Scholarly article on topic 'Contrasting effects of hypoxia on copper toxicity during development in the three-spined stickleback (Gasterosteus aculeatus)'

Contrasting effects of hypoxia on copper toxicity during development in the three-spined stickleback (Gasterosteus aculeatus) Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Environmental Pollution
OECD Field of science
Keywords
{Oxygen / "Toxic metal" / Freshwater / "Combined stressors" / Fish}

Abstract of research paper on Biological sciences, author of scientific article — Jennifer A. Fitzgerald, Ioanna Katsiadaki, Eduarda M. Santos

Abstract Hypoxia is a global problem in aquatic systems and often co-occurs with pollutants. Despite this, little is known about the combined effects of these stressors on aquatic organisms. The objective of this study was to investigate the combined effects of hypoxia and copper, a toxic metal widespread in the aquatic environment. We used the three-spined stickleback (Gasterosteus aculeatus) as a model because of its environmental relevance and amenability for environmental toxicology studies. We focused on embryonic development as this is considered to be a sensitive life stage to environmental pollution. We first investigated the effects of hypoxia alone on stickleback development to generate the information required to design subsequent studies. Our data showed that exposure to low oxygen concentrations (24.7 ± 0.9% air saturation; AS) resulted in strong developmental delays and increased mortalities, whereas a small decrease in oxygen (75.0 ± 0.5%AS) resulted in premature hatching. Stickleback embryos were then exposed to a range of copper concentrations under hypoxia (56.1 ± 0.2%AS) or normoxia (97.6 ± 0.1%AS), continuously, from fertilisation to free swimming larvae. Hypoxia caused significant changes in copper toxicity throughout embryonic development. Prior to hatching, hypoxia suppressed the occurrence of mortalities, but after hatching hypoxia significantly increased copper toxicity. Interestingly, when exposures were conducted only after hatching, the onset of copper-induced mortalities was delayed under hypoxia compared to normoxia, but after 48 h, copper was more toxic to hatched embryos under hypoxia. This is the second species for which the protective effect of hypoxia on copper toxicity prior to hatching, followed by its exacerbating effect after hatching is demonstrated, suggesting the hypothesis that this pattern may be common for teleost species. Our research highlights the importance of considering the interactions between multiple stressors, as understanding these interactions is essential to facilitate the accurate prediction of the consequences of exposure to complex stressors in a rapidly changing environment.

Academic research paper on topic "Contrasting effects of hypoxia on copper toxicity during development in the three-spined stickleback (Gasterosteus aculeatus)"

Environmental Pollution xxx (2016) 1—11

Contents lists available at ScienceDirect

Environmental Pollution

journal homepage: www.elsevier.com/locate/envpol

Contrasting effects of hypoxia on copper toxicity during development in the three-spined stickleback (Gasterosteus aculeatus)*

Jennifer A. Fitzgerald a b *, Ioanna Katsiadaki b, Eduarda M. Santos a **

a Biosciences, College of Life & Environmental Sciences, Geoffrey Pope Building, University of Exeter, Exeter, EX4 4QD, UK b Centre for Environment, Fisheries and Aquaculture Science, Barrack Road, The Nothe, Weymouth, Dorset, DT4 8UB, UK

ARTICLE INFO

ABSTRACT

Article history: Received 1 August 2016 Received in revised form 29 November 2016 Accepted 6 December 2016 Available online xxx

Keywords:

Oxygen

Toxic metal

Freshwater

Combined stressors

Hypoxia is a global problem in aquatic systems and often co-occurs with pollutants. Despite this, little is known about the combined effects of these stressors on aquatic organisms. The objective of this study was to investigate the combined effects of hypoxia and copper, a toxic metal widespread in the aquatic environment. We used the three-spined stickleback (Gasterosteus aculeatus) as a model because of its environmental relevance and amenability for environmental toxicology studies. We focused on embryonic development as this is considered to be a sensitive life stage to environmental pollution. We first investigated the effects of hypoxia alone on stickleback development to generate the information required to design subsequent studies. Our data showed that exposure to low oxygen concentrations (24.7 ± 0.9% air saturation; AS) resulted in strong developmental delays and increased mortalities, whereas a small decrease in oxygen (75.0 ± 0.5%AS) resulted in premature hatching. Stickleback embryos were then exposed to a range of copper concentrations under hypoxia (56.1 ± 0.2%AS) or normoxia (97.6 ± 0.1%AS), continuously, from fertilisation to free swimming larvae. Hypoxia caused significant changes in copper toxicity throughout embryonic development. Prior to hatching, hypoxia suppressed the occurrence of mortalities, but after hatching hypoxia significantly increased copper toxicity. Interestingly, when exposures were conducted only after hatching, the onset of copper-induced mortalities was delayed under hypoxia compared to normoxia, but after 48 h, copper was more toxic to hatched embryos under hypoxia. This is the second species for which the protective effect of hypoxia on copper toxicity prior to hatching, followed by its exacerbating effect after hatching is demonstrated, suggesting the hypothesis that this pattern may be common for teleost species. Our research highlights the importance of considering the interactions between multiple stressors, as understanding these interactions is essential to facilitate the accurate prediction of the consequences of exposure to complex stressors in a rapidly changing environment.

© 2016 Published by Elsevier Ltd.

1. Introduction

For most aquatic organisms, oxygen is essential for life but its levels in the environment can vary widely, and when dissolved oxygen concentrations drop below the levels required to sustain aerobic life (hypoxia), changes in ecosystem dynamics are likely to occur (Diaz and Breitburg, 2009). Areas of low oxygen can occur naturally in both freshwater and marine ecosystems, and evidence

* This paper has been recommended for acceptance by Prof. von Hippel Frank A.

* Corresponding author.

** Corresponding author.

E-mail addresses: jf277@exeter.ac.uk (J.A. Fitzgerald), e.santos@exeter.ac.uk (E.M. Santos).

http://dx.doi.org/10.1016/j.envpol.2016.12.008 0269-7491/© 2016 Published by Elsevier Ltd.

has shown that freshwater systems are more prone to hypoxia and anoxia (Jenny et al., 2016). In temperate freshwater lakes, for example, hypoxia can occur when factors affect vertical mixing, such as wind and temperature, or if a lake is covered by snow or ice preventing reaeration (Richards, 2009). In tropical freshwater systems, oxygen concentrations can vary depending on the amount of rainfall; during the dry season oxygen concentrations can drop as the water becomes stagnant, whereas rain promotes oxygenation of the water through increasing flow and mixing (Richards, 2009).

Hypoxia has increased rapidly in recent years and is a growing threat to aquatic ecosystems worldwide (Gewin, 2010). The causes of the increase in frequency, duration and severity of hypoxic events have been attributed to excessive anthropogenic nutrients entering water bodies (Gamenick et al., 1996), observed in both

marine (Assessment, 2005; Diaz and Rosenberg, 2008; Friedrich et al., 2014) and freshwater systems (Jenny et al., 2016; Richards, 2009). Excess nutrients in aquatic systems cause increased proliferation of algae at the surface and decomposition of organic materials in deep waters (eutrophication) resulting in oxygen depletion. Other factors can affect oxygen solubility, including changes in temperature, stratification of water bodies and metabolism of organisms, and contribute to the changing oxygen budget in the aquatic environment.

Periods of lowest oxygen concentrations often coincide with the presence of the highest contaminant concentrations (van der Geest et al., 2002), so it is fundamental to understand how hypoxia influences chemical toxicity. To date, studies investigating whether hypoxia modifies chemical toxicity in fish have considered endocrine disrupting chemicals (Brian et al., 2008), metals (Chun et al., 2000; Hattink et al., 2005, 2006; Heath, 1991; Lloyd, 1961; Kienle et al., 2008; Mustafa et al., 2012; Sampaio et al., 2008), poly-aromatic hydrocarbons (Prasch et al., 2004) and pharmaceuticals (Prokkola et al., 2015; Lubiana et al., 2016). Together, these studies suggest that the effects of hypoxia on chemical toxicity can vary widely as a function of the chemicals considered, the species and its life stage. This highlights a critical research need to understand how the effects of a stressor are influenced by the presence of other stressors, in order to accurately predict the consequences of exposure to combined stressors in the aquatic environment.

