in feral carp caught from PCBcontaminated sites in the Kalamazoo River Superfund area with gene expression in carp from relatively noncontaminated upstream reference sites. We quantified mRNA expression levels in liver in laboratory and feral carp for genes that have been shown to be regulated at the level of mRNA and may indicate exposure, oxidative stress responses, and changes in metabolic function related to PCBs. We first tested if carp were responding to PCB exposure by examining the induction of cytochrome P450 1A (CYP1A) mRNA in liver. CYP1A is a phase I drugmetabolizing enzyme induced by a broad range of environmental contaminants that bind to the aryl hydrocarbon receptor (AhR; Goldstein and Safe 1989). Cytochrome P450 1A induction has been used as a biomarker for exposure to Ah

15

Induction of CYP1A in PCB Exposed Carp

Fig. 1. Map of the Kalamazoo River, Michigan. Carp were sampled from reference sites (Ceresco and upstream of Morrow Pond; dark arrows) located upstream of the Superfund site and from PCBcontaminated sites within the Superfund sites (Lake Allegan, Trowbridge; light arrows). Date of capture and number of carp caught are shown.

receptor ligands (e.g., dioxins and PCBs) in a wide variety of animals (Peakall 1992) including fishes (Sivarajah et al. 1979; Melancon and Lech 1983; Payne et al 1987; Van der Weiden et al. 1989; Stegeman and Lech 1991; Haasch et al. 1993; Whyte et al. 2000; Van der Oost et al. 2003) and is hypothesized to be induced in PCB-exposed carp. In addition to examining CYP1A mRNA expression, we measured expression of specific genes (gamma-glutamylcysteine synthetase [c-GCS] and magnesium superoxide dismutase [Mn-SOD]) predicted to be upregulated to counterbalance the potentially harmful effects of CYP1A activity. c-GCS is the rate-limiting enzyme in glutathione synthesis. Cellular glutathione (L-c-glutamyl-L-cysteinylglycine [GSH]) is a major cellular antioxidant involved in detoxification of electrophiles and has been shown to be increased in fish liver by environmental contaminants including PCB exposure (Thomas and Wofford 1984; Otto and Moon 1995) and b-napthoflavone, an Ah receptor agonist (Hughes and Gallagher 2004). SOD functions to remove O2) electrophiles formed by uncoupling of CYP1A and PCBs (Schlezinger et al. 1999). SOD (Roberts et al. 1987; Livingstone et al. 2000) and MnSOD, the mitochondrial form of SOD (Meyer et al. 2003), have been shown to have increased activities in fish exposed to contaminated sediments. Thus, upregulation of c-GCS and Mn-SOD is expected if potentially damaging reactive oxygen species are produced as a consequence of CYP1A induction. We also used real-time RT-PCR to examine the expression of specific metabolic genes, phosphoenolpyruvate carboxykinase (PEP-CK) and nucleolin, predicted to be down regulated (PEP-CK) and up-regulated (nucleolin) by PCB exposure. PEP-CK is the rate-limiting enzyme in the gluconeogenesis pathway, and PEP-CK mRNA expression has been

shown to be decreased in response to 2,3,7,8-tetrachlorodibenzo-p-dioxin an Ah receptor agonist (Weber et al. 1991; Stahl et al. 1993). Nucleolin is a nucleolar protein that represses ribosomal RNA synthesis in cold-adapted carp (Alvarez et al. 2003) and also modulates DNA and RNA metabolism after oxidative stress (Daniely et al. 2002). In this study, real-time RT-PCR was used to quantify expression levels of several genes hypothesized to be differentially expressed in carp exposed to PCBs. Laboratory feeding and field studies were performed to assess responses to PCB exposure. Our results show that carp from the Kalamazoo River Superfund site and carp exposed to PCBs in the laboratory respond to exposure and, interestingly, do not show differences in select oxidative stress and metabolic genes, also hypothesized to be affected by exposure.

Materials and Methods Laboratory Raised Carp Maintenance. Juvenile carp were purchased from Aquatic Research Organisms fish hatchery (New Hampshire). Carp were allowed to acclimatize for 2 months before the start of the experiment. Light and dark cycles were set for 12 hours light and 12 hours dark. Water temperature in the tanks was 18 € 2C, and water pH ranged from 7.5 to 8.0 during the experiment. Carp were fed 0.6% of their body weight 2 · d. Health, feed, oxygen checks, and tank siphoning were performed daily, and aquaria NH3, NO2 and NO3 levels were monitored every 2 weeks to ensure that all levels were non-toxic or zero. There was no fish mortality or signs of health problems during the

M. A. Fisher et al.

16

acclimatization period or the experiment. All experiments were performed according to an approved Institutional Animal Care and Use Committee at Western Michigan University.

PCB-Exposures in Diet. Eighty carp (8.02 € 0.38 g) were randomly placed in 10 10-gallon aquaria. Carp were fed a standard diet (Laguna Goldfish and Koi Chow, Rolf C. Hager (USA) Corp. Lansfield, MA, medium floating pellets) either mixed with the commercial PCB mixture Aroclor 1242 (AccuStandard, New Haven, CT) or vehicle only. Fish in five tanks were fed food mixed with PCBs, and fish in the other 5 tanks were fed food with vehicle only. Treatment food was made by adding 7.2 ml 500 ug/g stock solution of Aroclor 1242 in hexane to 343 ml 95% ethanol for a calculated concentration of 12 ug/g Aroclor 1242. Control food was made by adding 7.2 ml hexane only, without Aroclor 1242, to 343 ml 95% ethanol. The treatment and control solutions were each added to 300 g fish food and left to mix on a rotary shaker in a ventilated fume hood for 48 hours. A second batch of food was made from the same stock solution of 500 ppm Aroclor 1242 in hexane, but it was scaled to add 150 g fish food for the remainder of the experiment. Control food was made similarly by adding hexane alone to 95% ethanol. A target dose of 12 lg/g (ppm) diet PCB concentration for the PCB treatment food was chosen for this laboratory feeding experiment. 12 lg/g diet is believed to reflect source PCB levels in sediments and lower-trophic organisms present in the Kalamazoo River Superfund site areas. Control food had a low but detectable level of PCB contamination (0.7 lg/g dry wt and 0.6 lg/g dry wt). Treatment food had a PCB concentration of 14.9 lg/g dry wt. and 18.1 lg/g dry wt., respectively.

