Gene expression, glutathione status, and indicators of hepatic oxidative stress in laughing gull (Larus atricilla) hatchlings exposed to methylmercury (original) (raw)

Abstract

Despite extensive studies of methylmercury (MeHg) toxicity in birds, molecular effects on birds are poorly characterized. To improve our understanding of toxicity pathways and identify novel indicators of avian exposure to Hg, the authors investigated genomic changes, glutathione status, and oxidative status indicators in liver from laughing gull (Larus atricilla) hatchlings that were exposed in ovo to MeHg (0.05–1.6 µg/g). Genes involved in the transsulfuration pathway, iron transport and storage, thyroid‐hormone related processes, and cellular respiration were identified by suppression subtractive hybridization as differentially expressed. Quantitative polymerase chain reaction (qPCR) identified statistically significant effects of Hg on cytochrome C oxidase subunits I and II, transferrin, and methionine adenosyltransferase RNA expression. Glutathione‐_S_‐transferase activity and protein‐bound sulfhydryl levels decreased, whereas glucose‐6‐phosphate dehydrogenase activity increased dose‐dependently. Total sulfhydryl concentrations were significantly lower at 0.4 µg/g Hg than in controls. Together, these endpoints provided some evidence of compensatory effects, but little indication of oxidative damage at the tested doses, and suggest that sequestration of Hg through various pathways may be important for minimizing toxicity in laughing gulls. This is the first study to describe the genomic response of an avian species to Hg. Laughing gulls are among the less sensitive avian species with regard to Hg toxicity, and their ability to prevent hepatic oxidative stress may be important for surviving levels of MeHg exposures at which other species succumb. Environ. Toxicol. Chem. 2012; 31: 2588–2596. © 2012 SETAC

INTRODUCTION

Methylmercury (MeHg), the most toxic form of Hg, readily accumulates up the food chain, posing a threat to both terrestrial and aquatic wildlife. Birds are especially vulnerable to the effects of Hg because of their trophic position and longevity, and tissue accumulation of Hg has been documented across avian species, in piscivores as well as terrestrial insectivores 1. Physiological effects of Hg are well documented in birds and include central nervous system disruptions, brain lesions, and kidney and liver damage 2. Female birds are known to deposit large fractions of their MeHg body burden into their eggs, resulting in reduced hatchability, embryo death, and developmental abnormalities in chicks 3, 4. Mallard hatchlings exposed in ovo to MeHg later exhibited a variety of reproductive effects as adults 5. In general, embryos and young birds are the most sensitive avian life stage.

Organisms have a variety of mechanisms for coping with MeHg, including active efflux, sequestration, and demethylation 6, 7. Previous studies have suggested that MeHg tolerance depends on the cellular levels of protective pathways, and damage occurs once the capacity of these protective mechanisms is exceeded 8. The mechanisms influencing an organism's sensitivity originate at the molecular level, where certain genetic polymorphisms can increase or decrease susceptibility to MeHg. For example, in rats and humans, certain polymorphisms of glutathione (GSH), an antioxidant that protects cells from reactive oxygen species (ROS), including those generated by xenobiotics, result in greater or lesser sensitivity to the effects of MeHg 9, 10. This suggests that differences in sensitivity to Hg or other xenobiotics between individuals or species are directly related to variations in genetic sequences and expression. Alterations in the expression of multiple genes, including those associated with the endocrine, immune, and nervous systems, have been reported in mammals, fish, and amphibians exposed to Hg 9, 11, 12. The expression and genetics of various oxidative stress–related genes have also been studied in mammals exposed to MeHg 9, 10. However, information on avian genetic responses to MeHg is limited.

In the present study, we investigated genomic changes in livers from laughing gull (Larus atricilla) hatchlings exposed to graded doses of MeHg in ovo and compared expression patterns across dosages. Laughing gulls are largely a marine species, with a diet that consists primarily of fish but includes anthropogenically derived food items such as meat, corn, and peanuts (L.M. Showalter, 2010, Master's thesis, University of Alabama, Tuscaloosa, AL, USA). A related study, from which the presently analyzed samples were collected, found that laughing gull embryos were among the least sensitive bird species exposed to MeHg, as indicated by a median lethal concentration (LC50) of 1.25 µg/g for MeHg injected into fertile eggs 13. We elected to study laughing gulls based on these findings, to gain insight into the molecular pathways that respond to MeHg exposure in a species that can survive at relatively higher Hg levels.

As the primary site for MeHg demethylation, the liver plays a critical role in Hg detoxification in birds and is vital for reducing the toxicity of MeHg to the avian neurological system 6, 14. In seabirds, MeHg accumulates at higher levels in the liver than in other organs 14. Despite its importance in avian responses to MeHg toxicity, genomic data on avian liver have not been previously reported. However, a study of zebrafish comparing the genetic effects of dietary MeHg in liver, skeletal muscle, and brain showed that although the brain accumulated the highest levels of Hg, expression of genes involved in antioxidant defenses, active efflux of organic compounds, and DNA repair did not vary between controls and dosed animals 12. Liver, however, exhibited significant changes in gene expression in response to MeHg, and, as reported in birds, was also found to be an important organ for demethylation. To better characterize the response of laughing gulls to MeHg, we also analyzed liver samples for a number of biochemical bioindicators of GSH and oxidative status. Oxidative stress is one of the most extensively studied mechanisms of MeHg toxicity, and the formation of highly reactive free radicals, mostly ROS, is induced by MeHg exposure 15. Reduced GSH and associated antioxidant enzymes are major combatants of oxidative stress and are commonly used as indicators of sublethal toxicity in birds 16. Identification of differentially expressed genes and analysis of their expression patterns and relationship with biochemical indicators under varying levels of Hg provides a foundation for understanding the molecular mechanisms involved in the response and tolerance of laughing gulls to MeHg stress. To our knowledge, this is the first study to assess the effects of MeHg on messenger RNA (mRNA) expression in a wild bird species.