Among the contaminants in aquatic systems, copper is an important chemical of concern, due to its widespread distribution and high toxicity at concentrations found in some of the most polluted environments. Copper concentrations vary naturally in freshwater systems, however as a consequence of both contemporary and historical mining activities, industrial processes, and urban and agricultural runoff, many water bodies have become significantly contaminated (Batty et al., 2010; Ohmichi et al., 2006; Snook and Whitehead, 2004; Uren Webster et al., 2013). Copper was also identified as the metal causing the greatest threat to the aquatic environment in the UK, based on its current measured concentrations and toxic effects (Donnachie et al., 2014). It has been widely documented that decreased water hardness dramatically increased copper toxicity to fish, and copper bioavailability is usually higher in freshwater than seawater (Flemming and Trevors, 1989). Copper speciation and bioavailability in freshwaters may also be influenced by pH and the amount of natural organic matter (Erickson et al., 1996; De Schamphelaere and Janssen, 2004). As a result, copper is particularly toxic to freshwater fish species. Excess copper can cause toxicity by affecting osmoregulation (Chowdhury et al., 2016; Grosell and Wood, 2002; Lauren and McDonald, 1987; Wood et al., 2012), causing metabolic disruption as a result of inhibition of respiratory enzymes in the mitochondria, and compromising gas exchange by inducing gill damage (Beaumont et al., 2003; Chowdhury et al., 2016; McDonald and Wood, 1993). In addition, copper can also induce oxidative stress, via increased reactive oxygen species, resulting in DNA damage and lipid per-oxidation (Craig et al., 2007; Nawaz et al., 2006; Sanchez et al., 2005; Sevcikova et al., 2011). In fish, the most sensitive stages of development are thought to be the embryo-larval or early juvenile stages (Flemming and Trevors, 1989). Observed effects during early life in zebrafish (Danio rerio) included the reduction in yolk sac utilisation and decreased body length in exposed larvae, suggesting that copper may delay development and growth. In addition, exposure to copper caused a significant reduction in neuromasts in zebrafish embryos, potentially affecting their ability to survive due to lateral line dysfunction and behaviour impairment (Herneandez et al., 2006; Johnson et al., 2007; Linbo et al., 2006).

In our previous work we demonstrated that copper toxicity strongly decreased under low dissolved oxygen conditions during

the first 24 h of development in zebrafish embryos, but in contrast, hypoxia significantly increased copper toxicity after hatching (Fitzgerald et al., 2016). In the common carp (Cyprinus carpio), copper toxicity was increased under hypoxia compared to nor-moxia in adults (Malekpouri et al., 2016; Mustafa et al., 2012). Based on this evidence, we hypothesise that the combined effects of hypoxia and copper on fish are dependent on life stage, with critical switch points occurring during development. To test this, more studies are needed to elucidate the interactions between copper and hypoxia for a range of representative model species, in particular during development.

The three-spined stickleback (Gasterosteus aculeatus) was used as a model organism for this study because of its ecological relevance to temperate freshwater and marine ecosystems in the Northern hemisphere and its well established role as a model species for environmental toxicology (Katsiadaki et al., 2007). Studies investigating the responses of this species to hypoxia have reported behavioural effects (fewer aggressive acts (Sneddon and Yerbury, 2004) and loss of inquisitive and active behaviour (Leveelahti et al., 2011)), physiological responses (increased rate of gill movements (Jones, 1952)) and biochemical effects (increases in tissue L-lactate (Sneddon and Yerbury, 2004)). During development sticklebacks are deemed to be sensitive to low oxygen, a characteristic associated with the parental behaviour of adult males (Green and McCormick, 2005; Pollock et al., 2007). Male sticklebacks invest significant time and energy resources to fanning their embryo-bearing nest, resulting in an increased oxygen concentration around the embryos. Under hypoxic conditions increased fanning tempo was observed resulting in increased oxygenation of the nest (Reebs et al., 2007). However, the effects of hypoxia on embryo development have not been previously documented. This information is essential to facilitate the use of stickleback embryos as models to investigate the combined effects of hypoxia and chemical stressors.

The aim of this study was to investigate the effects of reduced oxygen concentrations, alone and combined with copper, on the stickleback throughout development. We first established the responses of stickleback to various concentrations of oxygen during development and determined the sensitivity of this species to hypoxia during early life. These data informed the design of experiments investigating whether mild hypoxia affects copper toxicity during development. To do this, embryos were exposed continuously throughout development until they reach the stage of free feeding larvae (which is 48 h after hatching at 19 °C), or only after hatching, to a range of concentrations of copper under hypoxia and normoxia. The proportion of dead and hatched embryos was recorded to produce concentration-response curves to determine if there were differences in copper toxicity when exposures were conducted under normoxia or hypoxia.

2. Material and methods

2.1. Fish source, culture and husbandry

A population of freshwater three-spined stickleback (originating from the River Erme, Devon, United Kingdom, kindly provided by the University of Plymouth) was maintained in the Aquatic Resource Centre at the University of Exeter in mixed sex stock tanks (120 L), supplied with aerated synthetic freshwater. Before it was supplied to each aquarium, mains tap water was filtered by reverse osmosis (Environmental Water Systems (UK) Ltd) and reconstituted with Analar-grade mineral salts to standardized synthetic freshwater (final concentrations resulting in a conductivity of 300 mS), aerated, and heated to 19 ± 1 °C in a reservoir. Throughout the experimental period all adults were maintained under summer

JA. Fitzgerald et al. / Environmental Pollution xxx (2016) 1—11

conditions which corresponded to a temperature of 19 ± 1 °C and a photoperiod of 18:6 h light/dark (with a 30 min dawn/dusk transition period). Fish were fed to satiation every day with blood worm (Chironomus sp.; Tropical Marine Centre, Chorleywood, UK).

The stickleback embryos included in this study were obtained via artificial fertilisation (method adapted from Barber and Arnott (2000)). Unfertilised eggs were identified by visual observation as described by Swarup (1958), using a dissection microscope (Motic DM143, Hong Kong) and removed. Fertilised embryos were incubated in aerated artificial freshwater (according to the ISO-7346/3 guideline, ISO water, diluted 1:5 (International Organization for Standardization, 1996); 11.76 g/L CaCl2-2H2O, 4.93 g/L MgSO47H2O, 2.59 g/L NaHCO3, 0.23 g/L KCl), in experimental exposure tanks, from 1hr post fertilisation, set up as described below. All fish were maintained under approved protocols, according to the UK Home Office regulations for the use of animals in scientific procedures.

2.2. Embryo exposures to decreased oxygen levels

Stickleback embryos were exposed to different levels of air saturation (AS) to determine the effects to hypoxia on development.

To produce water with the required percentage of AS, water was aerated to produce 100% AS, or bubbled with nitrogen for 1hr to remove all dissolved oxygen, then allowed to equilibrate to 19 ± 1 °C for at least 1 h before the start of the exposures. Water was then mixed at the appropriate proportion to obtain 100%, 80%, 60%, 40% and 20% AS.