Samples Procedure for Real-Time RT-PCR and PCB Analysis. After 1, 2, 3, and 4 months of feeding, 1 carp from each of 10 tanks was sampled for real-time RT-PCR and PCB analysis, and 1 carp from each of 10 tanks was sampled for histologic examination. For real time RT-PCR analysis, carp were randomly chosen, netted, killed, and completely processed within 30 minutes. Carp were killed by treatment with 1:500 tricaine methanesulfonate (MS-222) for approximately 1 minute. The weight and length were measured and the viscera removed, immediately frozen in liquid N2, and stored at )80C until use. Muscle filets from these carp were stored at )20C until PCB analysis. For histologic sampling, carp were killed by treatment with 1:500 tricaine methanesulfonate (MS-222) for approximately 1 minute, the visceral region was exposed and the carp were fixed in 10% neutral buffered formalin. Subsequent histologic analysis of visceral sections (6 lm) batch-stained using hematoxylin and eosin did not show significantly different hepatocyte profiles, and was not pursued further.

Kalamazoo River Carp Sampling for Real-Time RT-PCR Processing and PCB Analysis. A total of 18 carp were caught from the Kalamazoo River in late spring and summer 2001. Carp were sampled from two upstream Kalamazoo River sites that have low background levels of PCB contamination (Ceresco and just upstream of Morrow Lake) and from two Kalamazoo River sites known to be contaminated with PCBs (above Trowbridge Dam and in Lake Allegan) (Fig. 1). Average PCB values of the surface layer from sediment cores sampled below Morrow Lake (0 to 2 inches), above Trowbridge Dam (0 to 6 inches), and Lake Allegan (0 to 2 inches) are 0.9 € 2.7, 12.3 € 4.0, and 3.4 € 3.0 ppm, respectively, with PCB values as high as 86.0 ppm (Blasland et al. 2000).

Carp were caught with a seine net (Trowbridge Dam n = 4), fishing rods (Lake Allegan n = 8 and upstream of Morrow Lake n = 2), or electroshocked (Ceresco n = 4). Carp were immediately brought back to the laboratory live in holding pens, killed with MS-222 (1:2000) for approximately 1 minute, measured, and dissected. For real time RTPCR analysis, liver was removed, immediately frozen with liquid N2, and stored at –80C until use. Additional liver tissue was sampled and stored at )20C for PCB analysis.

RNA Extraction Total RNA was extracted using Qiagen RNeasy midipreps from either 150 (€ 30) mg wet wt. (Kalamazoo River carp and laboratory carp after 1, 2, and 3 months of feeding) or 240 (€ 30) mg wet wt. (laboratory carp after 4 months of feeding) liver tissue. In addition, 50 ng yeast mRNA (Clonetech) was spiked into each total RNA sample before binding of RNA to the RNeasy membrane as an exogenous internal control for RNA extraction efficiency. Total RNA integrity for extractions was verified with denaturing 1.25% precast agarose gels (Sigma) and quantified using a GeneQuant pro Spectrophotometer (Pharmacia Biotech, Cambridge, UK). After total RNA extraction, each sample was extracted for messenger RNA using oligo d(T) cellulose (Ambion MicroPoly (A) Pure). Messenger RNA for each carp liver was quantified using a GeneQuant Pro Spectrophotometer and stored at )80C until use.

Real-Time RT-PCR Primer and Fluorogenic Probe Design. Gene sequences used for primer and probe design for reverse transcription of carp liver mRNA and subsequent PCR cycling were obtained from the National Center for Biotechnology Information Web site. Primers and dual-labeled fluorogenic probes were designed using Primer Express software (version 1.1; Applied Biosystems, Foster City, CA). Sequences and concentrations for forward and reverse primers are listed in Table 1. Primer concentrations were optimized according to the manufacturer's recommendations. Probes were run using a concentration of 200 nM.

Sample Preparation. Real-time RT-PCR for each carp liver mRNA sample was performed using Taqman One-Step RT-PCR Master Mix Reagent Kit (Applied Biosystems). For each gene, an identical amount (4 ng/ll/well) of mRNA template was added. Taqman probe (0.5 ll/well), optimized forward primer, and optimized reverse primer were also added for each gene. Diethylpyrocarbonate treated water was additionally added to bring the final well volume to 25 ll. For each gene tested in each individual liver mRNA sample, a ‘‘no reverse transcriptase enzyme’’ control was included on the PCR plate. This control, used for each gene, allows a quantitative measure of unwanted amplified DNA. For all PCR runs, ‘‘no reverse transcriptase enzyme’’ control threshold values for all genes were a minimum of 10 PCR cycles behind threshold values obtained with reverse transcriptase present. Thus, DNA amplification accounted for no greater than 0.1% of nucleic acid quantified for all samples. Additionally, for each gene, a four-point standard curve and a ‘‘no mRNA template’’ control were also included on each 96-well PCR plate. Standard curves were prepared using 1, 2, 4, and 8 ng/well mRNA for all genes for all laboratory fish and field fish examined (except for Mn-SOD after 90 and 120 days of exposure, for which 2, 4, 8, and 16 ng were used). Standard curve template mRNA values were chosen that generated correlation coefficients near r2 = 0.99 and had a slope of )3.3 € 0.3, indicating high PCR reaction efficiencies. Amplification was per-

