Oxidative status in nestlings of three small passerine species exposed to metal pollution

Science of the Total Environment 454–455 (2013) 466–473

Contents lists available at SciVerse ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Oxidative status in nestlings of three small passerine species exposed to metal pollution Miia J. Rainio a,⁎, Mirella Kanerva b, Juha-Pekka Salminen c, Mikko Nikinmaa b, Tapio Eeva a a b c

Department of Biology, Section of Ecology, University of Turku, FI-20014 Turku, Finland Department of Biology, Division of Genetics and Physiology, University of Turku, FI-20014 Turku, Finland Laboratory of Organic Chemistry and Chemical Biology, 20014 University of Turku, FI-20014 Turku, Finland

H I G H L I G H T S • Species-specific oxidative status was studied in birds exposed to metal pollution. • Species showed interspecific variation in their antioxidant enzyme activities. • Oxidative status was only weakly related to fecal metal exposure.

a r t i c l e

i n f o

Article history: Received 4 December 2012 Received in revised form 31 January 2013 Accepted 10 March 2013 Available online 9 April 2013 Keywords: Antioxidants Birds Enzyme activities Glutathione Oxidative stress ROS

a b s t r a c t Antioxidant defense has an important role in the protection of organisms against oxidative stress caused by reactive oxygen species (ROS). Many metals are capable of generating ROS and inducing oxidative damage, and may therefore lead to changes in oxidative regulation. We studied species-specific variation in the oxidative status of great tit (Parus major), blue tit (Cyanistes caeruleus) and pied flycatcher (Ficedula hypoleuca) nestlings in a vicinity of a non-ferrous smelter. Non-enzymatic (glutathione [tGSH], GSH:GSSG ratio, and carotenoids) and enzymatic (glutathione peroxidase [GP], glutathione-S-transferase [GST], superoxide dismutase [SOD], and catalase [CAT]) antioxidants were evaluated to determine the effects of metal exposure on the oxidative status of the birds. We found strong evidence of interspecific variation in CAT and SOD activities, whereas less variation was observed in parameters related to glutathione metabolism. Oxidative state (in terms of tGSH and GSH: GSSG) did not vary between species, suggesting that different species may employ different antioxidant pathways to achieve the same oxidative state. Oxidative status was only weakly related to metal exposure, and these associations were further obscured by species-specific environmental effects. Our results indicate that effects on oxidative status observed in one species cannot be generalized to other ones. Future work should attempt to incorporate species-specific biology and environmental context into assessments of contaminant impacts on oxidative regulation of passerine birds. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Metals appear naturally in the environment, and several of them participate in important metabolic and signaling pathways (Valko et al., 2005). However, at high concentrations these metals can be harmful to organisms, possibly owing to metal-induced oxidative stress caused by increased production of reactive oxygen species (ROS) (Halliwell and Cross, 1994; Halliwell and Gutteridge, 2007). As with metals, ROS play an important role in cell signaling and regulation of redox status (Jackson, 2005; Thannickal and Fanburg, 2000), but may pose hazards to the organisms when produced in excess (e.g. via oxidation of DNA, proteins, and lipids) (Bae et al., 2009; Beckman and Ames, 1998; Harman, 1956). ⁎ Corresponding author. Tel.: +358 2 333 6050; fax: +358 2 333 6550. E-mail addresses: miikoi@utu.fi (M.J. Rainio), mmkane@utu.fi (M. Kanerva), j-p.salminen@utu.fi (J.-P. Salminen), miknik@utu.fi (M. Nikinmaa), teeva@utu.fi (T. Eeva). 0048-9697/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2013.03.033

Metals can be found in different forms in the environment, their toxicity being related to their oxidative state and reactivity with other compounds (Scheuhammer, 1987; Valko et al., 2005; Walker, 1995). Metals can be divided either redox-active or -inactive metals according to their function. Redox-active metals (e.g. copper and iron) catalyze Fenton reactions, which generate reactive hydroxyl radicals and are commonly associated with membranous fractions such as mitochondria, microsomes and peroxisomes, whereas redox-inactive metals (e.g. lead, nickel and cadmium) deplete major cellular antioxidants, such as glutathione. Thus, metals are able to either increase ROS production or reduce antioxidant defense (Ercal et al., 2001; Stohs and Bagchi, 1995; Valko et al., 2005). Antioxidants are molecules that protect against oxidative damage by reducing ROS (or preventing their formation), thereby rendering them unable to cause further damage to the cell. Antioxidants may be taken up directly from the environment (as in the case of carotenoids), or they may be generated internally by a series of enzymatic and