Exposures were initiated at 1 h post fertilisation (hpf; corresponding to the one cell stage; blastodermic cap), when 20 fertilised eggs were randomly allocated to 600 ml acid-washed glass tanks and sealed with a glass lid, to prevent re-oxygenation, and all treatments were run in triplicate. To avoid changes in the water characteristics by the metabolic activity of the embryos, a large volume of water (30 ml per embryo) was used. Within this experimental setup, water pH was constant over time and it was not affected by the exposure condition, as described in Fitzgerald et al., 2016. After 24 h of exposure, the percentage of AS was measured in each exposure tank using an optical dissolved oxygen meter (Mettler SevenGo Pro OptiOx, U.S.) and was found to be within 5% of the nominal AS (measured AS were 96.5 ± 0.3%, 75.0 ± 0.5%, 57.6 ± 0.6%, 40.9 ± 0.5% and 24.7 ± 0.9%, for the 100%, 80%, 60%, 40% and 20% treatments, respectively). Individual embryos were observed using a dissection microscope (Motic DM143, Hong Kong) and the proportion of mortalities, hatched embryos, the stage of development and any developmental abnormalities were recorded. From each tank, 1 embryo was removed, anesthetized using tricane (Sigma Aldrich) and mounted on a slide using methylcellulose (2%; Sigma Aldrich) for imaging. Photographs were taken using a compound microscope (Nikon SM21500, Japan) equipped with a digital camera, to allow visualisation of any developmental effects resulting from the exposure to low oxygen concentrations. A complete water change was carried out at each 24 h period, as described above, and the experiment was maintained until 217 hpf. Up until this time point under our experimental conditions, sticklebacks have yolk sacs and as such are not considered to be free feeding; this has regulatory implications as under UK legislation, fish embryos become protected under the Animals (Scientific Procedures) Act from the free feeding stage onwards.

2.3. Copper exposures under normoxia and hypoxia

Embryos were continuously exposed to a range of copper concentrations (copper sulphate; Fisher, Fair Lawn NJ), 0, 0.015, 0.025,

0.05, 0.0625, 0.075, 0.0875, 0.1, 0.125, 0.15 mg Cu/L, throughout development (1-217 hpf). The range of concentrations was selected to encompass concentrations causing 0% and 100% mortalities, based on preliminary experiments. The concentrations of copper in the exposure water were determined for a selection of exposure tanks during the preliminary experiments by ICP-MS (as described in Fitzgerald et al., 2016). The measured concentrations were between 88.9% and 104% of the nominal concentrations throughout the range of concentrations tested. Based on the experiments described above, 50% AS was selected as the nominal level of AS for the combined exposures, because this was the lowest level of AS that did not result in observable developmental delays. The measured levels of AS were 97.6 ± 0.1% and 56.1 ± 0.2% for exposures conducted under normoxia and hypoxia, respectively. Throughout this paper, in the context of our experiments, we refer to 56% AS as hypoxia, and this corresponds to a level of air saturation that causes physiological changes without causing alterations in morphology or developmental delays. For each treatment, 20 embryos were randomly allocated to 600 ml acid washed glass tanks, with water prepared as above, containing the appropriate concentration of copper and at the appropriate AS. After each 24 h exposure period, the percentage of AS was immediately measured and the proportion of mortalities, hatched embryos, the stage of development and any developmental abnormalities were recorded for each exposure tank. Water was then completely replaced with freshly made exposure water at the appropriate AS and copper concentration, as described above. All exposures were conducted in triplicate.

2.4. Copper exposures to hatched embryos under normoxia and hypoxia

Additional exposures were conducted during the post hatch embryo stage (169-217 hpf). Embryos were randomly allocated to exposure tanks as described above and incubated under control conditions (97.6 ± 0.1% AS, 0 mg Cu/L) up to the start of the exposure period (169 hpf). Embryos were then exposed to a range of copper concentrations (from 0 to 0.15 mg/L) under normoxia and hypoxia (97.6 ± 0.1% and 56.1 ± 0.2%, respectively), as described above. Similarly to the continuous exposure, oxygen levels were measured and mortality and developmental delays monitored before water was replaced after 24 h of exposure, and were maintained until 48 h post exposure. All combined exposures to copper and hypoxia were conducted in triplicate.

2.5. Statistical analysis

Statistical analyses were conducted in R (Team, 2014). Differences between the proportion of mortality and hatching following exposure to copper under hypoxia or normoxia were identified using generalized linear models. A separate model was used for each time period after fertilisation, using a quasibinomial error structure and logit link to test for effects of copper concentration on the proportion of mortality (as a continuous variable), hypoxia or normoxia (as a categorical variable) and the interaction between the two. Minimum adequate models were derived by model simplification using F tests based on analysis of deviance (Crawley, 2012). A similar approach of model simplification of generalized linear models with quasibinomial error structure was used to test for the effects of the exposures on the proportion of hatched embryos following exposure to copper under hypoxia or normoxia and their interaction. F tests reported refer to the significance of removing terms from the models. All data was considered statistically significant when P < 0.05.

J.A. Fitzgerald et al. / Environmental Pollution xxx (2016) 1—11

3. Results

3.1. Effects of different oxygen concentrations on stickleback development

Embryos exposed to the lowest oxygen concentration (24.7± 0.9% AS) showed the strongest adverse response (Fig. 1; Table 1), including increased mortality (Table 1). In addition, the proportion of hatched embryos appeared to be reduced (despite the high variability) and developmental delays occurred at this level of AS (Fig. 1; Table 1), with embryos observed to be smaller and having less developed pigmentation compared to the controls (Fig. 1). Delays in development were also observed at 40% AS, with embryos appearing to be smaller than those in the 100% control group, as seen in the photographs of Fig. 1, although no significant effects on hatching or mortalities were observed at this oxygen concentration (Fig. 1; Table 1). For concentrations of oxygen above 60% AS, no developmental delays were observed in our study (Fig. 1; Table 1). However embryos exposed to 80% AS (measured: 75.0 ± 0.5% AS), hatched on average 24 h earlier than the controls (Fig. 1; Table 1).

3.2. Effects of hypoxia on copper toxicity throughout development

Copper caused mortalities to stickleback embryos under both hypoxic and normoxic conditions after 49 hpf, and the proportion of mortalities increased progressively throughout the exposure period, (Fig. 2). The effects of hypoxia on copper toxicity were dependent on the time point analysed. For the first 48 h of exposure, there was no significant difference between the two different AS treatments (P = 0.0609; Fig. 2). From 73 to 145 hpf (48-144 h of exposure), copper was more toxic to embryos under normoxia than under hypoxia (P < 0.01; Fig. 2). During the period when hatching occurs (from 145 to 193 hpf), there was no significant difference in copper-induced mortalities under hypoxia compared to normoxia (1-169 hpf - P = 0.1554,1-193 hpf - P = 0.3591; Fig. 2). In contrast,

during the final 24 h of exposure, copper caused significantly more mortalities to stickleback embryos under hypoxia compared to normoxia (P < 0.01; Fig. 2).

Copper reduced the proportion of embryos that hatched by169 hpf (P < 0.001, Fig. 3a), 193 hpf (P < 0.001, Fig. 3b) and 216 hpf (P < 0.001, Fig. 3c). Hypoxia had no effect on the proportion of embryos that had hatched following exposure to copper at 169 hpf (P = 0.411; Fig. 3a) and 193 hpf (P = 0.730; Fig. 3b). However, by 217 hpf, the reduction in the proportion of hatched embryos caused by copper exposure was more severe under normoxia compared to hypoxia (P = 0.0305; Fig. 3c).

3.3. Effects of hypoxia on copper toxicity to hatched embryos

Copper was significantly more toxic to hatched embryos under normoxia than under hypoxia during the first 24 h of exposure (P < 0.01; Fig. 4). Maintaining the exposure for a further 24 h resulted in a dramatic change, with copper significantly more toxic to hatched embryos under hypoxia than under normoxia (P < 0.05; Fig. 4).

4. Discussion

4.1. Impacts of hypoxia on stickleback development

Hypoxia caused a range of developmental effects on stickleback embryos, with their severity proportional to the levels of oxygen depletion. The stickleback population used in this study was found to be comparatively sensitive to hypoxia when compared with some cyprinid species (crucian carp (Nilsson and Renshaw, 2004) and zebrafish (Padilla and Roth, 2001)), or the annual killifish (Austrofundulus limnaeus (Podrabsky et al., 2007)), which have been shown to tolerate low concentrations of oxygen and even anoxia during development or as adult fish, but greater than the more sensitive teleost species, including salmonids (Bickler and Buck, 2007).