17

Induction of CYP1A in PCB Exposed Carp

Table 1. Primer and probe sequences and volumes used for real time RT-PCRa Gene CYP1A Forward primer Reverse primer Probe Mn-SOD Forward primer Reverse primer Probe c-GCS Forward primer Reverse primer Probe PEP-CK Forward primer Reverse primer Probe Nucleolin Forward primer Reverse primer Probe Yeast actin Forward primer Reverse primer Probe

Sequence

Amount used

CTGTGGCCAACGTGATCTGT CGAACTCGTCGCTCAAATTG 6FAM-TGCTTCGGCCGGCGCTACA-TAMRA

300 nM 300 nM 200 nM

GGCTTTGATAAGGACAGTGGAAGA GAAGTGGGACGAGACCTGTAGTG 6FAM-TCTTGGTTCCCACATGCAGCAATCCT-TAMRA

900 nM 300 nM 200 nM

AGACTCTTCTCACAGTTACGTCATTCC CAACAGGTGTGGGTTTGTATTCTG 6FAM-AGGCTGCCCAGGCTTCACTCAGC-TAMRA

900 nM 300 nM 200 nM

TCACATCTTCCTGACACCAGACA GCAATGCGTGAGGCGATAC 6FAM-CCGTAACCGCTGCCGAAGGACA-TAMRA

300 nM 300 nM 200 nM

TGCACGCACGTTGTTCGT TTAAACTCCAGGTATGCGATTCC 6FAM-AAGAACCTGCCCTACTCCATAACGCAGGA-TAMRA

50 nM 300 nM 200 nM

TGGATTCCGGTGATGGTGTT TCAAAATGGCGTGAGGTAGAGA 6FAM-CTCACGTCGTTCCAATTTACGCTGGTTT-TAMRA

50 nM 300 nM 200 nM

CYP1A-Cytochrome P450 1A. GCS-Gamma-glutamylcysteine synthetase. Mn-SOD-Magnesium superoxide dismutase. PEP-CK-Phosphoenolpyruvate Carboxykinase. RT-PCR-Reverse transcriptase–polymerase chain reaction. a Primer and probe concentrations were 10 lM.

formed using an ABI Prism 7700 Sequence Detection System. Thermal cycling parameters for all RT-PCR runs were set to 48C, 30 minutes, 10 minutes at 95C, then 40 cycles of 95C for 15 seconds alternating with 60C for 1 minute. Data Analysis. After each PCR run, a threshold value (CT) was manually set in the exponential phase of amplification of the PCR reaction to relate the normalized reporter fluorescent signal to PCR cycle number. Data were then quantified using the relative standard curve method (Applied Biosystems). Using this method, four-point standard curves for each gene were used to relate (CT) to the log of starting mRNA concentrations. Messenger RNA from control carp after 1, 2, 3, and 4 months of feeding, or from carp caught from reference sites in the Kalamazoo River, were used for standard curves. Using the standard curve for each gene, the relative amount of unknown mRNA was then determined from CT values using linear regression analysis. Samples for standard curves and unknowns were run in duplicate. To correct for extraction efficiency, 50 ng yeast mRNA was spiked into each carp liver sample before total RNA extraction to serve as an exogenous internal control as previously reported (Jelaso et al. 2003). Using an exogenous control circumvents normalizing by ‘‘housekeeping’’ genes, which have been shown by way of real-time RT-PCR to vary (Kim and Kim 2003). In each real-time RT-PCR run, primers and probe for yeast actin mRNA were additionally used to quantify relative yeast actin values for each individual. A four-point standard curve for yeast actin mRNA was also generated using identical mRNA template concentrations. Final PCR values from individual carp liver

mRNA analyses were normalized by dividing target gene mRNA PCR values by yeast actin mRNA PCR values.

PCB GC/MS Analysis Extraction and GC/MS Analysis of PCB Congeners. In laboratory-exposed carp, muscle tissue (approximately 0.5 g) was scraped using a spatula between the pectoral and anal fin. In carp caught from the Kalamazoo River, liver tissue (approximately 0.5 g) was analyzed. Carp samples were analyzed for PCBs using a matrix solid-phase dispersion method as previously reported (Fisher et al. 2003) and the PCB congeners were quantified by the multiple selected-ion monitoring, gas chromatography–mass spectometry method adapted from Means (1998) and McMillin and Means (1996). Analysis was performed on 2 ll samples of the extracts using an Agilent Technologies (Palo Alto, CA) 6890A gas chromatograph equipped with the capillary column (AG DB-5MS) (30 m x 0.025 mm i.d.), which was directly interfaced to an AG 5973N mass selective detector equipped with a 7683 autosampler. The congener groups were quantified using a calibration standard containing a representative congener from each chlorination group (AccuStandard, New Haven, CT).

Determination of Method Detection Limits. Detection limits for each analyte in the sample matrix type were estimated from statistical information derived from standard calibration curves (Taylor 1987). For carp tissue, this limit was typically 5 to 12 ng/g wet wt (ppb) with

M. A. Fisher et al.

18

a mean value of 9 lg/g. Samples were spiked immediately before injection with 10 ll of 100 lg/ml solution 4,4'-dibromo-octafluoro biphenyl as an internal standard.

Statistical Analysis. Statistical analyses were performed using one-way analysis of variance (ANOVA), Student t tests, and simple regression using StatView software (SAS, Cary, NC). Fisher's post hoc test was used after ANOVA to compare differences between treatment groups.