M.J. Rainio et al. / Science of the Total Environment 454–455 (2013) 466–473

non-enzymatic reactions (Sies, 1993). Glutathione is an antioxidant that has been found in almost all organisms examined to date (Andrews, 2000; Pinto et al., 2003). It functions by binding metals at sulfhydryl groups, thereby preventing them from creating ROS (Andrews, 2000; Pinto et al., 2003). Each glutathione molecule cycles between reduced (GSH) and oxidized (GSSG) states, with the overall ratio of GSH:GSSG serving as an indicator of current oxidative onslaught. Cycling between GSH and GSSG is catalyzed by a number of enzymes, including glutathione peroxidase (GP) and glutathione-S-transferase (GST) (Halliwell and Gutteridge, 2007). In normal cells, we expect to see a relatively high GSH:GSSG ratio, and low activity of associated enzymes. In contrast, a low GSH:GSSG ratio is indicative of oxidative stress (Halliwell and Gutteridge, 2007; Hoffman, 2002; Isaksson et al., 2005; Stephensen et al., 2002). In the event that metal concentrations overwhelm the protective capacity of glutathione, cells also have a number of mechanisms through which ROS can be deactivated directly. Two enzymes, superoxide dismutase (SOD) and catalase (CAT), are among these regulatory antioxidants (Ercal et al., 2001; Pinto et al., 2003). In this paper, the term oxidative state is used to refer to oxidative state in terms of tGSH and the ratio of GSH:GSSG, and the term oxidative status is used to cover both enzyme activities (GP, GST, SOD and CAT) and the antioxidant GSH and the GSH:GSSG ratio. In this paper, we report on antioxidant (tGSH, GSH:GSSG ratio, and carotenoids) levels and antioxidant enzyme (GP, GST, SOD, and CAT) activities in nestlings of three passerine species, great tits (Parus major), blue tits (Cyanistes caeruleus), and pied flycatchers (Ficedula hypoleuca), living near a metal smelter in Harjavalta, Finland. Previous studies have reported reductions in survival, reproductive success, genetic integrity, plumage brightness, carotenoid levels, and food availability associated with living on metal-contaminated areas (Berglund et al., 2010; Dauwe et al., 2006; Eeva et al., 2006; Eeva and Lehikoinen, 1996; Eeva et al., 2003, 1998; Geens et al., 2010), but little is known about management of oxidative stress (Bel'skii and Stepanova, 1995; Berglund et al., 2007; Geens et al., 2009; Koivula et al., 2011). Studies that have sought to address this question have yielded conflicting results, with some researchers reporting increased oxidative stress at polluted sites, and others reporting no apparent association (Berglund et al., 2007; Isaksson et al., 2005, 2009; Koivula et al., 2011). Part of the reason for this discrepancy may be that different species regulate antioxidant metabolism in different ways. The goals of our study were two-fold: First, we sought to generate baseline data on regulation of antioxidants in three avian species while accounting for variation due to environment and phylogeny. Second, we attempted to determine whether oxidative regulation of nestlings reared on contaminated areas deviated from these baselines as a result of increased exposure to metals. The use of several species in the same study set-up enables us to standardize many sources of variation, such as habitat quality, breeding season, tissue used in the analyses and measurement techniques. Fecal samples, which have been proven to be good indicators of metal pollution, especially when studying food chain contamination (Berglund et al., 2011; Dauwe et al., 2000, 2004; Eeva and Lehikoinen, 1996), were used to measure actual metal exposure. In insectivorous birds the transfer of metals via the food chain has been considered as a major source of metal contamination (Berglund et al., 2009). Plasma carotenoids were also examined, because carotenoids are small-molecule antioxidants and their levels have been shown to vary in relation to environmental pollution due to pollution-related changes in food webs (Eeva et al., 2008, 2009b; Geens et al., 2009).

467

smelter area are nickel (Ni) and copper (Cu), but arsenic (As), zinc (Zn), lead (Pb) and sulfuric oxides are also emitted in appreciable amounts (Kiikkilä, 2003). Studies of wild birds have been ongoing in this area for the past 21 years, and several effects of pollution exposure have been reported in past until today (Eeva et al., 2009a; Eeva and Lehikoinen, 1996, 1998; Koivula et al., 2011), in spite of fact that pollution levels have declined considerably since the 1990s (Kozlov et al., 2009; Kubin et al., 2000). We used thirteen study sites, each with 40–50 nest boxes at a distance of 0.6–11.2 km from the smelter, in three main directions (SW, SE and NW) (Fig. 1). Seven sites were in the polluted area (zone 1; b2.5 km from the smelter) and six in the unpolluted area (zone 2; >2.5 km from the smelter), where the pollution levels are close to the background levels (Eeva et al., 2008). We used similar wooden nest boxes for each species that were hanged 2 m above the ground level onto the trees, at the range of 35–40 m from each other. A description of the nest boxes is given in Lambrechts et al. (2010). The habitat type was barren Scots pine (Pinus sylvestris) dominated forest in all our study areas to avoid habitat-related variation between the study areas. All three study species are abundant at our sites and breed in the nest boxes. Altogether 243 nestlings of the great tits (n = 93), blue tits (n = 47) and pied flycatchers (n = 103) were included in this study. Approximately half of the nestlings were born in polluted area and half in unpolluted areas (great tit; 47 in zone 1 and 46 in zone 2, blue tit; 31 in zone 1 and 16 in zone 2, pied flycatcher; 52 in zone 1and 51 in zone 2). We visited nest boxes daily throughout the breeding season in order to collect information on clutch size, hatching success, and fledging success. The study was performed under the licenses of the Animal Care & Use Committee of Turku University (ESLH-2008-02274/Ym-23) and Regional Environment Centre (LOS-2008-L-224-254). 2.2. Sampling methods 2.2.1. Sampling in the field All nestlings were uniquely ringed at either 6 (pied flycatchers) or 7 (great tits and blue tits) days of age. At 9 (pied flycatchers) or 11 (great and blue tits) days old, we randomly selected two nestlings per brood (excluding runts) on which to perform measurements and collect blood samples. The difference between the times of sample collection was due to difference in the nestling development rate between pied flycatcher and the Parid species, pied flycatchers developing faster than Parids. The blood samples were taken by brachial venipuncture using 75 μl heparinized capillary tubes, and centrifuged immediately for 5 min at 4000 rpm to separate the plasma from red blood cells. Plasma and red blood cells were then immediately placed in liquid nitrogen until permanent storage at −80 °C. 2.2.2. Carotenoid analyses We determined plasma concentrations of lutein, zeaxanthin, and β-carotene using high performance liquid chromatography (HPLC). Briefly, a known volume of plasma (10–35 μl) was extracted 3× with 100% acetone. The solvent was evaporated from the combined extract under vacuum and the residue was dissolved into a small volume of 80% acetone. The carotenoid composition of the extracts (lutein, zeaxanthin, and β-carotene) was analyzed with HPLC at 450 nm using a Merck Purospher STAR RP-18 (55 × 2mm, i.d., 3 μm) column (Darmstadt, Germany). β-carotene was quantified as β-carotene and other carotenes as lutein equivalents.

2. Materials and methods 2.1. Study area and study species The study was conducted in summer 2008 in the vicinity of a metal smelter complex in the town of Harjavalta (61°20′ N, 22°10′ E), one of the most metal polluted areas in Finland. The main pollutants in the

2.2.3. Fecal metal analyses As for blood samples, we analyzed fecal metal concentrations of two nestlings per brood. Following collection, fecal sacs were pooled by brood and dried at 50 °C for 72 h. Thereafter, we combined 0.15–0.20 g of fecal material with 2 ml of Supra-pure HNO3 and 0.5 ml of H2O2 were added to the samples in Teflon bombs for digestion with a microwave system

468

M.J. Rainio et al. / Science of the Total Environment 454–455 (2013) 466–473

Fig. 1. Map of the study areas in the town of Harjavalta. Black dots represent the study areas and the crosses indicate the city centers. The circle around the smelter represents the polluted area, less than 2.5 km from the smelter.