MATERIALS AND METHODS

Exposures and sample collection

The liver samples employed in this study were collected from laughing gull hatchlings that had been exposed in ovo to MeHg in a study conducted by Heinz et al. 13. The eggs of several wild bird species were collected from the wild and injected with graded doses of MeHg according to an optimized egg injection protocol 17 under which embryo sensitivity is comparable to that observed when MeHg is naturally deposited by the mother. As described by Heinz et al. 13, laughing gull eggs were collected from the field in areas known to have low Hg contamination. The eggs were randomized and separated into nine treatment groups, each containing 20 to 30 eggs. When candling demonstrated that the embryos had reached the appearance of a 3‐d‐old chicken embryo, the eggs in seven groups were injected into the air cell (at 1 µl/g egg) with a single dose of Hg as MeHg chloride dissolved in corn oil that would result in 0.05, 0.1, 0.2, 0.4, 0.8, 1.6, and 3.2 µg/g Hg on a wet weight basis in the egg. Two types of controls were established. One group received a single dose of corn oil into the air cell (at 1 µl/g egg), whereas the eggs in the second control group were not injected with the corn oil vehicle. Further details of the egg injection protocol, exposures, and incubation can be found in Heinz et al. 13, 17. Eggs were allowed to hatch, and within 24 hours liver tissue was collected from five hatchlings at each dosage level, except at 1.6 µg/g Hg, where only three eggs hatched. Only one egg hatched at 3.2 µg/g, and thus this dosage was dropped from the analysis. Because only a small difference was observed in survival between the controls that received corn oil (63%) and those that did not (66%), only those samples that received corn oil were analyzed. The liver samples were preserved in RNA Later (Ambion) and stored at −80°C. Treatment of all animals was performed according to the policies and with the approval of the Institutional Animal Care and Use Committee of the USGS Patuxent Wildlife Research Center.

Total RNA extraction and polyadenylated RNA isolation

Total RNA was extracted from approximately 50 mg laughing gull livers. Livers were homogenized in 1 ml TRI Reagent RT (Molecular Research Center) with three 2.0‐mm zirconium oxide beads using a Bullet Blender (Next Advance) for 2 min at a speed of 8. The homogenized mixture was transferred to a new tube to which 4 µl PolyAcryl Carrier (Molecular Research Center) was added to improve isolation. Membranes, polysaccharides, and high‐molecular‐weight DNA were removed by a 5‐min incubation period and 10‐min centrifugation at 4°C (12,000 g). The remainder of the total RNA extraction procedure followed the manufacturer's instructions. Total RNA concentration and A260/A280 were determined by NanoDrop spectrophotometry (Thermo Scientific), and RNA quality was assessed visually by gel electrophoresis using the NorthernMax‐Gly Sample Loading Dye protocol (Applied Biosystems/Ambion). Three RNA samples (one control, two 0.8 µg/g) appeared degraded and were excluded from subsequent analyses.

Polyadenylated RNA (poly A+ mRNA) was isolated from total RNA using the Oligotex mRNA Mini Kit (Qiagen). Poly A+ mRNA was concentrated by adding 2 µl linear acrylamide, 1:10 3M sodium acetate, and 4 volumes of 100% ethanol, followed by overnight incubation at −20°C. The precipitate was centrifuged for 30 min at 4°C, and then the supernatant was removed from the pellet. The mRNA pellet was washed twice with 70% ethanol and then resuspended in the appropriate volume of RNA storage solution (1 mM sodium citrate, pH 6.4 ± 0.2) (Applied Biosystems/Ambion).

Suppression subtractive hybridization

Differentially expressed genes were identified by suppression subtractive hybridization (SSH) using the Clontech PCR‐Select™ cDNA Subtraction Kit (Clontech Laboratories). In one library, Poly A+ mRNA from a laughing gull exposed to 0.4 µg/g Hg was used as the tester population, and an undosed control sample was used as the driver in the forward subtraction (up‐regulated genes). In a second library, the experimentally dosed mRNA served as the driver and the undosed control sample served as the tester in the reverse subtraction, which produced differential cDNA sequences specific to the undosed sample (down‐regulated). We selected 0.4 µg/g Hg because at this dose survival of laughing gull embryos through 90% of incubation (48.5%) was substantially lower than in controls 13.

We made the following modifications to the manufacturer's SSH protocol: The Transcriptor First Strand cDNA Sythesis Kit (Roche) was used instead of the avian myeloblastosis virus reverse transcriptase (AMV‐RT) provided in the Clontech kit. The _Rsa_I digestion step used enzyme, 10× restriction buffer, and 0.1 µg/µl acetylated bovine serum albumin from Promega, with incubation at 37°C for 4 hours. For verification of adaptor ligation, we used avian‐specific genetic sexing primers (2718R/2550F) 18. FastStart Taq DNA Polymerase (Roche) was used instead of the cDNA polymerase mix from the Clontech kit.

Primary PCR consisted of sterile water, 10× PCR reaction buffer (Roche) (50 mM Tris‐HCl, 10 mM KCl, 5 mM [NH4]2SO4, 2 mM MgCl2), 200 µM deoxynucleotide triphosphate mix, 0.4 µM PCR primer 1 (Clontech), and 1 U FastStart Taq polymerase (Roche), for a total volume of 25 µl. Polymerase chain reaction cycling was performed in a C1000 Thermal Cycler (Biorad) and consisted of an initial incubation step of 75°C for 5 min to extend the adaptors, followed by heating to 95°C for 6 min. Samples were subjected to 30 cycles of 95°C for 30 s, 66°C for 30 s, and 72°C for 90 s, followed by a final extension at 72°C for 7 min. Two microliters of a 1:10 dilution of the primary PCR product were used in the secondary PCR reaction, which contained a mixture of sterile water, 10× PCR reaction buffer (Roche), 200 µM deoxynucleotide triphosphate mix, 0.4 µM each of nested primer 1 and nested primer 2R (Clontech), and 1 U of FastStart Taq Polymerase (Roche), for a total volume of 25 µl. Amplification was performed with an initial step of 95°C for 6 min, followed by 15 cycles of 95°C for 30 s, 68°C for 30 s, and 72°C for 90 s, with a final extension step of 72°C for 7 min. Reaction products were analyzed on a 1.5% sodium borate agarose gel to verify success of subtractions and efficiency of PCRs.

Cloning of secondary PCR products

Secondary PCR products (forward‐subtracted, reverse‐subtracted, and their respective unsubtracted cDNAs) were purified using the Wizard SV Gel and PCR Clean‐Up System (Promega) and cloned using the StrataClone PCR Cloning Kit following the manufacturer's instructions (Stratagene) to create forward‐subtracted and reverse‐subtracted cDNA libraries. Differentially expressed sequences from positive, white colonies of the forward‐ and reverse‐subtracted cDNA libraries were PCR amplified under the following conditions: sterile water, 10× PCR reaction buffer (Roche), 200 µM deoxynucleotide triphosphate mix, 0.2 µM each of M13 forward primer and M13 reverse primer, 1 µl of colony DNA, and 1 U of FastStart Taq Polymerase (Roche), for a total volume of 25 µl. Polymerase chain reaction was initiated at 95°C for 6 min, followed by 30 cycles of 94°C for 30 s, 56°C for 30 s, and at 72°C for 90 s, with a final extension step of 72°C for 10 min. Amplified cDNA inserts were analyzed on a 1.5% sodium borate gel to check for fragment size. This procedure was repeated separately for the forward‐subtracted and the reverse‐subtracted cDNA libraries so that two cDNA libraries each consisting of 96 colonies were created.