Results Laboratory-Raised Carp PCB Fed Carp Showed No Differences in Size Compared with Control Carp. Juvenile carp exposed to the PCB mixture A1242 through diet did not show a difference in weight, length, or condition factor compared with vehicle-only fed carp during the duration of the experiment (1, 2, 3, and 4 months). Weight (p = 0.0291) and length (p = 0.0034) in A1242-fed and control-fed carp both increased during the experiment.

Muscle PCB Values. Levels of PCBs measured in muscle tissue increased significantly in carp fed PCBs versus carp fed vehicle treated food during the duration of the experiment (ANOVA p < 0.0001; Fig. 2). Muscle PCB concentrations in carp fed PCBs had average values at 1, 2, 3, and 4 months of feeding of 10.0 € 0.5, 13.0 € 0.6, 21.3 € 1.1, and 50.0 € 1.6 lg/g, respectively, and differed significantly by time point (p < 0.0001). There was no difference in muscle PCB values in carp fed control food (with background levels of PCBs only) among the different time points. Muscle PCB concentrations in these carp averaged 3.9 € 0.2 lg wet weight.

Fig. 2. Muscle PCB levels are significantly different between carp fed PCBs (Aroclor 1242) and carp fed vehicle only (ANOVA; p < 0.0001). Muscle PCB levels also differed significantly by time point (p < 0.0001; pairwise comparisons all significant). ANOVA = analysis of variance; PCB = polychlorinated biphenyl

Fig. 3. CYP1A mRNA expression was significantly increased in laboratory carp fed PCBs (Aroclor 1242) compared with the vehiclefed control carp (p = 0.0011). CYP1A mRNA levels were not significantly different among time points (n = 5 for all time points except for 1 month of feeding [n = 4] and after 4 months of feeding a control diet [n = 4; n = 37 total])

CYP1A Real-Time RT-PCR mRNA Values Increased in PCB-Fed Carp. CYP1A mRNA expression levels from carp liver were significantly increased in carp fed A1242 in laboratory experiments compared with control carp fed vehicle (Fig. 3; p = 0.0011). CYP1A mRNA expression in PCB-treated fish was increased 1.4-fold after 1 month and 12.6-fold after 4 months of treatment compared with vehicle-fed control carp. However, differences were not significant among time points. CYP1A mRNA expression levels for untreated carp did not significantly change during the course of the experiment. CYP1A expression levels were positively correlated with muscle PCB levels (p = 0.0006), although the strength was low (r2 = 0.29) in part because of the high variation in the carp fed PCBs for 4 months (16.4 € 7.3).

Kalamazoo River Carp

Real-Time RT-PCR mRNA Values Involved in Oxidative Stress and Metabolism Did Not Change with PCB Ingestion. There was no significant difference in carp liver mRNA in response to PCBs for two genes involved in re-

PCB Values. PCB values were significantly higher in carp liver sampled from PCB sites (15.42 € 1.95 lg/g wet weight) than from carp liver sampled from reference sites (1.19 € lg/g wet weight) (Student t-test p = 0.0001; n = 18;

sponses to toxic oxidation products, Mn-SOD and c-GCS and two genes involved in metabolism, PEP-CK and nucleolin (Table 2).

Size. Carp were sampled from two PCB-contaminated sites within the Superfund site and from two upstream sites (Fig. 1) in the Kalamazoo River during summer 2001. Carp from the reference sites were longer (62.6 € 4.0 cm; n = 6) than the carp from the PCB-contaminated Superfund sites (41.3 € 1.9 cm; n = 11; p < 0.0001).

Induction of CYP1A in PCB Exposed Carp

19

Fig. 4). Furthermore, there was no significant difference in carp liver PCB values between the two reference sites or between the two PCB sites.

CYP1A Real-time RT-PCR mRNA Values Increased in PCB-Contaminated Carp. CYP1A mRNA expression levels from carp liver were significantly increased in carp from PCB-contaminated sites (3.4-fold) compared with upstream reference sites (Student t-test, p 0.0302; Fig. 5). CYP1A expression levels in liver were also positively correlated with liver PCB values (p = 0.0490), although the correlation was weak (r2 = 0.22).

Real-time RT-PCR mRNA Values Involved in Oxidative Stress and Metabolism Did Not Change with PCB Exposure. Liver mRNA expression levels for Mn-SOD, c-GCS, PEP-CK, and nucleolin were not significantly different between PCB-exposed field carp compared with upstream reference carp (Table 3). Two carp from a PCB-contaminated site (Lake Allegan) displayed high Mn-SOD mRNA expression compared with other fish (reflected by a high standard error). One of these carp also displayed a high PEP-CK mRNA level, also reflected in the relatively large SE for that gene.

Fig. 4. PCB values for carp liver sampled from Kalamazoo River reference (n = 6) and PCB-contaminated sites (n = 12) in 2001. Liver PCB values in carp from PCB-contaminated sites were significantly higher than liver values in carp from reference sites (Student t-test p = 0.0001). PCB = polychlorinated biphenyl