(Milestone High Performance Microwave Digestion Unit mls 1200 mega). The samples were then diluted to 50 ml with de-ionized water. Concentrations of Ca, Cu, Ni, As, Cd, and Pb were determined with ICP-MS (Elan 6100 DRC, Perkin-Elmer-Sciex, Boston, USA). Instrument calibration was performed using certified solution (Claritas PPT, Multi element solution 2A) from Spex Certiprep, and samples were prepared according to the method instructions (see more detailed description in Eeva et al., 2008). The detection limit for most of the metals was around 1 ng/l. Certified mussel tissue (ERM-CE278; Ni and Ca not included) was used as a reference material. Mean recoveries (±SE) in four reference samples were as follows: As 100 ± 0.84%, Cd 100 ± 1.68%, Cu 105 ± 0.82% and Pb 103 ± 1.25%. Since concentrations of different metals were positively correlated, principal component analyses were used to generate a single metric of metal exposure for each study species. Since the first principal component (PC1) explained 69.8–78.7% of the variation in our data (pied flycatcher 72.3% (eigenvalue 3.61), great tit 78.7% (eigenvalue 3.93) and blue tit 69.8% (eigenvalue 3.49)), it was used in the models as an explanatory variable to describe the general level of metal exposure. 2.2.4. Biomarker analyses We analyzed antioxidant levels in the red blood cells of 238 nestlings. All enzyme activity measurements were done in triplicate (intra-assay coefficient of variability [CV] b 10% in all cases) using either 96- (CAT) or 384-well (GP, GST, SOD, tGSH and GSH:GSSG ratio) microplates with Envision microplate reader (Perkin-Elmer Wallac, Turku, Finland). Three control samples were used in every plate to be able to correct the intra-assay precision with the ratio specific to the particular plate (range 0.8–1.2). As we attempted to minimize sample volume, we also used correspondingly smaller reagent volumes than recommended by analytical kits. GP, GST and CAT were measured using Sigma kits (Sigma Chemicals, St. Louis, Missouri, USA), and SOD was measured with a Fluka kit (Fluka, Buchs, Germany) according to

kit instructions. The protein concentration for all samples was measured according to Bradford method (Bradford, 1976) using BioRad stock (BioRad, Espoo, Finland) diluted to dH2O (1:5) and BSA (1 mg/ml) (Sigma) as a standard and measured with Envision microplate reader at absorbance of 595 nm. Total GSH and the GSH:GSSG ratio were measured with ThioStar® glutathione detection reagent (Arbor Assays, Michigan, USA) using GSH as the standard (Sigma Chemicals, St. Louis, Missouri, USA). Briefly, each sample (containing 5% SSA, sulfosalicylic acid) was diluted 1:5 with 100 mM Na-phosphate buffer and 5 mM EDTA (pH 7.5) to obtain 1% SSA. Standards were diluted with 100 mM Na-phosphate + 5 mM EDTA + 1% SSA. For each measurement, 12.5 μl of sample, standard, or blank was pipetted onto a plate, and 6.5 μl of ThioStar reagent was added. The plate was incubated for 15 min in the dark and the fluorescence was measured (excitation 405 nm, emission 510 nm) to determine GSH concentration. After the first measurement, 6.5 μl of 4 mM NADPH + 8 U ml−1 GR (glutathione reductase) in 100 mM Na-phosphate buffer + 5 mM EDTA (pH 7.5) was added to the wells. The plate was again incubated for 15 min in the dark, after which time fluorescence was again measured to determine the total glutathione concentration (GSH + GSSG).

2.2.5. Temperature data All our study species have nestlings that are ectothermic during early development. This means that variation in ambient temperature likely influences nestling body temperature, metabolic rate and quite possibly, oxidative stress response. In fact, previous work in this population of great tits demonstrated a positive correlation between temperature and GP activity during early nestling development (Koivula et al., 2011). Therefore, we calculated for each nestling the mean (daily) temperature during the first 9 (pied flycatchers) or 11 (great tit and blue tit) days of life to use it as a covariate in our analyses. We used temperature data collected by Finnish Meteorological

M.J. Rainio et al. / Science of the Total Environment 454–455 (2013) 466–473

Institute, at the station within our study area (the meteorological station of Peipohja, Kokemäki, 61°16′ N, 22°15′ E). 2.2.6. Statistics Statistical analyses were conducted with SAS statistical software 9.2 (SAS, 2008). All enzyme activities were log-transformed before the analyses to normalize distributions. Oxidative status measures below the instrument detection limit (DL) were replaced by an estimated value using the log–probit regression method (Helsel, 1990; Sinha et al., 2006). In general, fewer than 6% of values fell below the DL. The only exceptions were CAT activity in pied flycatchers (72.6% below DL) and GP activity in blue tits (12.8% below DL). Nestling fecal metal concentrations in zones 1 and 2 were compared for each study species using a linear mixed model (LMM, Type III analysis). Interspecific and zone differences in oxidative status biomarkers (GP, GST, tGSH, GSH:GSSG, CAT and SOD) were analyzed with generalized linear mixed models (GLMM) with a Gaussian distribution and identity link function (GLIMMIX procedure in SAS). Species, zone and their two-way interaction were used as independent variables and brood as a random factor to account for non-independence of the two nestlings sampled from the same brood. Degrees of freedom were calculated with Kenward–Roger method, and post-hoc pairwise comparisons were carried out using Tukey's test. The lower number of individuals used in the measurement of tGSH level and the GSH: GSSG ratio is due to missing values, where either tGSH or GSH:GSSG ratio could not be calculated from the sample and thus had to be excluded from the data. Three exceptionally high values (pied flycatcher with 25.5 nmol GP/min/mg; blue tit with 180.2 μmol tGSH/mg; great tit with 28.7 GSH:GSSG ratio) were excluded from the data as outliers. Zone-wide differences in oxidative status biomarkers may suggest broad variation between metal-polluted and reference areas (including pollution-related habitat effects), but they cannot directly implicate metal exposure per se as a causal agent of change. In order to investigate more closely the direct influence of metal exposure on antioxidant activity, we performed a second set of analyses (GLMM with Gaussian distribution and identity link function, as above), where variables affecting oxidative status were analyzed using PC1 of fecal metal concentration, body mass, brood size, temperature, and total carotenoid concentration as explaining factors and brood as a random factor in the model. We performed a separate GLMM for each study species and, once again, non-significant terms were dropped from the final models. Spatial autocorrelation between the nest boxes was analyzed using Moran's I. Since none of the species showed spatial autocorrelation, there was no need to take that into account in the models. 3. Results 3.1. Metal concentrations between polluted and unpolluted areas Fecal concentrations of As, Cu, Ni, Pb and Cd were significantly higher in polluted area (zone 1) than in unpolluted area (zone 2) in all of our study species (Table 1). The only exception was Ca, which did not show any difference between the study areas.