Time of development (hpf) 97 121 145

o «

# # — <1!LI

'¿J /3 * \ e* ) r « - - 1 » 1

O "H

1 V jf 0 # «Mb ID

Scale Bar: — 1mm

Fig. 1. Stickleback embryos exposed to different air saturations (AS) throughout development. Video-captured images (in a Nikon SM21500, Japan) of embryos from 1 to 217 hpf, exposed to 100%, 80%, 60%, 40% and 20% AS (measured average AS: 96.5 ± 0.3%, 75.0 ± 0.5%, 57.6 ± 0.6%, 40.9 ± 0.5% and 24.7 ± 0.9%, respectively). Developmental delays were observed in embryos exposed to 20% and 40% AS, with embryos appearing to be smaller and with less pigmentation than the controls. Hatching occurred 24 h earlier in the 80% AS treatment compared to the control.

Table 1

Proportion of embryo survival and hatching (±SEM) following exposure of stickleback embryos to a range of air saturation (AS) levels during development.

AS (%) Hours post fertilisation (hpf)

25 49 73 97 121 145 169 193 217

a) Mortality

100 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00)

80 0.00 (±0.00) 0.01 (±0.01) 0.01 (±0.01) 0.01 (±0.01) 0.01 (±0.01) 0.01 (±0.01) 0.01 (±0.01) 0.01 (±0.01) 0.01 (±0.01)

60 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00)

40 0.00 (±0.00) 0.00 (±0.00) 0.00 (±0.00) 0.01 (±0.01) 0.01 (±0.01) 0.02 (±0.02) 0.04 (±0.02) 0.04 (±0.02) 0.04 (±0.02)

20 0.00 (±0.00) 0.02 (±0.02) 0.02 (±0.02) 0.04 (±0.02) 0.05 (±0.02) 0.06 (±0.01) 0.17 (±0.02) 0.24 (±0.04) 0.39 (±0.04)

b) Hatching

100 - - — — — 0.08 (±0.04) 1.00 (±0.00) 1.00 (±0.00) 1.00 (±0.00)

80 - - — — — 0.96 (±0.03) 1.00 (±0.00) 1.00 (±0.00) 1.00 (±0.00)

60 - - — — — 0.03 (±0.03) 1.00 (±0.00) 1.00 (±0.00) 1.00 (±0.00)

40 - — — — — 0.01 (±0.01) 0.92 (±0.03) 1.00 (±0.00) 1.00 (±0.00)

20 - — — — — 0.01 (±0.01) 0.05 (±0.03) 0.28 (±0.15) 0.47 (±0.22)

For the rainbow trout (Oncorhynchus mykiss), developmental delays and mortalities where observed when embryos were exposed to 40% AS and lower (Roussel, 2007), a similar result to what was observed in our experiments for the stickleback, which showed developmental delays at 40% AS or below, but mortalities only for 20% AS. In the environment, hypoxia-sensitive species such as the rainbow trout and the stickleback have adopted ecological strategies to avoid low oxygen conditions during development, reducing the risk of adverse effects when developing in hypoxic environments. Rainbow trout spawn in fast flowing, well oxygenated rivers and streams, resulting in constant high oxygen concentrations in the water surrounding the embryos (Ciuhandu et al., 2007). Stickleback, however, are typically found in still or slow flowing water of lakes, ponds, lowland streams and sheltered coastal bays, associated with their slow swimming behaviour (Wootton, 1984). To avoid the costs of low oxygen on development, stickleback males have evolved behaviours associated with parental care, and nurture embryos as they are incubated in nests throughout development. The male commits large amounts of time and energy to maintaining the nest, including vigorously fanning the nest and removing decaying embryos (Van Iersel, 1953; Wootton, 1984). Lipid and glycogen levels have been shown to fall by 43% and 37%, respectively, during the reproductive season in breeding males, and the protein concentration in the carcass of breeding males was shown to fall by 70% compared to pre-breeding levels (Chellappa et al., 1989), demonstrating the high energetic cost of parental care in breading males during the reproductive season. Male sticklebacks trade-off time and energy to ensure embryo survival and avoid the reduced growth and developmental delays which would result in a longer incubation time, reducing the predation risk and ensuring the fitness and survival of the embryos (Ostlund-Nilsson et al., 2006).

No developmental delays were observed in embryos exposed to concentrations of oxygen of 60% AS and above. Oxygen concentrations in stickleback nests were shown to vary depending on whether the male is attending the nest or not. When parental fanning occurred, AS levels were reported to be 95—100% AS, but dropped to a minimum of 67% when the male abandoned the nest (Green and McCormick, 2005). This corresponds to the AS range where no developmental delays were observed in our study, indicating that for AS levels occurring naturally within nests, normal development would occur.

For embryos exposed to 80% AS, embryos hatched approximately 24 h earlier than the controls. The metabolic rate of the embryo is known to increase throughout development, causing an increase in their oxygen demands (Green and McCormick, 2005). For the 80% AS treatment there was no sign of either delayed or accelerated development, suggesting they were at the same stage

of development and in turn required the same oxygen concentrations as the control, despite less environmental oxygen available. Premature hatching may, therefore, have occurred to compensate for the insufficient diffusion of oxygen to the embryonic cells, as previously seen for the Atlantic salmon (Salmo salar) and the rainbow trout (Hamor and Garside, 1976; Wu, 2009). After embryos have hatched, they are able to move and seek oxygen rich areas, and gas exchange can occur more effectively via the skin, potentially explaining the advantage of hatching prematurely when exposed to reduced oxygen.

4.2. Combined effects of copper and hypoxia

The effects of hypoxia on copper toxicity were strongly dependent on the stage of development with contrasting effects observed prior to and after hatching. During early development, hypoxia significantly suppressed copper toxicity, but during the hatching period this effect was no longer present and hypoxia did not alter copper toxicity. To our knowledge, there is only one other dataset investigating the effects of hypoxia on copper toxicity during embryonic development in fish. In that study, hypoxia strongly modified copper toxicity to zebrafish embryos, in a process dependent on developmental stage, and hypoxia protected embryos from copper toxicity during early development (Fitzgerald et al., 2016). In contrast to the stickleback where copper at the concentrations tested had no toxic effect until 49 hpf, in zebrafish the majority of mortalities occurred during the first 24 h of exposure (Fitzgerald et al., 2016). The differences in tolerance between these species during early development may have been as a result of the chorion in sticklebacks providing a greater barrier to copper in comparison to the zebrafish (Hartmann and Englert, 2012; Hosemann et al., 2004), as this would likely affect the amount of copper reaching embryonic tissues.