Discussion Biomarkers are commonly used as tools to predict exposure and contaminant-induced health effects in organisms and are helpful in understanding integrated biologic responses to environmental contamination. In this study, real-time RT-PCR was used to test for potential biomarkers of exposure in PCBexposed carp, Cyprinus carpio. Carp liver was tested in an acute laboratory feeding study, in which carp were fed PCBs, and in a field study, in which adult carp from the Kalamazoo River Superfund Site were compared with carp from upstream Kalamazoo River reference sites. We demonstrated that CYP1A mRNA is induced in carp liver in the laboratory and in the field. Significantly increased CYP1A mRNA gene expression in the laboratory carp demonstrated that CYP1A mRNA increased in response to the PCB mixture Aroclor 1242. Significantly increased CYP1A mRNA levels in the field suggest that carp are most likely responding to PCB contamination at exposure levels present in the Kalamazoo River. Because exposure to high levels of PCBs adversely affects health (Faroon et al. 2001; Carpenter 2000), it is of concern that PCBs may be present in the Kalamazoo River at levels that may affect resident aquatic organisms. Many previous studies have validated the induction of CYP1A activity for use as a biomarker in fish exposed to Ah agonists such as planar PCBs (i.e., Van der Weiden et al. 1989; Stegeman and Lech 1991; Husoy et al. 1996; Whyte et al. 2000), consistent with the idea that the induction of CYP1A is indicative of environmental contamination and exposure. Induction of CYP1A messenger RNA expression using quantitative PCR technologies has also been used in field studies (Campbell et al. 1996;

Fig. 5. CYP1A mRNA expression was significantly increased in carp sampled at Kalamazoo River PCB sites compared with carp sampled at reference sites (Student t p-test t = 0.0302; n = 18)

Wong et al. 2001; Roy et al. 2002; Rees et al. 2003; McClain et al. 2003). In the present study, real-time RT-PCR was used to examine if PCBs may be present at levels to elicit a biologic response in carp. CYP1A mRNA was induced in liver in laboratory PCB-fed and field-caught carp from the Kalamazoo River Superfund Site and can be viewed as a biomarker of exposure. Significantly increased PCB body burdens in carp caught from the Superfund Site compared with those from upstream sites suggest that PCBs are contributing to the increase of CYP1A mRNA. In the laboratory, carp were dosed with PCBs by feeding with a PCB mixture (Aroclor 1242) estimated to be the most prevalent mixture in the Kalamazoo River (Blasland et al 2000). Experiments using AhR ligands such as PCB congeners or model inducers are commonly used to render information regarding speciesspecific induction capacities or to examine mechanisms of induction. However, in this initial study, exposing carp to an environmentally relevant mixture through diet was more pertinent to our objective of finding genes whose expression may also differ in a contaminated field setting. Future induction studies will be needed to further define biologic responses to specific PCB congeners in carp. Metabolic and oxidative stress genes, also predicted to be regulated by PCB exposure, were not differentially expressed in carp. The lack of PEP-CK mRNA regulation suggests that

M. A. Fisher et al.

20

Table 2. Real-time RT-PCR liver mRNA values for genes involved in oxidative stress (Mn-SOD and c-GCS) and metabolism (PEP-CK and Nucleolin)a Gene Mn-SOD

c-GCS

PEP-CK

Nucleolin

Months of exposure

Controlab

PCBab

1c 2 3 4 1c 2 3 4 1c 2 3 4 1c 2 3 4

0.85 1.15 0.85 0.75 1.09 2.22 2.21 0.59 0.96 2.71 1.66 1.39 0.79 1.94 2.06 1.71

0.77 1.3 1.09 1.38 1.50 2.26 2.38 1.27 0.55 3.12 1.98 2.34 1.48 1.54 1.20 3.26

€ € € € € € € € € € € € € € € €

0.29 0.54 0.12 0.22c 0.38 0.7 1.22 0.21c 0.27 1.04 0.26 0.3d 0.24 0.78 1.05 0.46d

€ € € € € € € € € € € € € € € €

0.24 0.53 0.14 0.51 0.75 0.69 0.59 0.57 0.2 0.84 0.21 1.05 0.72 0.45 0.27 1.42

a

Carp were raised in the laboratory and exposed to PCBs through the diet for up to 4 months. N = 5 carp treatment except where stated. Differences between control and PCB-exposed fish were not statistically significant. b Values are in relative units. c n = 4. d n = 3. CYP1A-Cytochrome P4501A. GCS - Gamma-glutamylcysteine synthetase. Mn-SOD-Magnesium superoxide dismutase. PEP-CK-Phosphoenolpyruvate carboxykinase. RT-PCR-Reverse transcriptase–polymerase chain reaction.

glycogen mobilization may not be affected by PCB exposure levels encountered in this study or is not affected in carp compared with mammals (Weber et al. 1991; Stahl et al. 1993). The lack of nucleolin regulation suggests that protein synthesis may not be compromised by PCB exposure levels encountered in this study. Both c-GCS and Mn-SOD liver mRNA expression levels were also not significantly altered in carp exposed to PCBs, although significant evidence suggests that oxidative stress increases in response to Ah receptor ligands (Amaro et al. 1996; Ferriera et al. 2005; Nebert et al. 2000; Orbea et al. 2002; Sneft et al. 2002; Stohs 1990; Van der Oost et al. 2003). In previous studies using carp, glutathione content and activity (using 1-chloro-2,4-dinitrobenzene; CDNB) and cytochrome P-450 content were upregulated after 2 weeks in carp fed 100 lg/g (ppm) Kasei PCB-48 (equivalent to Aroclor 1248) (Kobayashi et al. 1987). Also, SOD, glutathione S-transferase, uridine phospho-glucuronosyltransferase, and CYP enzyme activities were increased in caged carp in a river heavily polluted with PCBs and organochlorine pesticides (Van der Oost et al. 1998). It has also been shown that c-GCS mRNA, or glutamate-cysteine ligase catalytic subunit mRNA, is upregulated along with glutathione content and ethoxyresorufin O-deethylase activities, although CDNB activity was unchanged in largemouth bass (Micropterus salmoides) exposed to b-napthoflavone, an Ah ligand (Hughes and Gallagher 2004). Interestingly, in a Kalamazoo River study, liver and spleen SOD activity were decreased and CYP1A protein