had significantly higher SOD (t = 11.11107.2, p ≤ 0.0001), but lower CAT (t = − 22.33100.5, p ≤ 0.0001) activity than great tits (Fig. 2). Pied flycatcher and blue tit were likewise similar in their activities of GST (t = 2.06102, p = 0.103), but differed in activities of GP (t = 6.31111.8, p ≤ 0.0001), SOD (t = 6.79113.2, p ≤ 0.0001) and CAT (t = − 13.08103.6, p ≤ 0.0001), blue tit showing lower GP and SOD activities, but higher CAT activity than pied flycatcher (Fig. 2). The CAT activity, however, needs to be considered with caution, because in pied flycatcher nestlings most of the activities (72.6%) were under the detection limit (0.97 nmol/min/mg), indicating very low CAT activity in the nestlings of pied flycatcher. Parid species differed from each other in GP (t = −6.57112, p ≤ 0.0001), GST (t = −3.25103.1, p = 0.004) and CAT (t = −4.41105.8, p ≤ 0.0001) activities, great tit having significantly higher activities, but had similar activity of SOD (t = 1.87113.1, p = 0.151) (Fig. 2). Oxidative status differed also between zone 1 and zone 2, with GP (F = 6.471, 117.3, p = 0.012) and SOD (F = 6.301, 108.3, p = 0.014) activities, and the ratio of GSH:GSSG (F = 4.661, 95.67, p = 0.033), polluted areas showing higher GP and SOD activities and lower GSH: GSSG ratio (Fig. 2). However, post-hoc pairwise comparisons revealed that zone differences existed only for GP activity (F = 6.581, 20.49, p = 0.018) and for the ratio of GSH:GSSG (F = 5.831, 18.53, p = 0.026) (Fig. 2) in blue tits, whereas none of the species exhibited significant differences in SOD activity. 3.3. Oxidative status in relation to metal concentration, body mass, brood size and temperature 3.3.1. Great tits The fecal metal concentration did not significantly affect the oxidative status of great tit nestlings (Table 2). Instead, the tGSH decreased significantly with warmer developmental temperatures. Also GP activity had negative, but statistically insignificant association with temperature (Table 2). CAT activities increased with increasing body mass (Table 2). GST and SOD activities and the GSH:GSSG ratio did not show any association with studied parameters (Table 2). 3.3.2. Blue tits GP and GST activities increased significantly with increasing fecal metal concentration in blue tit nestlings (Table 2). However, the ratio of GSH:GSSG remained unaffected. GST activity also showed up-regulation in nestlings raised under warmer conditions and in those with higher

Table 1 Mean (±standard error) metal (μg/g; dry weight) and Ca (mg/g; dry weight) concentrations in the feces of Parus major, Cyanistes caeruleus and Ficedula hypoleuca nestlings in polluted (Zone 1) vs. unpolluted (Zone 2) areas. Species

Metal

Zone 11

P. major

As Cu Ni Pb Cd Ca As Cu Ni Pb Cd Ca As Cu Ni Pb Cd Ca

8.97 166.3 25.8 6.02 2.90 14.2 5.96 176.7 31.1 8.58 2.63 5.7 7.67 269.4 35.2 7.76 4.21 3.9

C. caeruleus

3.2. Oxidative status in relation to species and area Oxidative state was similar across the three study species (tGSH; F = 0.082, 99.01, p = 0.926 and GSH:GSSG; F = 0.612, 92.8, p = 0.544), but there was substantial variation in the enzyme activities used to achieve this oxidative state (GP; F = 24.832, 114.8, p ≤ 0.0001, GST; F = 5.312, 106.3, p = 0.006, SOD; F = 65.632, 110, p ≤ 0.0001 CAT; F = 263.542, 102.9, p ≤ 0.0001; Fig. 2). Pied flycatcher and great tit exhibited similar GP (t = − 0.46119.4, p = 0.888) and GST (t = − 1.53112.4, p = 0.282) activity profiles, but pied flycatchers

469

F. hypoleuca

1 2

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

Zone 21 0.07 0.06 0.09 0.11 0.07 0.16 0.06 0.07 0.06 0.16 0.06 0.23 0.09 0.11 0.11 0.12 0.05 0.21

1.34 68.4 4.53 1.64 1.16 11.5 1.12 72.0 7.50 3.44 1.48 3.9 0.92 116.9 4.05 1.92 3.48 2.6

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

Fdf2 0.06 0.06 0.05 0.05 0.06 0.11 0.08 0.11 0.11 0.13 0.08 0.28 0.02 0.04 0.05 0.06 0.04 0.21

274.911, 124.971, 227.301, 81.111, 41.551, 1.321, 133.051, 51.831, 139.581, 11.791, 12.931, 1.051, 286.031, 54.311, 325.411, 76.041, 5.371, 2.441,

p 89 89 89 89 89 89 45 45 45 45 45 45 101 101 101 101 101 101

b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 0.2530 b0.0001 b0.0001 b0.0001 0.0013 0.0008 0.3119 b0.0001 b0.0001 b0.0001 b0.0001 0.0225 0.1212

Geometric means. Generalized linear mixed model used to test the significance between the areas.

470

M.J. Rainio et al. / Science of the Total Environment 454–455 (2013) 466–473

b) GSH:GSSG ratio

12 10

3

a

a

8 6 4 2

21 13

40 27

3.5

a

a

b

3

*

2.5 2 1.5 1

45 45

31 16

51 49

GST activity (nmol/min/mg protein) +/- 95% CI

GP activity (nmol/min/mg protein) +/- 95% CI

22 13

d) GST activity a

b

45 45

31 16

3

a, b

2.5 2 1.5 1 0.5

50 51

0

0

f) CAT activity

30

a

a

b

20 15 10 5

44 45 P. major

27 14

47 48

C. caeruleus F. hypoleuca

CAT activity (µmol/min/mg protein) +/- 95% CI

e) SOD activity SOD activity (inhibition %) +/95% CI

1

31 32

4

0

1.5

0

c) GP activity

25

a

2

40 29

0

0.5

2.5

0.5

34 34

3.5

a *

a

a GSH:GSSG ratio

tGSH concentration (µmol/mg protein) +/- 95% CI

a) tGSH concentration

a

120

b

c

100 80 60 40 20

41 40 0

P. major

Species

26 12

46 49

C. caeruleus F. hypoleuca

Species

Fig. 2. Species- and site-specific variation in oxidative status of great tit, blue tit, and pied flycatcher nestlings. Bars represent the geometric means (±95% CI) of (a) non-enzymatic antioxidant tGSH (glutathione) and (b) the ratio of GSH:GSSG (reduced:oxidized glutathione), and the activities of antioxidant enzymes (c) GP (glutathione peroxidase), (d) GST (glutathione-S-transferase), (e) CAT (catalase) and (f) SOD (superoxide dismutase) measured from the blood. Polluted and unpolluted areas are indicated by gray and white bars, respectively. Letters above bars indicate significant differences between species (GLMM, Tukey's test, p b 0.05). Star above bars indicate significant differences between areas (GLMM, p b 0.05). Sample sizes vary due to missing values.