In contrast to what occurred during early development, after hatching hypoxia increased copper toxicity to stickleback embryos. This is similar to that observed for the zebrafish, where copper toxicity was greater under hypoxia after hatching (Fitzgerald et al., 2016). This suggests that independently of the species and its tolerance to hypoxia, after hatching hypoxia results in an increase in copper toxicity. This is supported by the results of exposures conducted in adult fish, which reported increased copper toxicity under hypoxia. For example, Mustafa et al. (2012) showed that carp were more sensitive to copper under hypoxia compared to nor-moxia, as combined exposures induced a significantly higher level of oxidative DNA damage. The greater metabolic demands of embryos after hatching compared to early stages of embryogenesis may also explain some of the differences observed. This is supported by observations in the common carp where hypoxia

J.A. Fitzgerald et al. / Environmental Pollution xxx (2016) 1—11

Fig. 2. Embryo mortality curves following continuous exposure to copper under normoxia or hypoxia throughout development. Each point on the graph represents the proportion of mortality in an individual replicate tank containing 20 embryos, black and white symbols represent groups exposed to copper under normoxia (97.6% ± 0.1 air saturation (AS)) or hypoxia (56.1% ± 0.2 AS), respectively, and the lines represent the best fit model for the data, calculated using generalized linear models in R (model output summarised in Table S1a). From 1 to 25 hpf, there was no significant effect of AS or copper on the mortality (P = 0.31). For all other time periods, copper caused an increase in mortality both under hypoxia and normoxia (P < 0.001). There was a significant increase in copper-induced mortality under normoxia compared to under hypoxia from 73 to 145 hpf (1-73 hpf, P < 0.01; 1-97 hpf, P < 0.01; 1-121 hpf, P < 0.001; 1-145 hpf, P < 0.05). For the 1—49,1—169 and 1-193 hpf periods, there was no significant difference in mortalities between the embryos exposed to copper at different AS. Copper toxicity was significantly greater under hypoxia compared to normoxia for the final time stage, 1-217 hpf (P < 0.01).

exacerbated the impacts of copper on its metabolic capabilities, compared to that observed in fish exposed to copper alone (Malekpouri et al., 2016). Furthermore, the similar trends in toxicity between hatched embryos and adult fish may be explained by the mode of uptake, which in both cases involve direct contact of fish with copper contaminated water and uptake via the skin and/or gills (Grosell and Wood, 2002; van Heerden et al., 2004; Zimmer et al., 2012, 2014). In contrast, prior to hatching, embryos are

protected by the chorion, which will likely modify the exposure dynamics and how much copper reaches the embryonic tissues. In the zebrafish, copper was shown to bind to the chorion of embryos (Fitzgerald et al., 2016), potentially preventing copper from reaching the embryonic cells, supporting the hypothesis that the presence of the chorion may play a role in copper bioavailability and contribute to the differences observed before and after hatching.

The hypoxia-induced changes in copper toxicity before and after

n' e k a

a -a c re

m op 3 re

0 d 3 J

1 O u °

o (re 6 f

■a y o

.00 vv

o re o

Fig. 3. Proportion of hatching in embryos exposed to copper under normoxic or hypoxic conditions, for the time stages; 145—169,145—193 and 145-217 hpf. Each point on the graph represents the proportion of hatched embryos in an individual tank containing 20 embryos. White and black symbols represent hypoxic (56.1% ± 0.2 air saturation (AS)) or normoxic (97.6% ± 0.1 AS) treatments, respectively, and the lines represent the best fit model for the data, calculated using a generalized linear model in R (model output summarised in Table S2). There was a significant effect of copper on the proportion of embryos that had hatched under normoxia and hypoxia for all time periods (P < 0.001). For 145—169 and 145-193 hpf, there was no significant difference in the proportion of hatched embryos exposed to copper under normoxia compared to hypoxia (145-169 hpf, P = 0.411; 145-193 hpf, P = 0.730). For 145217 hpf, there was a significant increase in the proportion of hatched embryos exposed to copper under normoxia compared to hypoxia (P = 0.0305).

A - 169-193hpf B - 169-217hpf

n-1-1-n n-1-1-n

0.00 0.05 0.10 0.15 0.00 0.05 0.10 0.15

Cu Concentration (mg/L) Cu Concentration (mg/L)

Fig. 4. Cumulative mortality curves following exposure to copper under normoxia or hypoxia in hatched embryos, A) 24 h after exposure (169-193 hpf) and B) 48 h after exposure (169-217 hpf). Each point on the graph represents the proportion of mortality in an individual replicate tank containing 20 embryos, black and white symbols represent groups exposed to copper under normoxia (97.6% ± 0.1 air saturation (AS)) or hypoxia (56.1% ± 0.2 AS), respectively, and the lines represent the best fit model for the data, calculated using generalized linear models in R (model output summarised in Table S1b). Copper caused an increase in mortality both under hypoxia and normoxia over the whole exposure period (P < 0.001). There was a significantly greater mortality under normoxia compared to under hypoxia, after the first 24 h of copper exposure (P < 0.001). After 48 h of exposure, copper toxicity was significantly greater under hypoxia compared to normoxia (P < 0.001).

hatching may also be due to different molecular responses to hypoxia in early embryos compared to later life stages. Around hatching there is a change in the expression, at the transcript and protein levels, of the hypoxia inducible factor (HIF) alpha molecules in embryos (Koblitz et al., 2015; Rytkonen et al., 2013, 2014), with consequent shifts in the molecular and physiological responses to hypoxia. These changes may result in alterations in the physiological processes regulated by HIF, including red blood cell formation and oxygen transport, changes in energy metabolism, suppression of cell growth and proliferation, and oxidative stress response (Lushchak, 2011; Nikinmaa and Rees, 2005; Richards, 2009). It has been shown in the zebrafish that hatching is a critical transition period for the transcription and protein expression of the HIF alpha molecules (Kopp et al., 2011; Rytkonen et al., 2014), and it is possible that similar changes in the dynamics of which HIF isoforms are expressed, and the physiological changes they regulate, also take place in the three spined stickleback, potentially explaining the opposing effects of hypoxia on copper toxicity as development progresses. Further work, including studies in other fish species, is essential for confirmation of these hypotheses and to document the mechanistic basis of the shift in the effects of hypoxia on copper toxicity, from a protective effect during early development to an increase in toxicity in hatched embryos.

Copper significantly reduced the proportion of hatched embryos both under hypoxia and normoxia and this may be as a result of copper disrupting the activity of chorionase, an enzyme involved in the process of hatching (Dave and Xiu, 1991; Jezierska et al., 2009; Johnson et al., 2007). Alternatively, this could also occur as a result of the metabolic effects of copper, which may have reduced ATP production and scope for growth and development (Johnson et al., 2007; Waser et al., 2010). There was no effect of AS level on the proportion of embryos that had hatched following exposure to copper up to 169 or 193 hpf, however by 217 hpf, the reduction of the proportion of hatched embryos was greater under normoxia compared to hypoxia. At this time point (217 hpf), copper also

caused greater proportion of mortality under hypoxia compared to normoxia, therefore the increased proportion of embryos that had hatched in exposures conducted under hypoxia could be associated with the observed increase in copper toxicity.

4.3. Effects of hypoxia on copper toxicity to hatched embryos

The concentration of oxygen in freshwater systems can vary rapidly over time in the course of days due to changes in temperature, precipitation and water mixing. It is therefore important, and environmentally relevant, to consider the effects of hypoxia on chemical toxicity during short developmental windows. Post-hatch embryos are considered to be the most sensitive stage of development for chemical exposures in sticklebacks (Ostlund-Nilsson et al., 2006), marking this developmental stage critically sensitive to the combined effects of copper and hypoxia. Therefore, we investigated the effect of hypoxia on copper toxicity during this life stage, in the absence of pre-exposure to these stressors.

For hatched embryos, during the first 24 h of exposure, copper was significantly more toxic under normoxia compared to hypoxia, but by 48 h copper was significantly more toxic under hypoxia than under normoxia. The data for this later time point concurs with what was observed for embryos exposed continuously throughout development and discussed above, with hypoxia increasing copper toxicity after hatching. We hypothesise that hypoxia may delay the uptake of copper into the embryos resulting in a later onset of toxicity, and/or that the cellular responses to hypoxia may initially protect from the toxic effects of copper, for example by the activation of oxidative stress defence mechanisms, a known effect of the activation of the HIF signalling pathway (Ransberry et al., 2016). Further work will be required to test these hypotheses, including by measuring the copper uptake in embryos exposed over a time series and investigating the molecular mechanisms of response to combinations of copper and hypoxia.