Differential screening

The cDNA clones were differentially screened using the method described in the PCR‐Select Differential Screening Kit User Manual (Clontech), with slight modification. To minimize background, the adaptors on the subtracted and unsubtracted samples to be used as probes were removed by _Rsa_I, _Eag_I, and _Sma_I restriction enzyme digestion following the PCR‐Select Differential Screening Kit protocol. For hybridization, we used the digoxigenin (DIG) High Prime DNA Labeling and Detection Starter Kit II (Roche). Membranes were incubated for 30 min at 42°C in a blocking buffer containing 50 µl 20× saline‐sodium citrate, 50 µl denatured blocking solution (Clontech), and 4 ml DIG Easy Hyb buffer (Roche). A 100‐ng aliquot of each forward‐subtracted, reverse‐subtracted, forward‐unsubtracted, and reverse‐unsubtracted sample was labeled with DIG following the manufacturer's protocol. The labeled probes were combined with 40 µl DIG Easy Hyb (Roche), 50 µl 20× saline‐sodium citrate, and 50 µl Clontech Blocking Solution and denatured at 95°C for 5 min. The reaction was then chilled on ice and added to 35 ml preheated DIG Easy Hyb buffer to a final concentration of 40 ng/ml. Replicate membranes were hybridized at 42°C overnight with 3.5 ml of one of the following probes: forward‐subtracted cDNA probe, reverse‐subtracted cDNA probe, unsubtracted tester probe, or unsubtracted driver probe.

Washing and immunological detection followed the instructions for the DIG High Prime DNA Labeling and Detection Starter Kit II (Roche) under stringent conditions. The ECF Substrate (GE Healthcare) was used for chemifluorescent detection of the alkaline phosphatase–conjugated probes instead of the disodium 3‐(4‐methoxyspiro {1,2‐dioxetane‐3,2′‐(5′‐chloro)tricyclo [3.3.1.13,7]decan}‐4‐yl)phenyl phosphate solution included in the Roche Kit. Each membrane was incubated in approximately 1.5 ml ECF Substrate for 1 minute and transferred to a plastic envelope. The envelopes were placed in aluminum foil in the dark for approximately 40 min until chemifluorescent analysis was performed on a Typhoon 9400 scanner (GE Healthcare) at a wavelength of 477 nm.

Sequencing

Plasmids containing differentially expressed genes were purified using the Bio101 Systems RPM Turbo Kit (QBioGene) according to the manufacturer's instructions. Plasmid concentrations were determined by NanoDrop spectroscopy (Thermo Scientific), and a 96‐well plate was prepared with the appropriate concentrations of each sample required for sequencing. Samples were sequenced by Michigan State University, Research Technology Support Facility, on an ABI 3730 Genetic Analyzer (Applied Biosystems). Sequence analysis was performed using the Basic Local Alignment Search Tool (BLAST) 19. Sequences were deposited in Genbank (Table 1). Based on a preliminary literature search of gene function, four down‐regulated and four up‐regulated sequences were chosen for further analysis.

Table 1.

Differentially expressed genes in laughing gulls exposed to MeHg as identified by SSHa

a

Subtractive libraries were constructed using a control hatchling and a hatchling exposed to 0.4 mg/kg MeHg. Accession IDs for sequences deposited in Genbank are provided.

MeHg = methylmercury; SSH = suppression subtractive hybridization; ROS = reactive oxygen species.

Table 1.

Differentially expressed genes in laughing gulls exposed to MeHg as identified by SSHa

a

Subtractive libraries were constructed using a control hatchling and a hatchling exposed to 0.4 mg/kg MeHg. Accession IDs for sequences deposited in Genbank are provided.

MeHg = methylmercury; SSH = suppression subtractive hybridization; ROS = reactive oxygen species.

Real time quantitative PCR

Forward and reverse primers were designed to amplify short segments (∼100 bp) within the seven differentially expressed sequences using Geneious Pro 4.8.3 (http://www.geneious.com). Exonic primers were also designed for four candidate reference genes (glyceraldehyde‐3‐phosphate dehydrogenase, β‐actin, and 28S and 18S ribosomal RNA) using conserved sequence regions from several avian species. Assay optimization/validation included primer specificity analysis using BLAST, gel electrophoresis, and melt curve analysis, annealing temperature and primer concentration optimization, and analysis of cDNA serial dilutions for each primer pair to calculate the reaction efficiency. Primer‐specific annealing temperatures, concentrations, abbreviations, sequences, efficiencies, and product lengths are provided in the Supplemental Data, Table S1.

The Applied Biosystems TURBO DNA‐free Kit was used to remove contaminating DNA from 10 µl of each sample. Samples were then reverse transcribed into cDNA using the SuperScript III First‐Strand Synthesis SuperMix system (Invitrogen). First‐strand cDNA was synthesized using 1 µg total RNA, 5 µM oligo dT, and 5 ng/µl random hexamers according to the manufacturer's protocol. The reaction was incubated for 10 min at 25°C, followed by 50 min at 50°C and terminated by heating to 85°C for 5 min and then chilled on ice. A 1‐µl aliquot of RNase H was added to each tube and incubated for 20 min at 37°C. Samples were stored at −20°C until analysis by quantitative PCR (qPCR).

Expression of seven differentially expressed genes and four reference genes was analyzed by qPCR using the DyNAmo Flash SYBR Green qPCR Kit (Finnzymes) on a Rotor‐gene 6000 real‐time PCR thermocycler (Corbett Robotics). Reactions were performed under the following cycling parameters: enzyme activation and initial denaturation at 95°C for 7 min, followed by 35 to 40 cycles of denaturation at 95°C for 10 s; annealing for 10 s; extension at 72°C for 15 s; and a 2 s acquiring temperature that was set for each specific primer pair based on melt‐curve analysis (Supplemental Data, Table S1). Reactive oxygen species modulator 1 (ROMO1) was analyzed using GoTaq qPCR Master Mix (Promega) and the following protocol: 95°C 2 min, followed by 40 cycles of 95°C 10 s, annealing 15 s, 72°C 15 s, and a 2 s acquiring temperature (Supplemental Data, Tables S1). All reactions were performed in duplicate, and the absence of nonspecific PCR products was verified by melt curve analysis and agarose gel electrophoresis. Quality control included analysis of no template controls and no‐reverse transcriptase (RT) control cDNA samples. Between three and five replicate samples were analyzed for each dose.