carotenoid concentrations (Table 2). Total GSH levels were also positively associated with temperature, while CAT and the ratio of GSH:GSSG were positively related to nestling body mass (Table 2). SOD did not show any association to the studied parameters.

3.3.3. Pied flycatchers The tGSH level increased significantly with increasing fecal metal concentration in pied flycatcher nestlings (Table 2). However, there was no association between the ratio of GSH:GSSG and metal exposure. GST activity and tGSH level were negatively associated with ambient temperature during early nestling period, while the ratio of GSH:GSSG was positively affected by increasing temperature (Table 2). GST activity was also affected by increases in brood size and nestling body mass, declining in both instances. GP was likewise negatively associated with nestling body mass, whereas the ratio of GSH:GSSG had positive association with body mass. Thus, both GP and GST activities and the GSH: GSSG ratio show similar response to nestling condition (in terms of body mass). Neither SOD nor CAT showed any association to the studied parameters (Table 2).

4. Discussion 4.1. Interspecific differences in oxidative status Although we found no difference in the overall oxidative state of our three study species, there was substantial variation in the manner in which this oxidative state was achieved. In fact, we observed significant interspecific variation in the activity of all four of the antioxidant enzymes we examined (GP, GST, SOD, and CAT). GP and GST are involved in glutathione metabolism, while SOD and CAT work to transform ROS into more stable chemical species. SOD transforms superoxides (O2−) to hydrogen peroxide (H2O2), which is further catalyzed to water and molecular oxygen by CAT enzyme (Fridovich, 1997; Halliwell and Gutteridge, 2007). The fact that both types of enzymes varied according to species suggests that there is significant flexibility in how different species regulate their oxidative state. The fact that two closely related species, the great and blue tit, varied in their oxidative profiles suggests that antioxidant activity may be determined more by contemporary environmental conditions than by shared evolutionary history.

M.J. Rainio et al. / Science of the Total Environment 454–455 (2013) 466–473

471

Table 2 The relationship between environmental factors and antioxidant activities in (1) Parus major, (2) Cyanistes caeruleus and (3) Ficedula hypoleuca. All enzyme activities were analyzed in separately for nestlings of each species using PC1 of fecal metal levels, body mass (g), brood size (number of hatchlings), total carotenoid concentration in plasma (μg/ml) and temperature (°C) as explanatory factors (GLMM with Gaussian distribution and identity link function). Non significant terms were dropped sequentially from each model. The terms left in final model are shown in bold. Biomarkers

GP (nmol/min/mg)

Source of variation

Fdf (Est, SE)

2

p

GST (nmol/min/mg) Fdf (Est, SE)

Model

P. major

Fecal metal level (PC1) 0.061, 39 Body mass 0.801, 57.1

0.807 3.091, 0.376 2.401,

Brood size Carotenoids Temperature

0.266 0.151, 0.815 1.221, 0.064 0.481,

1.271, 40.07 0.061, 75.7 3.621, 42.67

56.33

36.21 77.50 38

0.022 11.231, 21.06 (0.13, 0.04) 0.641 1.351, 23.63

Brood size Carotenoids

0.011, 19.45 0.591, 34.08

Temperature

0.041, 17.26

0.932 0.001, 18.18 0.449 4.681, 36.12 (0.01, 0.004) 0.839 5.561, 17.99 (0.10, 0.04) 0.407 1.491, 48.12

Body mass Brood size Carotenoids Temperature

2

39.79

Fecal metal level (PC1) 6.091, 21.36 (0.11, 0.04) Body mass 0.221, 21.56

F. hypoleuca Fecal metal level (PC1) 0.701, 59.63

1

tGSH (μmol/mg) 2

p

GSH:GSSG ratio

p

Fdf (Est, SE)

Fdf (Est, SE)

0.087 0.127

0.091, 28.4 1.081, 48.88

0.760 2.791, 35.15 0.304 0.011, 48.9

0.698 0.273 0.492 0.003

0.021, 26.28 0.011, 44.74 6.331, 44.44 (−0.12, 0.05) 0.011, 14.51

0.257

0.061, 22.26

0.965 0.037

0.191, 13.98 0.961, 25.5

0.030

0.007

8.271, 16.92 (0.14, 0.05) 7.371, 44.49 (0.10, 0.04) 1.131, 38.64

0.001

0.001, 34.34

2

p

SOD (inhibition %)

CAT (μmol/min/mg)

Fdf

Fdf (Est, SE)2 p

p

1

Species

C. caeruleus

2

3.971, 76.63 0.050 7.801, 66.32 (−0.07, 0.04) (−0.07, 0.03) 0.427 12.021, 52.13 0.641, 54.14 (−0.10, 0.03) 0.445 1.031, 61.03 0.591, 78.21 0.281, 44.13 0.600 12.751, 46.01 (−0.25, 0.07)

0.228

0.145 0.037

0.880 1.001, 25.64 0.923 0.501, 46.41 0.016 0.001, 36.33

0.104 0.071, 39.73 0.786 2.231, 32.08 0.935 0.301, 50.65 0.584 4.651, 42.02 (0.06, 0.03) 0.326 1.231, 44.40 0.273 0.021, 32.39 0.481 1.651, 73.13 0.203 0.251, 71.28 0.975 0.011, 38.56 0.923 2.041, 35.68

0.919 0.821, 12.63

0.382 2.061, 16.85 0.170 0.311,

0.585

0.816 4.231, 27.09 (0.13, 0.06) 0.673 0.261, 13.36 0.335 0.691, 26.99

0.049 0.191, 32.26 0.666 4.141, 30.88 (0.19, 0.09) 0.619 1.721, 22.24 0.203 0.061, 20.02 0.413 0.051, 30.8 0.820 1.441, 25.25