content did not differ in smallmouth bass (Micropterus dolomieu) inhabiting a PCB-contaminated site compared with reference fish (Anderson et al. 2003). The reason why CYP1A mRNA activity was increased but c-GCS and Mn-SOD oxidative stress genes were not affected in carp examined in this study remains an open question. CYP1A mRNA expression is inducible, mediated through activation of the Ah receptor. The CYP1A protein can metabolize Ah- ligands, but it may also be a source of reactive oxygen species and activated metabolic breakdown products (Schlezinger et al. 1999). Superoxide anion and hydrogen peroxide may also be produced by CYP1A by uncoupling during CYP1A electron transfer (Schlezinger et al. 1999; Park et al. 1996; Alsharif et al. 1994). Several mechanisms that mediate oxidative stress were not examined in the present study. For example, it may be possible that glutathione may be used to mediate oxidative damage but regenerated by using means others than increased synthesis, i.e., increased glutathione reductase activity to convert oxidized glutathione to glutathione, increased availability of nicotinamide adenine dinucleotide phosphate energy equivalents, or cellular export of oxidized glutathione (Schafer and Buettner 2001). Phase II enzyme activities—such as glutathione S-transferases, UDP-glucuronosyltransferases, and DT diaphorase (NAD(P)H:quinone oxidoreductase)—function to mediate oxidative stress and have been shown to be regulated in response to PCBs in fish (Machala et al. 1998; Monody et al. 1988; Otto and Moon 1995, 1996; Perez-Lopez et al. 2002; Sturve et al. 2005), but they were not examined. It is therefore possible that carp use alternative mechanisms to limit the damaging effects of CYP1A metabolites and oxidative stress untested in this study. It is also possible that PCB levels present in the laboratory and field carp were not sufficient to significantly induce or sustain an oxidative stress response. In the laboratory dosing experiment, the most contaminated carp had an average of 50.0 € 1.6 lg/g wet wt muscle PCBs after 4 months of feeding, and carp from PCB sites in the Kalamazoo River had an average of 15.42 € 1.95 lg/g wet wt liver PCBs. After an extensive literature review, Niimi (1996) concluded that in PCB studies in aquatic organisms, tissue concentrations at approximately 50 to 100 lg/g wet weight may be needed to adversely affect growth and reproduction. Thus, body burdens £ 50 ppm in carp from the present study might be sufficient to activate enzymatic processes (such as CYP1A), yet not of sufficient magnitude to adversely affect major physiologic processes through generation of reactive oxygen species or other damaging metabolites.

Conclusion This study used real-time RT-PCR to examine expression levels in liver of select genes to address the question if carp may be responding to environmental contaminants in a PCB Superfund site. The induction of CYP1A mRNA in laboratory carp exposed to PCBs and in carp from PCB-contaminated Superfund site indicate that carp respond to environmental levels of PCBs as expected (Van der Oost et al. 1998; Payne et al. 1987; Haasch et al. 1993). However, the lack of response

21

Induction of CYP1A in PCB Exposed Carp

Table 3. Real-time RT-PCR liver mRNA values for genes involved in oxidative stress (Mn-SOD and c-GCS) and metabolism (PEP-CK and nucleolin)a Gene

Referenceab

PCBab

Mn-SOD c-GCS PEP-CK Nucleolin

0.72 0.49 1.98 2.47

1.76 0.79 3.26 1.80

€ € € €

0.21 0.15 0.64 0.63

€ € € €

5.96 0.22 1.43 0.42

a

Carp were caught from upstream reference sites or PCB-contaminated sites from the Kalamazoo River (n = 6 reference sites and n = 12 PCB sites). Differences between control and PCB-exposed fish were not statistically significant. b Values are in relative units. GCS - Gamma-glutamylcysteine synthetase. Mn-SOD-Magnesium superoxide dismutase. PEP-CK-Phosphoenolpyruvate Carboxykinase. RT-PCR-Reverse transcriptase–polymerase chain reaction.

of select oxidative stress genes tested suggests that carp may not be exposed to PCBs at a level sufficient to induce oxidative stress or that carp mediate stress by using alternative mechanisms.

Acknowledgments. This study was funded by USEPA Grant No. XP7528301-0; the College of Arts and Sciences at Western Michigan University Monroe-Brown Life Science Research Award to M. A. F. and a West Michigan Chapter of the Air and Waste Management Association research scholarship to M. A. F. The authors thank Brent Lehmkuhl for help catching carp and anonymous reviewers for their suggestions.

References Alsharif N, Lawson T, Stohs S (1994) Oxidative stress induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin is mediated by the aryl hydrocarbon (Ah) receptor complex. Toxicol Appl Pharmacol 92:39–51 Alvarez M, Quezada C, Navarro C, Molina A, Bouvet P, Krauskopf M et al. (2003) An increased expression of nucleolin is associated with a physiological nucleolar segregation. Biochem Biophys Res Commun 301:152–158 Amaro A, Oakley G, Bauer U, Spielmann H, Robertson L (1996) Metabolic activation of PCBs to quinones: Reactivity toward nitrogen and sulfur nucleophiles and influence of superoxide dismutase. Chem Res Toxieol 9:623–629 Anderson M, Cacela D, Beltman D, Teh S, Okihiro M, Hinton D et al. (2003) Biochemical and toxicopathic biomarkers assessed in smallmouth bass recovered from a polychlorinated biphenylcontaminated river. Biomarkers 8:371–393 Blasland, Bouck, and Lee (2000) Allied Paper, Inc./Portage Creek/ Kalamazoo River Superfund Site RI/FS. Remedial Investigation Report Phase 1. Blasland, Bouck, and Lee, Syracuse, NY Campbell PM, Kruzynski GM, Birtwell IK, Devlin RH (1996) Quantitation of dose-dependent increases in CYP1A messenger RNA levels in juvenile Chinook salmon exposed to treated bleached-kraft mill effluent using two field sampling techniques. Environ Toxicol Chem 15:1119–1123 Carpenter DO (1998) Polychlorinated biphenyls and human health. Int J Occup Med Environ Health 11:291–303