Mortality was greater when embryos were exposed only after

JA. Fitzgerald et al. / Environmental Pollution xxx (2016) 1—11

hatching compared to that measured at the same developmental stage for embryos exposed continuously throughout development. For example, for the continuous exposure throughout development, 0.1 mg Cu/L caused 68% and 87% mortalities under normoxia and hypoxia, respectively (Fig. 3), while when embryos were exposed only after hatching the same copper concentration resulted in 97% and 100% mortality under normoxia and hypoxia, respectively (Fig. 4). This may be as a result of the embryos in the continuous exposure having a period where acclimation to copper and hypoxia may have occurred. Although comparable results for developmental exposures in fish have not been previously reported, evidence exists for a protective effect of pre-exposure on subsequent tolerance to copper in other organisms. Herkovits and Perez-Coll (2007) showed that pre-exposure of copper resulted in a transient beneficial effect on embryo survival when re-exposed to the same metal in Bufo arenarum (Herkovits and Perez-Coll, 2007). Similarly, pre-exposure to arsenic decreased its toxicity during subsequent exposures in killifish (Shaw et al., 2007). This may be as a result of the activation of protective response mechanisms, including increased expression of metal binding proteins (such as metallothionein) or oxidative stress response mechanisms. Furthermore, pre-conditioning to hypoxia has also been shown to increase tolerance during subsequent exposures in a number of organisms, including in the zebrafish during embryogenesis (Manchenkov et al., 2015).

It is more likely that hatched embryos will experience hypoxia compared to those prior to hatching. Following hatching, stickleback males reduce fanning of the nest possibly because the presence of a current caused by fanning would add a high metabolic stress on the free swimming embryos (Garenc et al., 1999), reducing the flow of oxygenised water. If exposed to copper, the likelihood of the embryo surviving would be reduced as hatched embryos were most sensitive to the combined effects of copper and hypoxia during this developmental stage, both when exposures were conducted only after hatching and for exposures initiated immediately after fertilisation. Therefore, hatched embryos are a critical developmental stage at risk from the adverse effects of exposure to combinations of copper and hypoxia in the natural environment.

5. Conclusions

We report evidence for significant effects of hypoxia on copper toxicity during embryonic development for a teleost species. Similarly to that reported for the zebrafish, during the early stages of development, hypoxia protected embryos from copper toxicity, but in contrast, after hatching, hypoxia increased copper toxicity in hatched embryos. We suggest that this pattern of interaction between copper and hypoxia during development may be common across teleost species, and further data across a range of families of fish will be required to verify this hypothesis. Considering the significant increase in the incidence, prevalence and severity of hypoxic events in both marine and freshwater systems worldwide, the likelihood of aquatic organisms being exposed to hypoxia during development is high. Therefore, taking into account the combination of chemical stressors and hypoxia is important to generate environmentally relevant information on chemical effects on aquatic organisms, to support appropriate regulatory and management decisions. For example, our data demonstrate that the evaluation of the toxicity of either copper or hypoxia alone would have under-estimated the effects of these stressors when they occur in combination, for hatched embryos. It is not possible to study in detail the many thousands of chemicals that enter the aquatic environment, so intelligent testing strategies are needed to document the interactions between hypoxia and groups of chemicals with the same mode of action. This will facilitate the

development of predictive models to better inform risk assessment and environmental management and move towards a framework that is effective in protecting living organisms in environments affected by complex mixtures of stressors.

Funding

This work was supported by a PhD studentship supporting J. A. Fitzgerald, funded by the University of Exeter and the Centre for Environment Fisheries and Aquaculture Science under their Strategic Alliance.

Acknowledgements

We thank Jan Shears, Gregory Paull and the Aquatic Resources Centre technical team for support with stickleback husbandry at the University of Exeter, and Matthew Sanders for support with embryo experiments at Cefas Weymouth.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.envpol.2016.12.008.

References

Assessment, M.E., 2005. Ecosystems and human well-being: wetlands and water.

World Resources Institute, Washington, DC, p. 5. Barber, I., Arnott, S.A., 2000. Split-clutch IVF: a technique to examine indirect fitness consequences of mate preferences in sticklebacks. Behaviour 137,1129-1140. Batty, L.C., Auladell, M., Sadler, J., Hallberg, K., 2010. The impacts of metalliferous drainage on aquatic communities in streams and rivers, ecology of industrial pollution. Cambridge Univeristy Press, Cambridge, pp. 70-100. Beaumont, M., Butler, P., Taylor, E., 2003. Exposure of brown trout Salmo trutta to a sublethal concentration of copper in soft acidic water: effects upon gas exchange and ammonia accumulation. J. Exp. Biol. 206, 153-162. Bickler, P.E., Buck, L.T., 2007. Hypoxia tolerance in reptiles, amphibians, and fishes:

life with variable oxygen availability. Annu. Rev. Physiol. 69, 145-170. Brian, J.V., Beresford, N., Walker, J., Pojana, G., Fantinati, A., Marcomini, A., Sumpter, J.P., 2008. Hypoxia does not influence the response of fish to a mixture of estrogenic chemicals. Environ. Sci. Technol. 43, 214-218. Chellappa, S., Huntingford, F., Strang, R., Thomson, R., 1989. Annual variation in energy reserves in male three-spined stickleback, Gasterosteus aculeatm L. (Pisces, Gasterosteidae). J. Fish Biol. 35, 275-286. Chowdhury, M.J., Girgis, M., Wood, C.M., 2016. Revisiting the mechanisms of copper toxicity to rainbow trout: time course, influence of calcium, unidirectional Na+ fluxes, and branchial Na+, K+ ATPase and V-type H+ ATPase activities. Aquat. Toxicol. 177, 51-62.

Chun, Y.-S., Choi, E., Kim, G.T., Choi, H., Kim, C.H., Lee, M.J., Kim, M.-S., Park, J.W., 2000. Cadmium blocks hypoxia-inducible factor (HIF)-1-mediated response to hypoxia by stimulating the proteasome-dependent degradation of HIF-1a. Eur. J. Biochem. 267, 4198-4204. Ciuhandu, C.S., Wright, P.A., Goldberg, J.I., Stevens, E.D., 2007. Parameters influencing the dissolved oxygen in the boundary layer of rainbow trout (Onco-rhynchus mykiss) embryos and larvae. J. Exp. Biol. 210, 1435-1445. Craig, P.M., Wood, C.M., McClelland, G.B., 2007. Oxidative stress response and gene expression with acute copper exposure in zebrafish (Danio rerio). American Journal of Physiology-Regulatory. Integr. Comp. Physiol. 293, R1882-R1892. Crawley, M.J., 2012. The R Book. John Wiley & Sons.

Dave, G., Xiu, R., 1991. Toxicity of mercury, copper, nickel, lead, and cobalt to embryos and larvae of zebrafish, Brachydanio rerio. Arch. Environ. Contam. Toxicol. 21,126-134.

Diaz, R.J., Breitburg, D.L., 2009. Chapter 1 the hypoxic environment. In: Jeffrey, G.,

Richards, A.P.F., Colin, J.B. (Eds.), Fish Physiology. Academic Press, pp. 1-23. Diaz, R.J., Rosenberg, R., 2008. Spreading dead zones and consequences for marine

ecosystems. Science 321, 926-929. Donnachie, R.L., Johnson, A.C., Moeckel, C., Pereira, M.G., Sumpter, J.P., 2014. Using risk-ranking of metals to identify which poses the greatest threat to freshwater organisms in the UK. Environ. Pollut. 194, 17-23. De Schamphelaere, K.A., Janssen, C.R., 2004. Effects of dissolved organic carbon concentration and source, pH, and water hardness on chronic toxicity of copper to Daphnia magna. Environ. Toxicol. Chem. 23, 1115-1122. Erickson, R.J., Benoit, D.A., Mattson, V.R., Leonard, E.N., Nelson, H.P., 1996. The effects of water chemistry on the toxicity of copper to fathead minnows. Environ. Toxicol. Chem. 15, 181-193. Fitzgerald, J.A., Jameson, H.M., Dewar Fowler, V.H., Bond, G.L., Bickley, L.K., Uren Webster, T.M., Bury, N.R., Wilson, R.J., Santos, E.M., 2016. Hypoxia suppressed

copper toxicity during early development in zebrafish embryos in a process mediated by the activation of the HIF signaling pathway. Environ. Sci. Technol. 50 (8), 4502-4512.