Glutathione and oxidative status bioindicators

Because Hg exposure is known to elicit adaptive responses to oxidative stress and oxidative damage in birds, liver biochemistries were analyzed from a population of MeHg‐exposed laughing gull hatchlings from which we subsampled the hatchlings used for the qPCR analysis. Basic methods and assay conditions are described in Hoffman and Heinz 20. Hepatic oxidative status indicator assays of potential Hg‐related effects included (1) enzyme activities for total glutathione peroxidase (TGSHpx), selenium‐dependent glutathione peroxidase (SGSHpx), glutathione reductase (GR), glutathione‐_S_‐transferase (GST), and glucose‐6‐phosphate dehydrogenase (G6PDH); (2) thiol status measurements for reduced GSH, oxidized glutathione (GSSG), total sulfhydryl (TSH), and protein‐bound sulfhydryl (PBSH) concentrations; and (3) lipid peroxidation recorded as thiobarbituric acid reactive substances (TBARS). Protein‐bound sulfhydryl (PBSH; TSH minus GSH) and the ratio of GSSG to GSH (GSSG:GSH) were calculated using the measured endpoints. Portions of the liver were homogenized (1:10 weight/volume [w/v]) in ice‐cold 1.15% KCl‐0.01MNa, K‐phosphate buffer (pH 7.4). Whole homogenate was used as the starting basis for determining GSH, TSH, PBSH, GSSG, and TBARS concentrations. The 10,000 g supernatant was used for all of the assays of enzymes related to GSH metabolism and antioxidant activity as previously described 20.

Statistics

Optimal reference genes from the four potential candidates were evaluated using Normfinder 21. The threshold for determination of Cq was set in the exponential phase at 0.1 normalized fluorescence units. We used Genex Version 5.3.2 (MultiD Analyses AB) for preprocessing of data (efficiency correction, reference gene normalization, and calculation of relative quantities). We calculated the reaction‐specific efficiencies for each sample and primer pair using the Comparative Quantitation function included in the Rotor‐gene software and used the mean amplification efficiency for all samples per run for each gene in our calculations (Supplemental Data, Table S1). Relative quantities were calculated using the mean Cq for the control samples as the calibrator. Statistical analyses of qPCR data were performed using SPSS 13.0 (SPSS): p < 0.05 was considered statistically significant, and p < 0.07 was considered marginally significant. Data were analyzed for violations of assumptions for parametric analysis (heteroscedacity and normality of residuals). Analysis of variance (ANOVA) followed by post‐hoc analysis using Dunnett's two‐sided test on log2 transformed data was used to determine differences in gene expression between dosed samples and controls. Analysis of GSH and oxidative status indicators was performed using GraphPad Instat V3.1 and SPSS 13.0. Data were analyzed for violations of assumptions for parametric analysis (heteroscedacity and normality of residuals). One‐way ANOVA followed by Dunnett's post test or the Games‐Howell procedure (TSH) was used to compare treated samples with controls. Spearman's correlation coefficients were examined to identify any correlations between oxidative stress indicators and MeHg dose.

RESULTS

Suppression subtractive hybridization

Using SSH, we identified 15 (7 up‐regulated and 8 down‐regulated) homologs of previously published avian genes, which were putatively differentially expressed in response to Hg (Table 1). We chose eight of these genes for further analysis by qPCR: thyroid hormone–responsive spot 14 (THRSP), type 2 iodothyronine 5′‐deiodinase (DIO2), ROMO1, cytochrome c oxidase subunit I (COI), ferritin (FER), methionine adenosyltransferase 1 alpha (MAT1A), transferrin (Tf), and cytochrome c oxidase subunit II (COII).

Analysis of qPCR data

Normfinder analysis selected 28s ribosomal RNA and glyceraldehyde‐3‐phosphate dehydrogenase as the optimal reference genes from the four candidates. This pair was then used for normalization of samples. Expression data for each MeHg dose were analyzed and expressed relative to those of control samples. Quantitative PCR results were directionally consistent with SSH for six genes (ROMO1, THRSP, Tf, FER, COI, and DIO2).

Quantitative PCR revealed that MAT1A mRNA expression was up‐regulated relative to controls at all treatment levels except at 1.6 µg/g Hg (Fig. 1). Expression ratios generally increased with exposure, revealing a dose‐related response up to 0.8 µg/g Hg. Methionine adenosyltransferase 1 alpha expression at 0.4 and 0.8 µg/g Hg was fivefold higher than in controls. Statistically significant differences in MAT1A levels were observed between doses (ANOVA; F = 3.170, p = 0.02). Post‐hoc analysis identified a statistically significant increase in expression at 0.4 µg/g Hg (p = 0.028) relative to control levels and a marginally significant increase at 0.8 µg/g Hg (p = 0.06) relative to controls. Expression of the two cytochrome c oxidase subunit genes, COI (ANOVA, F = 6.303, p < 0.001) and COII (ANOVA, F = 3.545, p = 0.012), differed significantly between treatments (Fig. 1). Post‐hoc analysis revealed that at 0.4 µg/g Hg expression of COI was significantly up‐regulated (p = 0.004), and that of COII was marginally significant (p = 0.058). The expression of MAT1A and both CO genes returned to baseline levels at 1.6 µg/g Hg. Reactive oxygen species modulator 1 expression was not significantly affected by MeHg dose (Fig. 1).

Figure 1.

Relative expression of reactive oxygen species modulator 1 (ROMO‐1), methionine adenosyltransferase 1 alpha (MAT1A), cytochrome c oxidase subunit I (COI), and cytochrome c oxidase subunit II (COII) in laughing gull hatchlings exposed in ovo to methylmercury (MeHg). Error bars represent the standard error of the mean. Results were normalized to the expression of glyceraldehyde‐3‐phosphate dehydrogenase and 28s ribosomal RNA, and relative quantities were calculated based on the average mRNA expression of control samples. Analysis of variance (ANOVA) followed by Dunnett's two‐sided test was used to determine significant differences relative to controls. * Statistically significant difference in expression from the controls (p < 0.05). # Marginally statistically significant difference in expression from the controls (p < 0.07).

Figure 2 depicts expression profiles for FER and Tf. Peak expression for FER was observed at at 0.2 and 0.4 µg/g, although the expression levels did not differ significantly between doses (ANOVA, F = 1.478, p = 0.23). Transferrin expression remained near baseline levels until 0.1 µg/g Hg, but at 0.4 µg/g Hg and higher, expression decreased up to 1.6‐fold. Transferrin expression differed significantly between doses (ANOVA, F = 2.547, p = 0.049; post‐hoc not significant). Both THRSP and DIO2 expression did not show a consistent dose‐related response (Fig. 2). The changes observed in THRSP and DIO2 expression were not statistically significant.