Daniely Y, Dimitrova DD, Borowiec JA (2002) Stress-dependent nucleolin mobilization mediated by p53-nucleolin complex formation. Mol Cell Biol 22:6014–6022 Faroon O, Keith S, Jones D, de Rosa C (2001) Effects of polychlorinated biphenyls on development and reproduction. Toxicol Ind Health 17:63–93 Ferreira M, Moradas-Ferreira P, Reis-Henriques MA (2005) Oxidative stress biomarkers in two resident species, mullet (Mugil cephalus) and flounder (Platichthys flesus), from a polluted site in River Douro Estuary, Portugal. Aquat Toxicol 71:39–48 Fisher MA, Jelaso AM, Predenkiewicz A, Schuster L, Means J, Ide CF (2003) Exposure to the polychlorinated biphenyl mixture Aroclor 1254 alters melanocyte and tail muscle morphology in developing Xenopus laevis tadpoles. Environ Toxicol Chem 22:321–328 Goldstein JA, Safe S (1989) Mechanism of action and structureactivity relationship for the chlorinated dibenzo-p-dioxins and related compounds. In: Kimbrough RD, Jenson AA, (eds) Halogenated biphenyls, terphenyls, naphthalenes, dibenzodioxins and related products. Elsevier, Amsterdam, p.239–293 Haasch ML, Quardokus EM, Sutherland LA, Goodrich M, Lech JJ (1993) Hepatic CYP1A1 induction in rainbow trout by continuous flowthrough exposure to beta-naphthoflavone. Fundam Appl Toxicol 20:72–82 Hughes EM, Gallagher EP (2004) Effects of [beta]-naphthoflavone on hepatic biotransformation and glutathione biosynthesis in largemouth bass (Micropterus salmoides). Mar Environmental Res 58:675–679 Husoy A, Myers MS, Goksoyr A (1996) Cellular localization of cytochrome P450 (CYP1A) induction and histology in Atlantic cod (Gadus morhua L.) and European flounder (Platichthys flesus) after environmental exposure to contaminants by caging in Sorfjorden, Norway. Aquat Toxicol 36:53–74 Jelaso AM, Lehigh-Shirey E, Means J, Ide CF (2003) Gene expression patterns predict exposure to PCBs in developing Xenopus laevis tadpoles. Environ Mol Mutagen 42:1–10 Kim S, Kim T (2003) Selection of optimal internal controls for gene expression profiling of liver disease. BioTechniques 35:456–460 Kobayashi K, Oshima Y, Taguchi C, Wang Y (1987) Induction of drug-metabolizing enzymes by long-term administration of PCB and duration of their induced activities in carp. Nippon Suisan Gakkaishi 55:487–491 Livingstone DR, Mitchelmore CL, O'Hara SC, Lemaire P, Sturve J, Forlin L (2000) Increased potential for NAD(P)H-dependent reactive oxygen species production of hepatic subcellular fractions of fish species with in vivo exposure to contaminants. Mar Environ Res 51:57–60 Machala M, Drabek P, Neca J, Kolarova J, Svobodova Z (1998) Biochemical markers for differentiation of exposures to nonplanar polychlorinated biphenyls, organochlorine pesticides, or 2,3,7,8tetrachlorodibenzo-p-dioxin in trout liver. Ecotoxicol Environ Safe 41:107–111 McClain JS, Oris JT, Burton Jr, GA, Lattier D (2003) Laboratory and field validation of multiple molecular biomarkers of contaminant exposure in rainbow trout (Oncorhynchus mykiss). Environ Toxicol Chem 22:361–370 McMillin D, Means J (1996) Spatial and temporal trends of pesticides residues in water and particulates in the Mississippi River plume and the Northwestern Gulf of Mexico. J Chromatogr 754:169–185 Means J (1998) Compound-specific gas chromatographic/mass spectrometric analysis of alkylated and parent polycyclic aromatic hydrocarbons in water, sediments and aquatic organisms. J Assoc Off Anal Chem 81:657–672 Melancon MJ, Lech JJ (1983) Dose-effect relationship for induction of hepatic monooxygenase activity in rainbow trout and carp by Aroclor 1254. Aquat Toxicol 4:51–61 Meyer JN, Smith JD, Winston GW, Di Giulio RT (2003) Antioxidant defenses in killifish (Fundulus heteroclitus) exposed to contam-