0.809 0.241

0.011 0.001, 14.76

0.995 0.921, 17.39 0.351 0.001,

18.58

0.993

0.009 2.941, 36.37

0.095 0.281, 46.85 0.602 0.011,

48.44

0.937

0.294 6.221, 51.4 (0.30, 0.14) 0.961 0.331, 35.9

0.016 0.191, 51.27 0.664 0.491,

54.17

0.486

0.571 2.371, 46.05 0.130 0.061,

51.51

0.816

0.315 2.481, 57.15 0.121 0.051, 55.99 0.0008 10.971, 37.09 0.002 4.741, 36.65 (−0.43, 0.13) (0.11, 0.04)

0.820 0.711, 71.52 0.404 0.561, 0.036 2.711, 49.6 0.106 0.681,

19.79

77.69 47.73

0.892 0.621 0.162

0.051

0.456 0.412

Brood used as a random factor in the model. Est = estimate, SE = standard error.

4.2. Oxidative status and metal contamination One of the major goals of this study was to investigate whether passerines living on metal-polluted sites experienced increased oxidative stress or altered enzyme activity as a result of metal exposure. We approached this question in two ways, first, by looking for broad differences across areas, and second, by examining the effect of individual metal exposure on antioxidant profiles. The oxidative status did differ also between polluted and unpolluted areas, but further analyses revealed that this difference was largely driven by area-specific variation in GP and the GSH:GSSG ratio in blue tits. GP activity was higher in blue tits living on contaminated areas, while GSH:GSSG ratio was lower. Coots exposed to high levels of Pb have also exhibited elevated GP and SOD levels (Martinez-Haro et al., 2011), as did domestic ducks (Shaoxing duck, Anas platyrhynchos) exposed to high levels of selenium (Se) and mercury (Hg) (Ji et al., 2006). Both enzymes act as free radical scavengers, for example protecting against such dangers as cell injury or neuropathology (Ji et al., 2006). The up-regulation of GP and GST in blue tits might provide evidence of a protective response against metal exposure (Martinez-Haro et al., 2011), reflecting the activation of antioxidant defense (Halliwell and Gutteridge, 2007; Koivula and Eeva, 2010). However, if oxidative enzymes such as GP prove insufficient in combating metal exposure, increased oxidative stress may result, providing one possible explanation for the simultaneously high GP activity and low GSH:GSSG ratio observed in blue tits living on contaminated areas. In any case, the fact remains that the observed area-specific differences, even when statistically significant, were minor and not at all generalizable across passerine species. This point has seldom been discussed in the ecotoxicological literature. To date, there have been only few studies, where oxidative stress has been measured across multiple bird

species in their natural habitats (Costantini et al., 2007, 2010; Hegseth et al., 2011; Hoffman et al., 1998; Martinez-Haro et al., 2011) and even fewer in which similar tissues and measurement techniques have been used (Hegseth et al., 2011; Hoffman et al., 1998; Martinez-Haro et al., 2011). Thus, there is very little understanding of how species-specific biology influences antioxidant metabolism. Such knowledge may be important for researchers hoping to assess the impact of contaminants on various avian species, especially given the highly species-specific nature of the effects that we found. One alternative explanation for the patterns we observed is that metal accumulation, rather than oxidative regulation, varied according to species. In fact, previous work has demonstrated that pied flycatchers living in contaminated areas accumulate metals to a greater degree than Parids (Bel'skii et al., 1995; Berglund et al., 2011; Eeva and Lehikoinen, 1996, 2004). If species accumulate contaminants differently, then variation in oxidative status may be more a matter of degree than kind. In addition, metal-related variation in oxidative status may be masked in studies which only investigate site differences if proximity to the smelter does not accurately reflect individual metal exposure. We tested both of these possibilities by examining oxidative profiles of individual nestlings with respect to fecal metal concentration. Overall, metal profiles were relatively similar across the three study species, apart from somewhat higher Cu and Cd levels in pied flycatcher nestlings, suggesting that differences in accumulation may not be responsible for the species-specific variation in oxidative status that we observed. The main difference between species was in Ca levels, blue tits and pied flycatchers having significantly lower Ca levels than great tits, indicating that blue tit and pied flycatchers may be more vulnerable to metal related detrimental effects with low Ca level in their diet (Eeva et al., 2009a). Likewise, there was little evidence of an association between individual metal exposure and nestling oxidative status. Of all of the biomarkers that we examined, only GP and GST

472

M.J. Rainio et al. / Science of the Total Environment 454–455 (2013) 466–473

activities in blue tits, and tGSH level in pied flycatchers, showed any association with metal contamination.

4.3. Oxidative status and nestling condition Aspects of nestling condition were associated with oxidative state or enzyme activities in all three study species. However, the direction and strength of these relationships varied considerably. In great tits, CAT activity was positively related to nestling body mass. In blue tits, CAT activity likewise exhibited a positive correlation to mass, as did the GSH:GSSG ratio. In addition, GST activity was positively associated with carotenoid concentration. Pied flycatchers showed perhaps the greatest condition-dependent oxidative response. GST and GP activities were negatively affected by increasing mass, while GST was likewise influenced by variation in brood size. The GSH:GSSG ratio instead had positive association with increased body mass, showing similar response to body mass than the enzyme activities. Positive correlations between CAT and nestling mass indicate that heavier nestlings experienced higher CAT activity. Since CAT removes ROS from the cell environment (by transformation to water), higher CAT activity favors the maintenance of glutathione in its reduced state (GSH). Thus, heavier Parid nestlings may be better equipped to handle high levels of stress-induced ROS. In blue tits, this hypothesis may be evidenced by the corresponding association between mass and GSH:GSSG ratio: heavier nestlings tended to have higher GSH: GSSG ratios, indicating more effective defense against oxidative stress. It is interesting to note that blue tit nestlings living on contaminated sites tended to be smaller than conspecifics raised on uncontaminated sites. These same nestlings were also more likely to exhibit symptoms of oxidative stress (low GSH:GSSG ratio), perhaps indicating an indirect pollution effect. Although, in pied flycatchers, many of the associations between nestling condition and oxidative stress biomarkers were in the opposite direction of those found in Parids, the oxidative stress tended to be lower (high GSH:GSSG ratio) in nestlings of higher mass also in pied flycatchers. GP and GST were both up-regulated in lighter nestlings of pied flycatchers, while GST activity was also higher in nestlings from smaller broods. These findings in pied flycatcher nestlings, together with the higher ratio of GSH:GSSG in heavier nestlings, suggest that heavier nestlings may be less vulnerable to oxidative stress than the lighter ones, especially when taking into account the fact that smaller nestlings are often found in polluted areas.