Figure 2.

Relative expression of transferrin (Tf), ferritin (FER), type 2 iodothyronine 5′‐deiodinase (DIO2), and thyroid hormone–responsive spot 14 (THRSP) in laughing gull hatchlings exposed in ovo to methylmercury (MeHg). Error bars represent the standard error of the mean. Results were normalized to the expression of glyceraldehyde‐3‐phosphate dehydrogenase and 28s ribosomal RNA, and relative quantities were calculated based on the average mRNA expression of control samples. Analysis of variance (ANOVA) followed by Dunnett's two‐sided test was used to determine significant differences relative to controls.

Glutathione and oxidative status bioindicators

Analysis of variance identified significant effects of MeHg on TSH (F = 2.876, p = 0.019) and GST (F = 2.375, p = 0.024) activity in a larger population of MeHg‐exposed laughing gull hatchlings (Fig. 3). Glutathione‐S‐transferase activity gradually decreased in a dose‐responsive manner (p < 0.0001, r = −0.471), appearing significantly different from the control group at 0.4 µg/g Hg (Dunnett's, p = 0.006) and remaining approximately the same thereafter (Fig. 3). Total sulfhydryl was also significantly lower in the 0.4 ppm group than in controls (Games‐Howell, p = 0.009) but was not significantly correlated with dose. A statistically significant positive correlation with MeHg dose was observed for G6PDH activity (p = 0.052, r = 0.273), whereas a significant negative correlation was observed for PBSH concentration (p = 0.012, r = −0.347) (Fig. 3). In all, GSH, GSSG, and TBARS concentrations, the ratio of GSSG:GSH, and TGSHpx, SGSHpx, and GR activities did not differ significantly at any dose from their respective controls (Table 2; Fig. 3), nor did they exhibit any overall correlations with dose.

Figure 3.

Levels of glutathione and oxidative status indicators in laughing gull hatchlings. Error bars represent the standard deviation. Statistically significant correlations with methylmercury (MeHg) dose were observed for glucose‐6‐phosphate dehydrogenase (G6PDH; p = 0.0303, r = +0.3037), protein bound sulfhydryl (PBSH; p = 0.012, r = −0.347) and glutathione‐_S_‐transferase (GST) activity (p < 0.0001, r = −0.471). Analysis of variance (ANOVA) followed by Dunnett's two‐sided test or the Games‐Howell procedure (total sulfhydryl [TSH]) was used to determine significant differences relative to controls. *Statistically significant difference from controls (p < 0.05).

Table 2.

Concentrations (SD) or activity levels (SD) of oxidative status‐related bioindicators in laughing gull hatchlings

a

Activity expressed as nmoles/min/mg of 10,000 g supernatant protein.

MeHg = methylmercury;TGSHpx = total glutathione peroxidase; SGSHpx = selenium‐dependent glutathione peroxidase; GR = glutathione reductase; GSSG = oxidized glutathione; GSH = glutathione; TBARS = thiobarbituric acid reactive substances; SD = standard deviation.

Table 2.

Concentrations (SD) or activity levels (SD) of oxidative status‐related bioindicators in laughing gull hatchlings

a

Activity expressed as nmoles/min/mg of 10,000 g supernatant protein.

MeHg = methylmercury;TGSHpx = total glutathione peroxidase; SGSHpx = selenium‐dependent glutathione peroxidase; GR = glutathione reductase; GSSG = oxidized glutathione; GSH = glutathione; TBARS = thiobarbituric acid reactive substances; SD = standard deviation.

DISCUSSION

In the present study, we investigated hepatic gene expression and biochemical indicators in laughing gulls to better understand the molecular responses of an avian species that is relatively tolerant to MeHg stress. With the exception of the highest dose (1.6 µg/g Hg), all of the doses employed in the present study were below the LC50 calculated by Heinz et al. 13 for laughing gull embryos. Only limited data are available on Hg concentrations in laughing gulls collected in the wild. Egg concentrations of 0.019 to 0.093 µg/g have been reported in one study in Mobile Bay, Alabama, USA (L.M. Showalter, 2010, Master's thesis), which fall within range of the doses used here.

We identified 15 genes by SSH whose activity was putatively altered by MeHg exposure and focused on a subset of these for qPCR analysis. These eight genes were chosen to study a broad range of possible molecular responses from MeHg exposure that involved processes related to cysteine synthesis, thyroid hormone activity, iron transport and storage, and cellular respiration. The results for thyroid hormone‐related genes and ROMO1 suggest that the expression patterns are largely unaffected by MeHg. The differences in expression patterns observed between SSH and qPCR may be attributed to individual variations in expression of the studied genes. Because SSH used a single sample at each concentration to develop the libraries, the genes identified as differentially expressed by SSH are potentially unique to the individuals used. Because qPCR validation used multiple replicates, some discrepancy between the two methods is not unexpected.

Exposure of embryos to MeHg in the present study induced MAT1A mRNA levels in a dose‐dependent manner up to fivefold at treatments less than 1.6 µg/g Hg. Methionine adenosyltransferase, the enzyme encoded by MAT1A, catalyzes the formation of _S_‐adenosylmethionine, the initial reaction in the transsulfuration pathway that converts available methionine to homocysteine to produce cysteine 22. Cysteine is crucial for the formation of GSH, one of the most important scavengers of ROS and a conjugator of a wide range of xenobiotics, including Hg 15, 23. The metal‐scavenging protein metallothionein, which is also known to bind Hg, is composed of approximately one‐third cysteine and thus competes directly with GSH for available cysteine 24. Cysteine itself is also known to directly bind MeHg 25. Our results suggest that exposure to MeHg initiates an increase in the initial component of the transsulfuration pathway, potentially as a compensatory mechanism to generate more cysteine for GSH or metallothionein synthesis.

Two of the differentially expressed genes identified by SSH in the present study, FER and Tf, are involved in iron binding and storage. Transferrin serves as a transport vehicle for iron through serum to various tissues in the body, whereas FER is responsible for the storage and release of iron and is active in maintaining adequate levels of iron at times of deficiency or overload 26. Iron storage prevails over iron transfer and uptake in the presence of increased ROS to limit the amounts of free iron available to interact in the Fenton reaction, thereby reducing the production of the reactive hydroxyl radical OH · and the proliferation of oxidative stress 27. The iron‐modulated protective response to oxidative stress involves the repression of the Tf receptor and up‐regulation of FER 27. Ferritin is also known to sequester metallic Hg and other heavy metals 28, 29. Ferritin therefore plays a dual role—promotion of iron storage to minimize oxidative stress and sequestration of Hg. In the present study, at 0.4 µg/g Hg and above, Tf expression declined and FER was up‐regulated, although not significantly, suggesting that oxidative stress, although potentially increasing with dose, is limited.