Flemming, C., Trevors, J., 1989. Copper toxicity and chemistry in the environment: a review. Water, Air, Soil Pollut. 44,143-158.

Friedrich, J., Janssen, F., Aleynik, D., Bange, H.W., Boltacheva, N., ^agatay, M., Dale, A.W., Etiope, G., Erdem, Z., Geraga, M., 2014. Investigating hypoxia in aquatic environments: diverse approaches to addressing a complex phenomenon. Biogeosciences 10,12655-12772.

Gamenick, I., Jahn, A., Vopel, K., Giere, O., 1996. Hypoxia and sulphide as structuring factors in a macrozoobenthic community on the Baltic Sea shore: colonisation studies and tolerance experiments. Mar. Ecol. Prog. Ser. 144, 73-85.

Garenc, C., Couture, P., Laflamme, M.A., Guderley, H., 1999. Metabolic correlates of burst swimming capacity of juvenile and adult threespine stickleback (Gaster-osteus aculeatus). J. Comp. Physiol. B 169,113-122.

Gewin, V., 2010. Dead in the water. Nature 466, 812-814.

Green, B.S., McCormick, M.I., 2005. O2 replenishment to fish nests: males adjust brood care to ambient conditions and brood development. Behav. Ecol. 16, 389-39 .

Grosell, M., Wood, C.M., 2002. Copper uptake across rainbow trout gills mechanisms of apical entry. J. Exp. Biol. 205,1179-1188.

Hamor, T., Garside, E., 1976. Developmental rates of embryos of Atlantic salmon, Salmo salar L., in response to various levels of temperature, dissolved oxygen, and water exchange. Can. J. Zool. 54,1912-1917.

Hartmann, N., Englert, C., 2012. A microinjection protocol for the generation of transgenic killifish (Species: Nothobranchius furzeri). Dev. Dyn. 241,1133-1141.

Hattink, J., De Boeck, G., Blust, R., 2005. The toxicokinetics of cadmium in carp under normoxic and hypoxic conditions. Aquat. Toxicol. 75,1-15.

Hattink, J., De Boeck, G., Blust, R., 2006. Toxicity, accumulation and retention of zinc by carp under normoxic and hypoxic conditions. Environ. Toxicol. Chem. 25, 87-96.

Heath, A.G., 1991. Effect of water-borne copper on physiological responses of bluegill (Lepomis macrochirus) to acute hypoxic stress and subsequent recovery. Comp. Biochem. Physiol. Part C Comp. Pharmacol. 100, 559-564.

Herkovits, J., Perez-Coll, C.S., 2007. Acclimation to low level exposure of copper in Bufo arenarum embryos: linkage of effects to tissue residues. Int. J. Environ. Res. Public Health 4, 166-172.

Hernandez, P.P., Moreno, V., Olivari, F.A., Allende, M.L., 2006. Sub-lethal concentrations of waterborne copper are toxic to lateral line neuromasts in zebrafish (Danio rerio). Hear. Res. 213,1-10.

Hosemann, K.E., Colosimo, P.F., Summers, B.R., Kingsley, D.M., 2004. A simple and efficient microinjection protocol for making transgenic sticklebacks. Behaviour 141,1345-1355.

International Organization for Standardization, 1996. ISO, Water quality determination of the acute lethal toxicity of substances to a freshwater fish [Brachy-danio rerio Hamilton-Buchanan (Teleostei, Cyprinidae)]. Flow-Through Method (Part 3).

Jenny, J.P., Francus, P., Normandeau, A., Lapointe, F., Perga, M.E., Ojala, A., Schimmelmann, A., Zolitschka, B., 2016. Global spread of hypoxia in freshwater ecosystems during the last three centuries is caused by rising local human pressure. Glob. Change Biol. 22,1481-1489.

Jezierska, B., Lugowska, K., Witeska, M., 2009. The effects of heavy metals on embryonic development of fish (a review). Fish Physiol. Biochem. 35, 625-640.

Johnson, A., Carew, E., Sloman, K., 2007. The effects of copper on the morphological and functional development of zebrafish embryos. Aquat. Toxicol. 84,431-438.

Jones, J.E., 1952. The reactions of fish to water of low oxygen concentration. J. Exp. Biol. 29, 403-415.

Katsiadaki, I., Sanders, M., Sebire, M., Nagae, M., Soyano, K., Scott, A.P., 2007. Three-spined stickleback: an emerging model in environmental endocrine disruption. Environ. Sci. 14, 263-283.

Kienle, C., Köhler, H.-R., Filser, J., Gerhardt, A., 2008. Effects of nickel chloride and oxygen depletion on behaviour and vitality of zebrafish (Danio rerio, Hamilton, 1822)(Pisces, Cypriniformes) embryos and larvae. Environ. Pollut. 152, 612-620.

Koöblitz, L., Fiechtner, B., Baus, K., Lussnig, R., Pelster, B., 2015. Developmental expression and hypoxic induction of hypoxia inducible transcription factors in the zebrafish. PloS One 10, e0128938.

Kopp, R., Köblitz, L., Egg, M., Pelster, B., 2011. HIF signaling and overall gene expression changes during hypoxia and prolonged exercise differ considerably. Physiol. Genomics 43, 506-516.

Lauren, D.J., McDonald, D., 1987. Acclimation to copper by rainbow trout, Salmo gairdneri: physiology. Can. J. Fish. Aquat. Sci. 44, 99-104.

Leveelahti, Leskinen, P., Leder, E.H., Waser, W., Nikinmaa, M., 2011. Responses of threespine stickleback (Gasterosteus aculeatus, L) transciptome to hypoxia. Comp. Biochem. Physiol. Part D Genomics Proteom. 6, 370-381.

Linbo, T.L., Stehr, C.M., Incardona, J.P., Scholz, N.L., 2006. Dissolved copper triggers cell death in the peripheral mechanosensory system of larval fish. Environ. Toxicol. Chem. 25, 597-603.

Lloyd, R., 1961. Effect of dissolved oxygen concentrations on the toxicity of several poisons to rainbow trout (Salmo gairdnerii Richardson). J. Exp. Biol. 38, 447-455.

Lubiana, P., Prokkola, J.M., Nikinmaa, M., Burmester, T., Kanerva, M., Goötting, M., 2016. The effects of the painkiller diclofenac and hypoxia on gene transcription and antioxidant system in the gills of three-spined stickleback. Comp. Biochem. Physiol. Part C Toxicol. Pharmacol. 185,147-154.

Lushchak, V.I., 2011. Environmentally induced oxidative stress in aquatic animals. Aquat. Toxicol. 101,13-30.

Malekpouri, P., Peyghan, R., Mahboobi-Soofiani, N., Mohammadian, B., 2016. Metabolic capacities of common carp (Cyprinus carpio) following combined exposures to copper and environmental hypoxia. Ecotoxicol. Environ. Saf. 127, 1-11.

Manchenkov, T., Pasillas, M.P., Haddad, G.G., Imam, F.B., 2015. Novel genes critical for hypoxic preconditioning in zebrafish are regulators of insulin and glucose metabolism. G3 Genes|Genomes|Genetics 5,1107-1116.

McDonald, D., Wood, C., 1993. Branchial mechanisms of acclimation to metals in freshwater fish, Fish Ecophysiology. Springer, pp. 297-321.

Mustafa, S.A., Davies, S.J., Jha, A.N., 2012. Determination of hypoxia and dietary copper mediated sub-lethal toxicity in carp, Cyprinus carpio, at different levels of biological organisation. Chemosphere 87, 413-422.