4.4. Oxidative status and temperature Ambient temperature during nestling development strongly influenced oxidative regulation, in particular, glutathione metabolism. However, once again, many trends fell in opposite directions across our study species. In great tits, tGSH levels were negatively influenced by increasing temperature, as were tGSH level and GST activity in pied flycatchers. In pied flycatchers, there was also a positive association between temperature and GSH:GSSG ratio, indicating that nestlings raised in warmer conditions experienced less oxidative stress. In contrast, blue tits exhibited positive associations between GST activity, tGSH level and temperature. There was also strong positive correlation between hatching day and temperature in all three species (data not shown), indicating that later hatched nestlings faced also higher temperatures. It might be that some environmental factors such as precipitation or food quality have been changed in time, thus affecting the birds hatched later in summer, which could be reflected in the tGSH levels as well. However, more specific studies are needed to confirm the connection between glutathione metabolism and temperature.

4.5. Conclusions This study sought to determine whether oxidative status varied with metal exposure in three passerine species. While we found some evidence in favor of this view, perhaps our most noteworthy discovery was our inability to generalize results across study species. Indeed, each responded to metal exposure in different, and in some cases, contradictory ways. Even in the absence of metal pollution, birds living on uncontaminated sites employed wildly divergent methods of achieving a similar oxidative state, some of which were undoubtedly influenced by short-term environmental conditions. Such variation speaks to the importance of incorporating species-specific biology into assessments of ecotoxicological impact. More dose-dependent field studies are also needed to reveal what are the exposure levels of different contaminants that may cause increased oxidative stress in different species, as well as how to separate the variation caused by pollution exposure from that caused by other ambient factors. Acknowledgments We thank Hanna Tuominen, Jorma Nurmi and Jari Lehto for their great help during the field season. We also thank Paul Ek (Åbo Akademi) for metal analyses and Terhi Sundman (University of Turku) for performing carotenoid analyses. We are grateful to Kalle Rainio and Åsa Berglund for valuable comments on the manuscript. We also want to thank one anonymous referee for very valuable comments that markedly improved this manuscript. The study was financed by Academy of Finland (TE; Project 8119367) and The Finnish Cultural Foundation (MR). References Andrews GK. Regulation of metallothionein gene expression by oxidative stress and metal ions. Biochem Pharmacol 2000;59:95–104. Bae YA, Cai G-B, Kim S-H, Zo Y-G, Kong Y. Modular evolution of glutathione peroxidase genes in association with different biochemical properties of their encoded proteins in invertebrate animals. BMC Evol Biol 2009;9:72. Beckman KB, Ames BN. The free radical theory of aging matures. Physiol Rev 1998;78: 547–81. Bel'skii EA, Stepanova ZL. On industrial contamination effect on the state of nestlings of the hollow-nesting birds. A Tribute to Prof. V.V. StanchinskyThe 2nd International Proceeding, Smolensk, Russia; 1995. p. 96–9. Bel'skii EA, Bezel VS, Polents EA. Early stages of the nesting period of hollow-nesting birds under conditions of industrial pollution. Russ J Ecol 1995;26:38–43. Berglund ÅMM, Sturve J, Förlin L, Nyholm NEI. Oxidative stress in pied flycatcher (Ficedula hypoleuca) nestlings from metal contaminated environments in northern Sweden. Environ Res 2007;105:330–9. Berglund ÅMM, Klaminder J, Nyholm NEI. Effects of reduced lead deposition on pied flycatcher (Ficedula hypoleuca) nestlings: tracing exposure routes using stable lead isotopes. Environ Sci Technol 2009;43:208–13. Berglund ÅMM, Ingvarsson PK, Danielsson H, Nyholm NEI. Lead exposure and biological effects in pied flycatchers (Ficedula hypoleuca) before and after the closure of a lead mine in northern Sweden. Environ Pollut 2010;158:1368–75. Berglund ÅMM, Koivula MJ, Eeva T. Species- and age-related variation in metal exposure and accumulation of two passerine bird species. Environ Pollut 2011;159: 2368–74. Bradford MM. Rapid and sensitive method for quantitation of microgram quantities of protein utilizing principle of protein–dye binding. Anal Biochem 1976;72:248–54. Costantini D, Cardinale M, Carere C. Oxidative damage and anti-oxidant capacity in two migratory bird species at a stop-over site. Comp Biochem Physiol C 2007;144:363. Costantini D, Carello L, Fanfani A. Relationships among oxidative status, breeding conditions and life-history traits in free-living great tits Parus major and common starlings Sturnus vulgaris. Ibis 2010;152:793–802. Dauwe T, Bervoets L, Blust R, Pinxten R, Eens M. Can excrement and feathers of nestling songbirds be used as biomonitors for heavy metal pollution? Arch Environ Contam Toxicol 2000;39:541–6. Dauwe T, Janssens E, Bervoets L, Blust R, Eens M. Relationships between metal concentrations in great tit nestlings and their environment and food. Environ Pollut 2004;131:373–80. Dauwe T, Janssens E, Eens M. Effects of heavy metal exposure on the condition and health of adult great tits (Parus major). Environ Pollut 2006;140:71–8. Eeva T, Lehikoinen E. Growth and mortality of nestling great tits (Parus major) and pied flycatchers (Ficedula hypoleuca) in a heavy metal pollution gradient. Oecologia 1996;108:631–9. Eeva T, Lehikoinen E. Local survival rates of the pied flycatchers (Ficedula hypoleuca) and the great tits (Parus major) in an air pollution gradient. Écoscience 1998;5:46. Eeva T, Lehikoinen E. Rich calcium availability diminishes heavy metal toxicity in pied flycatcher. Funct Ecol 2004;18:548–53.