Not only is induction of ROS scavengers an important stress response, but so is repression of ROS generation. The avian mitochondrial enzyme cytochrome c oxidase (Complex IV: COX) is composed of 9 to 10 subunits and is involved in the final stage of the electron transport chain of aerobic respiration 30. Complex IV aids in depleting electron‐rich intermediates responsible for the formation of ROS in the mitochondrial respiratory chain, and increased COX activity suppresses ROS production 31. Under oxidative stress, COX activity and therefore expression of its core subunits have been found to increase. We observed differential expression of two COX subunits (COI and COII) in response to MeHg. The expression patterns of the two subunits mirrored one another. At low doses, expression was similar to controls but began to increase at 0.2 µg/g Hg to a maximum at 0.4 µg/g, and then decreased again at the two highest doses. Presumably, COI and COII mRNA expression increased to counter the production of electron‐rich intermediates by the electron transport chain and suppress production of ROS. Up‐regulation of the COI gene has also been identified in zebrafish exposed to MeHg 12 and in clams, mussels, and oysters exposed to chronic cadmium 32. In both cases the response was considered an adaptive mechanism for coping with increased oxidative stress.

At 1.6 µg/g, the expression of MAT1A, FER, and the CO genes returned to levels similar to those in controls. Heinz et al. 13 found that only 28% of laughing gull embryos survived through 90% of incubation at 1.6 µg/g Hg. The observed decline in mRNA levels may play a role in the laughing gull's low survival rate at high doses of MeHg. Similar patterns were observed in a study of human liver carcinoma cells exposed to graded doses of Hg, which found that expression of 13 stress‐response genes increased in a dose‐dependent manner at subacute exposures to Hg but fell sharply to baseline levels at doses surrounding the LC50 33. Tchounwou et al. 34 reported a similar decrease in metallothionein and 70 kDa heat shock protein mRNA expression at high concentrations of Cd and Hg in human hepatoma cells attributable to cell death. The return of gene expression to baseline levels in Figures 1 and 2 may be a culmination of adverse effects caused by MeHg toxicity.

The biochemical endpoints we analyzed in the present study supported the gene expression patterns and suggest that hepatic oxidative stress in laughing gull hatchlings is minimal. We observed little change in GSH status or levels of lipid peroxidation across doses in laughing gull hatchlings. Lipid peroxidation is a well‐studied and strong indicator of oxidative damage, which triggers the production of reactive intermediates and results in damage of proteins and DNA 35. No changes in TBARS concentrations were observed in our samples, indicating that lipid peroxidation levels did not vary between doses.

Glutathione‐_S_‐transferase remained near baseline levels through 0.4 µg/g Hg. At higher Hg doses (≥0.8 µg/g), GSH began to increase in a dose‐dependent manner parallel to increasing G6PDH activity, although the change was not statistically significant. Methylmercury complexes the cysteine of GSH and is excreted from hepatocytes into the bile 22. Therefore, GSH conjugation irreversibly consumes intracellular GSH. Under this scenario, to maintain hepatic GSH levels, GSH must be synthesized de novo or be recycled from GSSG. However, GR activity, which regenerates GSH from GSSG (the oxidized form) and GSHpx activity, which catalyzes the reaction whereby GSH is oxidized to GSSG, were maintained at baseline levels in our study. Likewise, GSSG levels were stable. This suggests that if the administered MeHg is being bound by GSH in the laughing gull hatchlings, the levels of GSH are being maintained by de novo synthesis. The up‐regulation of MAT1A we observed may in part be necessary for de novo formation of GSH, keeping concentrations stable at these MeHg doses and preventing GSH from otherwise being depleted by conjugation with Hg. Additional studies focusing on specific steps of the GSH synthetic pathway, specifically the rate‐limiting gamma‐glutamylcysteine synthetase step, could help clarify whether our observations are in fact a result of de novo formation.

We observed a qualitative similarity between the levels of GSH and G6PDH activity. Glucose‐6‐phosphate dehydrogenase is the first and rate‐limiting enzyme of the oxidative pentose phosphate pathway, which produces reduced nicotinamide adenine dinucleotide phosphate and various five‐carbon sugars. Reduced nicotinamide adenine dinucleotide phosphate plays a role in several detoxification pathways as a cofactor for antioxidant enzymes, including GR, catalase, and thioredoxin reductase 36. Because we saw no change in GR activity, the increase in G6PDH suggests that its activity is being directed toward a different purpose, perhaps fatty acid synthesis. Although at high doses Hg is a known inhibitor of this process 37, up‐regulation of fatty acid synthesis has been observed in the liver of zebrafish exposed to sublethal doses of HgCl2 38.

The observed decreases in TSH and PBSH, most likely attributable to the binding of Hg to various structural and functional proteins, have been reported previously in mallard and great egret livers after Hg exposure 20, 39. The binding of protein thiol groups is a major mechanism of enzyme inhibition and can alter the nature and activity of proteins within cells, disturbing both protein synthesis and energy production 40. However, a number of sulfhydryl‐containing proteins are known metal‐binding agents that thereby neutralize their toxic effects. Because GSH levels were primarily stable, the change in TSH can be attributed to a decrease in PBSH. Whether the decrease in PBSH is in fact a reflection of damage or of binding for the purpose of detoxification is unknown.

Hepatic GST activity also decreased in our samples with increasing Hg dose, suggesting a direct effect of Hg on GST activity. Although GSTs catalyze the nucleophilic attack of GSH on electrophilic/lipophilic substrates, Dierickx 41 reported that in rat liver, organomercurials are not conjugated to GSH by GST‐catalyzed reactions. Instead, Hg compounds interact with GSTs directly through binding, which serves a protective role 41. Henny et al. 42 reported that, compared to birds caught at a reference site, young double‐crested cormorants (Phalacrocorax auritus) exposed to MeHg at the Lower Carson River, Nevada, USA, exhibited a decrease in GST activity, despite liver total Hg concentrations that were approximately six times higher than in controls. Cormorants, like laughing gulls, exhibit relatively low sensitivity to injected MeHg 13. In contrast, Henny et al. 42 found that GST activity was higher in young snowy egrets (Egretta thula) from the high Hg site relative to the reference site. According to Heinz et al. 13, snowy egrets exhibit high sensitivity to MeHg. Studies involving other avian species at later stages of the avian life cycle found that Hg most often either stimulated GST activity or had no effect 16, 20, 42. In those studies, GST stimulation was identified as a protective compensatory mechanism in response to oxidative stress but was accompanied by increased lipid peroxidation (TBARS), which we did not observe here.