Nawaz, M., Manzl, C., Lacher, V., Krumschnabel, G., 2006. Copper-induced stimulation of extracellular signal-regulated kinase in trout hepatocytes: the role of reactive oxygen species, Ca2+, and cell energetics and the impact of extracellular signal-regulated kinase signaling on apoptosis and necrosis. Toxicol. Sci. 92, 464-475.

Nikinmaa, M., Rees, B.B., 2005. Oxygen-dependent gene expression in fishes. Am. J. Physiol.-Regul., Integr. Comp. Physiol. 288, R1079-R1090.

Nilsson, G.E., Renshaw, G.M., 2004. Hypoxic survival strategies in two fishes: extreme anoxia tolerance in the North European crucian carp and natural hypoxic preconditioning in a coral-reef shark. J. Exp. Biol. 207, 3131-3139.

Ohmichi, K., Seno, Y., Takahashi, A., Kojima, K., Miyamoto, H., Ohmichi, M., Matsuki, Y., Machida, K., 2006. Recent heavy metal concentrations in watarase basin around ashio mine. J. Health Sci. 52, 465-468.

Ostlund-Nilsson, S., Mayer, I., Huntingford, F.A., 2006. Biology of the three-spined stickleback. CRC Press.

Padilla, P.A., Roth, M.B., 2001. Oxygen deprivation causes suspended animation in the zebrafish embryo. Proc. Natl. Acad. Sci. 98, 7331 -7335.

Podrabsky, J.E., Lopez, J.P., Fan, T.W., Higashi, R., Somero, G.N., 2007. Extreme anoxia tolerance in embryos of the annual killifish Austrofundulus limnaeus: insights from a metabolomics analysis. J. Exp. Biol. 210, 2253-2266.

Pollock, M.S., Clarke, L.M.J., Dube, M.G., 2007. The effects of hypoxia on fishes: from ecological relevance to physiological effects. Environ. Rev. 15,1-14.

Prasch, A., Andreasen, E.A., Peterson, R.E., Heiderman, W., 2004. Interations between 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and hypoxia signaling pathways in zebrafish: hypoxia decreases responses to TCDD in zebrafish embryos. Toxicol. Sci. 78, 68-77.

Prokkola, J.M., Nikinmaa, M., Lubiana, P., Kanerva, M., McCairns, R.S., Gotting, M., 2015. Hypoxia and the pharmaceutical diclofenac influence the circadian responses of three-spined stickleback. Aquat. Toxicol. 158,116-124.

Ransberry, V.E., Blewett, T.A., McClelland, G.B., 2016. The oxidative stress response in freshwater-acclimated killifish (Fundulus heteroclitus) to acute copper and hypoxia exposure. Comp. Biochem. Physiol. Part C Toxicol. Pharmacol. 179,11-18.

Reebs, S.G., Whoriskey, F.G., Fitzgerald, G.J., 2007. Diel patterns of fanning activity, egg respiration and the nocturnal behaviour of male three-spined sticklebacks, Gasterosteus asuleatus L. Can. J. Zool. 62, 329-334.

Richards, J.G., 2009. Chapter 10 Metabolic and molecular responses of fish to hypoxia. In: Jeffrey, G., Richards, A.P.F., Colin, J.B. (Eds.), Fish Physiology. Academic Press, pp. 443-485.

Roussel, J.-M., 2007. Carry-over effects in brown trout (Salmo trutta): hypoxia on embryos impairs predator avoidance by alevins in experimental channels. Can. J. Fish. Aquat. Sci. 64, 786-792.

Rytkonen, K.T., Akbarzadeh, A., Miandare, H.K., Kamei, H., Duan, C., Leder, E.H., Williams, T.A., Nikinmaa, M., 2013. Subfunctionalization of cyprinid hypoxia-inducible factors for roles in development and oxygen sensing. Evolution 67, 873-882.

Rytkonen, K.T., Prokkola, J.M., Salonen, V., Nikinmaa, M., 2014. Transcriptional divergence of the duplicated hypoxia-inducible factor alpha genes in zebrafish. Gene 541, 60-66.

Sampaio, F.G., Boijink, C.d.L., Oba, E.T., Santos, L.R.B.d., Kalinin, A.L., Rantin, F.T., 2008. Antioxidant defenses and biochemical changes in pacu (Piaractus meso-potamicus) in response to single and combined copper and hypoxia exposure. Comp. Biochem. Physiol. Part C 147, 43-51.

Sanchez, W., Palluel, O., Meunier, L., Coquery, M., Porcher, J.-M., Aït-Aïssa, S., 2005. Copper-induced oxidative stress in three-spined stickleback: relationship with hepatic metal levels. Environ. Toxicol. Pharmacol. 19, 177-183.

Sevcikova, M., Modra, H., Slaninova, A., Svobodova, Z., 2011. Metals as a cause of oxidative stress in fish: a review. Veterinary Med. 56, 537-546.

Shaw, J.R., Jackson, B., Gabor, K., Stanton, S., Hamilton, J.W., Stanton, B.A., 2007. The influence of exposure history on arsenic accumulation and toxicity in the kil-lifish, Fundulus heteroclitus. Environ. Toxicol. Chem. 26, 2704-2709.

Sneddon, L.U., Yerbury, J., 2004. Differences in response to hypoxia in the three-spined stickleback from lotic and lentic localities: dominance and an anaerobic metabolite. J. Fish Biol. 64, 799-804.

Snook, D.L., Whitehead, P.G., 2004. Water quality and ecology of the River Lee: mass balance and a review of temporal and spatial data. Hydrol. Earth Syst. Sci. Discuss. 8, 636-650.

Swarup, H., 1958. Stages in the development of the stickleback Gasterosteus acu-leatus (L.). J. Embryol. Exp. Morphol. 6, 373-383.

Team, R.C., 2014. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, 2012, Available at: http:// cran.r-project.org.

Uren Webster, T.M., Bury, N., van Aerle, R., Santos, E.M., 2013. Global transcriptome profiling reveals molecular mechanisms of metal tolerance in a chronically exposed wild population of brown trout. Environ. Sci. Technol. 47, 8869—8877. van der Geest, H.G., Soppe, W.J., Greve, G.D., Kroon, A., Kraak, M.H.S., 2002. Combined effects of lowered oxygen and toxicants (Copper and diazinon) on the mayfly Ephoron virgo. Environ. Toxicol. Chem. 21, 431—436. van Heerden, D., Vosloo, A., Nikinmaa, M., 2004. Effects of short-term copper exposure on gill structure, metallothionein and hypoxia-inducible factor-1a (HIF-1a) levels in rainbow trout (Oncorhynchus mykiss). Aquat. Toxicol. 69, 271—280.

Van Iersel, J.J.A., 1953. An analysis of the parental behaviour of the male three-

spined stickleback (Gasterosteus aculeatus L.). Behaviour (Suppl. III), 159. Waser, W., Sahoo, T.P., Herczeg, G., Merila, J., Nikinmaa, M., 2010. Physiological differentiation among nine-spined stickleback populations: effects of copper

exposure. Aquat. Toxicol. 98,188-195.

Wood, C.M., Farrell, A.P., Brauner, C.J., 2012. Homeostasis and toxicology of essential metals. Academic Press.

Wootton, R.J., 1984. A functional biology of sticklebacks. Univ of California Press.

Wu, R.S., 2009. Effects of hypoxia on fish reproduction and development. Fish. Physiol. 27, 79-141.

Zimmer, A.M., Barcarolli, I.F., Wood, C.M., Bianchini, A., 2012. Waterborne copper exposure inhibits ammonia excretion and branchial carbonic anhydrase activity in euryhaline guppies acclimated to both fresh water and sea water. Aquat. Toxicol. 122-123, 172-180.

Zimmer, A.M., Brauner, C.J., Wood, C.M., 2014. Exposure to waterborne Cu inhibits cutaneous Na+ uptake in post-hatch larval rainbow trout (Oncorhynchus mykiss). Aquat. Toxicol. 150, 151-158.