M.J. Rainio et al. / Science of the Total Environment 454–455 (2013) 466–473 Eeva T, Lehikoinen E, Rönkä M. Air pollution fades the plumage of the Great Tit. Funct Ecol 1998;12:607–12. Eeva T, Lehikoinen E, Nikinmaa M. Pollution-induced nutritional stress in birds: an experimental study of direct and indirect effects. Ecol Appl 2003;13:1242. Eeva T, Belskii E, Kuranov B. Environmental pollution affects genetic diversity in wild bird populations. Mutat Res 2006;608:8–15. Eeva T, Sillanpää S, Salminen JP, Nikkinen L, Tuominen A, Toivonen E, et al. Environmental pollution affects the plumage color of great tit nestlings through carotenoid availability. Ecohealth 2008;5:328–37. Eeva T, Ahola M, Lehikoinen E. Breeding performance of blue tits (Cyanistes caeruleus) and great tits (Parus major) in a heavy metal polluted area. Environ Pollut 2009a;157:3126. Eeva T, Sillanpää S, Salminen J. The effects of diet quality and quantity on plumage colour and growth of great tit Parus major nestlings: a food manipulation experiment along a pollution gradient. J Avian Biol 2009b;40:491. Ercal N, Gurer-Orhan H, Aykin-Burns N. Toxic metals and oxidative stress part I: mechanisms involved in metal induced oxidative damage. Curr Top Med Chem 2001;1:529–39. Fridovich I. Superoxide anion radical (O-2 radical anion), superoxide dismutases, and related matters. J Biol Chem 1997;272:18515–7. Geens A, Dauwe T, Eens M. Does anthropogenic metal pollution affect carotenoid colouration, antioxidative capacity and physiological condition of great tits (Parus major)? Comp Biochem Physiol C 2009;150:155–63. Geens A, Dauwe T, Bervoets L, Blust R, Eens M. Haematological status of wintering great tits (Parus major) along a metal pollution gradient. Sci Total Environ 2010;408:1174–9. Halliwell B, Cross CE. Oxygen derived species — their relation to human disease and environmental stress. Environ Health Perspect 1994;102:5–12. Halliwell B, Gutteridge J. Free radicals in biology and medicine. New York: Oxford University Press; 2007. Harman D. Aging — a theory based on free-radical and radiation chemistry. J Gerontol 1956;11:298–300. Hegseth MN, Camus L, Helgason LB, Bocchetti R, Gabrielsen GW, Regoli F. Hepatic antioxidant responses related to levels of PCBs and metals in chicks of three Arctic seabird species. Comp Biochem Physiol C 2011;154:28–35. Helsel DR. Less than obvious — statistical treatment of data below the detection limit. Environ Sci Technol 1990;24:1766–74. Hoffman DJ. Role of selenium toxicity and oxidative stress in aquatic birds. Aquat Toxicol 2002;57:11–26. Hoffman DJ, Ohlendorf HM, Marn CM, Pendleton GW. Association of mercury and selenium with altered glutathione metabolism and oxidative stress in diving ducks from the San Francisco Bay region, USA. Environ Toxicol Chem 1998;17:167–72. Isaksson C, Oernborg J, Stephensen E, Andersson S. Plasma glutathione and carotenoid coloration as potential biomarkers of environmental stress in great tits. Ecohealth 2005;2:138–46. Isaksson C, Sturve J, Almroth BC, Andersson S. The impact of urban environment on oxidative damage (TBARS) and antioxidant systems in lungs and liver of great tits, Parus major. Environ Res 2009;109:46–50. Jackson MJ. Reactive oxygen species and redox-regulation of skeletal muscle adaptations to exercise. Philos Trans R Soc Lond B Biol Sci 2005;360:2285–91.

473

Ji XL, Hu WX, Cheng JP, Yuan T, Xu F, Qu LY, et al. Oxidative stress on domestic ducks (Shaoxing duck) chronically exposed in a mercury–selenium coexisting mining area in China. Ecotoxicol Environ Saf 2006;64:171–7. Kiikkilä O. Heavy-metal pollution and remediation of forest soil around the Harjavalta Cu–Ni smelter, in SW Finland. Silva Fenn 2003;37:399. Koivula MJ, Eeva T. Metal-related oxidative stress in birds. Environ Pollut 2010;158: 2359–70. Koivula MJ, Kanerva M, Salminen JP, Nikinmaa M, Eeva T. Metal pollution indirectly increases oxidative stress in great tit (Parus major) nestlings. Environ Res 2011;111: 362–70. Kozlov MV, Zvereva EL, Zverev VE. Impacts of point polluters on terrestrial biota: comparative analysis of 18 contaminated areas introduction, 15. Netherland: Springer; 2009. p. 466. Kubin E, Lippo H, Poikolainen J. Heavy metal loading. In: Mälkönen E, editor. Forest condition in a changing environment — the Finnish case. Kluwer Academic Publishers; 2000. p. 60–71. Lambrechts MM, Adriaensen F, Ardia DR, Artemyev AV, Atiénzar F, Banbura J, et al. The design of artificial nestboxes for the study of secondary hole-nesting birds: a review of methodological inconsistencies and potential biases. Acta Ornithol 2010;45:1–26. Martinez-Haro M, Green AJ, Mateo R. Effects of lead exposure on oxidative stress biomarkers and plasma biochemistry in waterbirds in the field. Environ Res 2011;111:530–8. Pinto E, Sigaud-Kutner TCS, Leitao MAS, Okamoto OK, Morse D, Colepicolo P. Heavy metal-induced oxidative stress in algae. J Phycol 2003;39:1008–18. SAS. SAS. Cary, NC, USA: SAS Institute, 2008. Scheuhammer AM. The chronic toxicity of aluminum, cadmium, mercury, and lead in birds — a review. Environ Pollut 1987;46:263–95. Sies H. Strategies of antioxidant defense. Eur J Biochem 1993;215:213–9. Sinha P, Lambert MB, Trumbull VL. Evaluation of statistical methods for left-censored environmental data with nonuniform detection limits. Environ Toxicol Chem 2006;25:2533–40. Stephensen E, Sturve J, Förlin L. Effects of redox cycling compounds on glutathione content and activity of glutathione-related enzymes in rainbow trout liver. Comp Biochem Physiol C 2002;133:435–42. Stohs SJ, Bagchi D. Oxidative mechanisms in the toxicity of metal ions. Free Radic Biol Med 1995;18:321–36. Thannickal VJ, Fanburg BL. Reactive oxygen species in cell signaling. Am J Physiol Lung Cell Mol Physiol 2000;279:L1005–28. Valko M, Morris H, Cronin MTD. Metals, toxicity and oxidative stress. Curr Med Chem 2005;12:1161–208. Walker CH. Biochemical biomarkers in ecotoxicology — some recent developments. Sci Total Environ 1995;171:189–95.