The genomic and biochemical responses of birds exposed to MeHg provide critical insights into activated or suppressed biological pathways that may allow a particular species or individual to survive and potentially thrive despite exposure to this common contaminant. Despite the increased mortality observed in laughing gulls at 0.4 µg/g Hg and higher 13, the surviving embryos appear not to experience hepatic oxidative stress. However, whether hepatic oxidative stress plays a direct role in the deaths of those embryos that did not hatch is unknown. The laughing gulls apparently have sufficient mechanisms by which hepatic oxidative stress is prevented, including suppression of electron‐rich intermediates and direct binding of Hg by cysteine‐containing compounds, GST, and protein‐bound thiols. The results also suggest that 0.4 µg/g Hg represents a potentially significant concentration for the hatchlings. At this point, the hatchlings begin to show the initiation of several potentially compensatory mechanisms, indicating that their ability to counter MeHg effects using baseline processes may be overwhelmed. Our results provide new data on the interplay of various interrelated responses to MeHg in birds and serve as a starting point for future investigations of genetic susceptibility to Hg in other less sensitive avian species.

SUPPLEMENTAL DATA

Table S1 (98 KB DOC).

Acknowledgements

We are indebted to G. Heinz for providing us with the dosed hatchlings used for these analyses and for his suggestions, discussions, and reviews of this manuscript. We thank Y. Chen and L. Robertson for their reviews of an earlier version of this manuscript. We thank C. Maddox for her assistance. Funding for this work was provided by the CALFED Bay‐Delta Program and the U.S. Geological Survey. Use of product names does not imply endorsement by the U.S. Government.

REFERENCES

1

Evers

D

,

Burgess

N

,

Champoux

L

,

Hoskins

B

,

Major

A

,

Goodale

W

,

Taylor

R

,

Poppenga

R

,

Daigle

T.

2005

.

Patterns and interpretation of mercury exposure in freshwater avian communities in northeastern North America

.

Ecotoxicology

14

:

193

–

221

.

2

Wolfe

MF

,

Schwarzbach

S

,

Sulaiman

RA.

1998

.

Effects of mercury on wildlife: A comprehensive review

.

Environ Toxicol Chem

17

:

146

–

160

.

3

Scheuhammer

AM

,

Perrault

JA

,

Bond

DE.

2001

.

Mercury, methylmercury, and selenium concentrations in eggs of common loons (Gavia immer) from Canada

.

Environ Monit Assess

72

:

79

–

94

.

4

Heinz

GH

,

Hoffman

DJ.

2003

.

Embryotoxic thresholds of mercury: Estimates from individual mallard eggs

.

Arch Environ Contam Toxicol

44

:

257

–

264

.

5

Heinz

GH.

1979

.

Methylmercury: Reproductive and behavioral effects on three generations of mallard ducks

.

J Wildl Manag

43

:

394

–

401

.

6

Eagles‐Smith

CA

,

Ackerman

JT

,

Yee

J

,

Adelsbach

TL.

2009

.

Mercury demethylation in waterbird livers: Dose‐response thresholds and differences among species

.

Environ Toxicol Chem

28

:

568

–

577

.

7

Chan

J

,

Huang

Z

,

Merrifield

ME

,

Salgado

MT

,

Stillman

MJ.

2002

.

Studies of metal binding reactions in metallothioneins by spectroscopic, molecular biology, and molecular modeling techniques

.

Coord Chem Rev

233–234

:

319

–

339

.

8

Mahapatra

CT

,

Bond

J

,

Rand

DM

,

Rand

MD.

2010

.

Identification of methylmercury tolerance gene candidates in Drosophila

.

Toxicol Sci

116

:

225

–

238

.

9

Brambila

E

,

Liu

J

,

Morgan

DL

,

Beliles

RP

,

Waalkes

MP.

2002

.

Effect of mercury vapor exposure on metallothionein and glutathione s‐transferase gene expression in the kidney of nonpregnant, pregnant and neonatal rats

.

J Toxicol Environ Health

65

:

1273

–

1288

.

10

de Nazare Klautau‐Guimarães

M

,

D'Ascenção

R

,

Caldart

FA

,

Grisolia

CK

,

Souza

JRd

,

Barbosa

AC

,

Cordeiro

CMT

,

Ferrari

I.

2005

.

Analysis of genetic susceptibility to mercury contamination evaluated through molecular biomarkers in at‐risk Amazon Amerindian populations

.

Gen Mol Biol

28

:

827

–

832

.

11

Monetti

C

,

Vigetti

D

,

Prati

M

,

Sabbioni

E

,

Bernardini

G

,

Gornati

R.

2002

.

Gene expression in Xenopus embryos after methylmercury exposure: A search for molecular biomarkers

.

Environ Toxicol Chem

21

:

2731

–

2736

.

12

Gonzalez

P

,

Dominique

Y

,

Massabuau

JC

,

Boudou

A

,

Bourdineaud

JP.

2005

.

Comparative effects of dietary methylmercury on gene expression in liver, skeletal muscle, and brain of the zebrafish (Danio rerio)

.

Environ Sci Technol

39

:

3972

–

3980

.

13

Heinz

GH

,

Hoffman

DJ

,

Klimstra

JD

,

Stebbins

KR

,

Kondrad

SL

,

Erwin

CA.

2009

.

Species differences in the sensitivity of avian embryos to methylmercury

.

Arch Environ Contam Toxicol

56

:

129

–

138

.

14

Kim

EY

,

Murakami

T

,

Saeki

K

,

Tatsukawa

R.

1996

.

Mercury levels and its chemical form in tissues and organs of seabirds

.

Arch Environ Contam Toxicol

30

:

259

–

266

.

15

do Nascimento

JLM

,

Oliveira

KRM

,

Crespo‐Lopez

ME

,

Macchi

BM

,

Maues

LAL

,

Pinheiro

MDN

,

Silveira

LCL

,

Herculano

AM.

2008

.

Methylmercury neurotoxicity and antioxidant defenses

.

Ind J Med Res

128

:

373

–

382

.

16

Hoffman

DJ

,

Eagles‐Smith

CA

,

Ackerman

JT

,

Adelsbach

TL

,

Stebbins

KR.