22

inated sediments and model prooxidants: Short-term and heritable responses. Aquat Toxicol 65:377–395 Monod G, Devaux A, Riviere J (1988) Effects of chemical pollution on the activities of hepatic xenobiotic metabolizing enzymes in fish from the River Rhone. Sci Total Environ 73:189–201 Nebert DW, Roe AL, Dieter MZ, Solis WA, Yang Y, Dalton TP (2000) Role of the aromatic hydrocarbon receptor and [Ah] gene battery in the oxidative stress response, cell cycle control, and apoptosis. Biochem Pharmacol 59:65–85 Niimi AJ (1996) PCBs in aquatic organisms. In: Beyer WN, Heins GH, Redmon-Norwood AW, (eds) Environmental Contaminations in Wildlife. Interpreting Tissue Concentrations. CRC Press, Boca Raton, FL, USA, pp 117–152 Orbea A, Ortiz-Zarragoitia M, Sole M, Porte C, Cajaraville MP (2002) Antioxidant enzymes and peroxisome proliferation in relation to contaminant body burdens of PAHs and PCBs in bivalve molluscs, crabs and fish from the Urdaibai and Plentzia estuaries (Bay of Biscay). Aquat Toxicol 58:75–98 Otto DM, Moon TW (1996) Phase I and II enzymes and antioxidant responses in different tissues of brown bullheads from relatively polluted and non-polluted systems. Arch Environ Contam Toxicol 31:141–147 Otto DM, Moon TW (1995) 3,3',4,4'-tetrachlorobiphenyl effects on antioxidant enzymes and glutathione status in different tissues of rainbow trout. Pharmacol Toxicol 77:281–287 Park JK, Shigenaga MK, Ames BN (1996) Induction of cytochrome P4501A1 by 2,3,7,8-tetrachlorodibenzo-p-dioxin or indolo(3,2-b) carbazole is associated with oxidative DNA damage. Proc Natl Acad Sci U S A 93:2322–2327 Payne JF, Fancy LL, Rahimtula AD, Porter EL (1987) Review and perspective on the use of mixed-function oxygense enzymes in biological monitoring. Comp Biochem Physiol C86:233–245 Peakall D (1992) Animal biomarkers as pollution indicators. Chapman & Hall, London, UK Perez-Lopez M, Novoa-Valinas MC, Melgar-Riol MJ (2002) Glutathione S-transferase cytosolic isoforms as biomarkets of polychlorinate biphenyl (Arochlor-1254) experimental contamination in rainbow trout. Toxicol Lett 136:97–406 Rees C, Li W (2004) Development and application of a real-time quantitative PCR assay for determining CYP1A transcripts in three genera of salmonids. Aquat Toxicol 66:357–368 Roberts MH, Sved DW, Felton SP (1987) Temporal changes in AHH and SOD activities in feral spot from the Elizabeth River, a polluted sub-estuary. Mar Environ Res 23:89–101 Roy N, Courtenay S, Maxwell G, Yuan Z, Chambers R, Wirgin I (2002) Cytochrome P4501A1 is induced by PCB 77 and benzo[a]pyrene treatment but not by exposure to the Hudson River environment in Atlantic tomcod (Microgadus tomcod) postyolk sac larvae. Biomarkers 7:162–173 Schafer F, Buettner G (2001) Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutatliione couple. Free Radic Biol Med 30:1191–1212 Schlezinger JJ, White RD, Stegeman JJ (1999) Oxidative inactivation of cytochrome P-450 1A (CYPIA) stimulated by 3,3¢,4,4¢-tetra-

M. A. Fisher et al.

chlorobiphenyl: production of reactive oxygen by vertebrate CYP1As. Mol Pharmacol 56:588–597 Senft A, Dalton T, Nebert D, Center M, Puga A, Hutchinson R et al. (2002) Mitochondrial reactive oxygen production is dependent on the aromatic hydrocarbon receptor. Free Radic Biol Med 33:1268–1278 Sivarajah K, Franklin CS, Williams WP (1978) Some histopathological effects of Aroclor 1254 on the liver and gonads of rainbow trout, Salmo gairdneri and carp, Cyprinus carpio. J Fish Biol 13:411–414 Stahl BU, Beer DO, Weber LWD, Rozman K (1993) Reduction of hepatic phosphoenolpyruvate carboxykinase (PEPCK) activity by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is due to decreased mRNA levels. Toxicology 79:81–95 Stegeman J, Lech J (1991) Cytochrome P-450 monooxygenase systems in aquatic species: Carcinogen metabolism and biomarkers for carcinogen and pollutant exposure. Environ Health Perspect 90:101–109 Stohs S (1990) Oxidative stress induced by 2,3,7,8-tetraclilorodibenzo-p-dioxin (TCDD). Free Radic Biol Med 9:79–90 Sturve J, Stephensen E, Forlin L (2005) Effects of redox cycling compounds on DT diaphorase activity in the liver of rainbow trout (Oncorhynchus mykiss). Comp Hepatol 4:4 Taylor J (1987) Quality assurance of chemical measurement. Lewis, Chelsea, MI Thomas P, Wofford HW (1984) Effects of metals and organic compounds on hepatic glutathione, cystein, and acid soluble thiol levels in mullet (Mugil cephalus L.). Toxicol Appl Pharmacol 76:172–182 Van der Oost R, Beyer J, Vermeulen NPE (2003) Fish bioaccumulation and biomarkers in environmental risk assessment: A review. Environ Toxicol Pharmacol 13:57–149 Van der Oost R, Lopes SCC, Komen H, Satumalay K, Van den Bos V, Heidet H et al. (1998) Assessment of environmental quality and inland water pollution using biomarker responses in caged carp (Cyprinus carpio): Use of a bioconcentration:detoxication ratio as a biotransformation index. Mar Environ Res 46:315– 319 Van der Weiden MEJ, Craane LHJ, Evers EHG, Kooke RMM, Olie K, Seinen W et al. (1989) Bioavailability of PCDDs and PCDFs from bottom sediments and some associated biological effects in the carp (Cyprinus carpio). Chemosphere 19:1009–1016 Weber LW, Lebofsky M, Stahl BU, Gorski JR, Muzi G, Rozman K (1991) Reduced activities of key enzymes of gluconeogenesis as possible cause of acute toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in rats. Toxicology 66(2): 133–144 Whyte JJ, Jung RE, Schmitt CJ, Tillitt DE (2000) Ethoxyresorufin-Odeethylase (EROD) activity in fish as a biomarker of chemical exposure. Crit Rev Toxicol 30:347–570 Wong CKC, Yeung HY, Woo PS, Wong MH (2001) Specific expression of cytochrome P4501A1 gene in gill, intestine and liver of tilapia exposed to coastal sediments. Aquat Toxicol 54:69–80