2011

.

Oxidative stress response of Forster's terns (Sterna forsteri) and Caspian terns (Hydroprogne caspia) to mercury and selenium bioaccumulation in liver, kidney, and brain

.

Environ Toxicol Chem

30

:

920

–

929

.

17

Heinz

GH

,

Hoffman

DJ

,

Kondrad

SL

,

Erwin

CA.

2006

.

Factors affecting the toxicity of methylmercury injected into eggs

.

Arch Environ Contam Toxicol

50

:

264

–

279

.

18

Fridolfsson

A

,

Ellegren

H.

1999

.

A simple and universal method for molecular sexing of non‐ratite birds

.

J Avian Biol

30

:

116

–

121

.

19

Altschul

SF

,

Gish

W

,

Miller

W

,

Myers

EW

,

Lipman

DJ.

1990

.

Basic local alignment search tool

.

J Mol Biol

215

:

403

–

410

.

20

Hoffman

DJ

,

Heinz

GH.

1998

.

Effects of mercury and selenium on glutathione metabolism and oxidative stress in mallard ducks

.

Environ Toxicol Chem

17

:

161

–

166

.

21

Andersen

CL

,

Jensen

JL

,

Ørntoft

TF.

2004

.

Normalization of real‐time quantitative reverse transcription‐PCR data: A model‐based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets

.

Cancer Res

64

:

5245

–

5250

.

22

Lu

SC.

1999

.

Regulation of hepatic glutathione synthesis: Current concepts and controversies

.

FASEB J

13

:

1169

–

1183

.

23

Dutczak

WJ

,

Ballatori

N.

1994

.

Transport of the glutathione‐methylmercury complex across liver canalicular membranes on reduced glutathione carriers

.

J Biol Chem

269

:

9746

–

9751

.

24

Susanto

I

,

Wright

SE

,

Lawson

RS

,

Williams

CE

,

Deneke

SM.

1998

.

Metallothionein, glutathione, and cystine transport in pulmonary artery endothelial cells and NIH/3T3 cells

.

Am J Physiol Lung Cell Mol Physiol

274

:

L296

–

L300

.

25

Clarkson

TW

,

Magos

L.

2006

.

The toxicology of mercury and its chemical compounds

.

Crit Rev Toxicol

36

:

609

–

662

.

26

Hinzmann

R.

1999

.

Ferritin and transferrin in iron deficiency and overload

.

Immunodiagnostics Today

12

:

1

–

11

.

27

Morel

Y

,

Barouki

R.

1999

.

Repression of gene expression by oxidative stress

.

Biochem J

342

:

481

–

496

.

28

Li

Q

,

Chen

H

,

Huang

X

,

Costa

M.

2006

.

Effects of 12 metal ions on iron regulatory protein 1 (IRP‐1) and hypoxia‐inducible factor‐1 alpha (HIF‐1alpha]) and HIF‐regulated genes

.

Toxicol Appl Pharmacol

213

:

245

–

255

.

29

Ogata

M

,

Ikeda

M.

1978

.

Mercury uptake by acatalasemia mice and their erythrocytes, lung and liver homogenates

.

Int Arch Occup Environ Health

41

:

87

–

93

.

30

Garcia‐Horsman

JA

,

Barquera

B

,

Rumbley

J

,

Ma

JX

,

Gennis

RB.

1994

.

The superfamily of heme‐copper respiratory oxidases

.

J Bacteriol

176

:

5587

–

5600

.

31

Campian

JL

,

Qian

M

,

Gao

X

,

Eaton

JW.

2004

.

Oxygen tolerance and coupling of mitochondrial electron transport

.

J Biol Chem

279

:

46580

–

46587

.

32

Achard‐Joris

M

,

Gonzalez

P

,

Marie

V

,

Baudrimont

M

,

Bourdineaud

J‐P.

2006

.

Cytochrome c oxidase subunit I gene is up‐regulated by cadmium in freshwater and marine bivalves

.

Biometals

19

:

237

–

244

.

33

Sutton

D

,

Tchounwou

P

,

Ninashvili

N

,

Shen

E.

2002

.

Mercury induces cytotoxicity and transcriptionally activates stress genes in human liver carcinoma (HepG2) cells

.

Int J Mol Sci

3

:

965

–

984

.

34

Tchounwou

PB

,

Ishaque

AB

,

Sutton

D

,

Ninashvili

N

,

Shen

E.

2002

.

Biomarkers of sensitivity and effect associated with cadmium and mercury toxicity in human liver carcinoma (HepG(2)) cells

.

Metal Ions Biol Med

7

:

132

–

137

.

35

Monaghan

P

,

Metcalfe

NB

,

Torres

R.

2009

.

Oxidative stress as a mediator of life history trade‐offs: mechanisms, measurements and interpretation

.

Ecol Lett

12

:

75

–

92

.

36

Koivula

MJ

,

Eeva

T.

2010

.

Metal‐related oxidative stress in birds

.

Environ Pollut

158

:

2359

–

2370

.

37

Thayer

RH

,

Donaldson

WE.

1975

.

Mercury inhibition of fatty acid synthesis in chicks

.

Chem‐Biol Interact

11

:

235

–

243

.

38

Ung

C

,

Lam

S

,

Hlaing

M

,

Winata

C

,

Korzh

S

,

Mathavan

S

,

Gong

Z.

2010

.

Mercury‐induced hepatotoxicity in zebrafish: in vivo mechanistic insights from transcriptome analysis, phenotype anchoring and targeted gene expression validation

.

BMC Genomics

11

:

212

.

39

Hoffman

DJ

,

Spalding

MG

,

Frederick

PC.

2005

.

Subchronic effects of methylmercury on plasma and organ biochemistries in great egret nestlings

.

Environ Toxicol Chem

24

:

3078

–

3084

.

40

Bando

I

,

Reus

MIS

,

Andrés

D

,

Cascales

M.

2005

.

Endogenous antioxidant defence system in rat liver following mercury chloride oral intoxication

.

J Biochem Mol Toxicol

19

:

154

–

161

.

41

Dierickx

PJ.

1985

.

In vitro interaction of organic mercury compounds with soluble glutathione s‐transferases from rat liver

.

Pharmacol Res Commun

17

:

489

–

500

.

42

Henny

CJ

,

Hill

EF

,

Hoffman

DJ

,

Spalding

MG

,

Grove

RA.

2002

.

Nineteenth century mercury: Hazard to wading birds and cormorants of the Carson River, Nevada

.

Ecotoxicology

11

:

213

–

231

.

Copyright © 2012 SETAC

This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/pages/standard-publication-reuse-rights)