RESEARCH ARTICLE Low-O2 Acclimation Shifts the Hypoxia Avoidance Behaviour of Snapper

369

RESEARCH ARTICLE

Low-O2 acclimation shifts the hypoxia avoidance behaviour of snapper (Pagrus auratus) with only subtle changes in aerobic and anaerobic function

Denham G. Cook1, Fathima I. Iftikar2, Daniel W. Baker2, Anthony J. R. Hickey2 and Neill A. Herbert1,* 1Leigh Marine Laboratory, The University of Auckland, Leigh, Warkworth 0941, New Zealand and 2School of Biological Sciences, The University of Auckland, Auckland 1142, New Zealand *Author for correspondence ([email protected])

SUMMARY It was hypothesised that chronic hypoxia acclimation (preconditioning) would alter the behavioural low-O2 avoidance strategy of fish as a result of both aerobic and anaerobic physiological adaptations. Avoidance and physiological responses of juvenile snapper (Pagrus auratus) were therefore investigated following a 6week period of moderate hypoxia exposure (10.2–12.1kPa PO2, 21±1°C) and compared with those of normoxic controls (PO2=20–21kPa, 21±1°C). The critical oxygen pressure (Pcrit) limit of both groups was unchanged at ~7kPa, as were standard, routine and maximum metabolic rates. However, hypoxia-acclimated fish showed increased tolerances to hypoxia in behavioural choice chambers by avoiding lower PO2 levels (3.3±0.7 vs 5.3±1.1kPa) without displaying greater perturbations of lactate or glucose. This behavioural change was associated with unexpected physiological adjustments. For example, a decrease in blood O2 carrying capacity was observed after hypoxia acclimation. Also unexpected was an increase in whole-blood P50 following acclimation to low O2, perhaps facilitating Hb–O2 off-loading to tissues. In addition, cardiac mitochondria measured in situ using permeabilised fibres showed improved O2 uptake efficiencies. The proportion of the anaerobic enzyme lactate dehydrogenase, at least relative to the aerobic marker enzyme citrate synthase, also increased in heart and skeletal red muscle, indicating enhanced anaerobic potential, or in situ lactate metabolism, in these tissues. Overall, these data suggest that a prioritization of O2 delivery and O2 utilisation over O2 uptake during long-term hypoxia may convey a significant survival benefit to snapper in terms of behavioural low-O2 tolerance. Key words: preconditioning, Sparidae, oxygen affinity, metabolism, swimming, activity, haemoglobin, mitochondria, respiration. Received 26 March 2012; Accepted 20 September 2012

INTRODUCTION undoubtedly help to ensure the survival of fish facing hypoxic The prevalence of low oxygen (environmental hypoxia) has conditions (Herbert et al., 2011; Poulsen et al., 2011). Some fish increased in coastal regions (Diaz and Rosenberg, 2008), and thus engage these strategies before encountering their respective Pcrit it is now more important than ever to understand how different fish (Herbert and Steffensen, 2005; Poulsen et al., 2011), but the New species will respond to environmental change. Because hypoxia Zealand snapper (Pagrus auratus, Sparidae) does not show these presents a significant metabolic challenge for most aquatic animals, responses (Cook et al., 2011). Snapper are late to leave hypoxic physiological adaptations to chronic hypoxia generally involve water because low O2 avoidance commences below their Pcrit (Cook enhancing the capacity for O2 uptake and delivery (Richards, 2009). et al., 2011). Moreover, snapper do not show any significant change For example, many fish species studied to date increase red blood in swimming speed during hypoxic exposure (Cook et al., 2011). cell numbers and haemoglobin (Hb) concentrations to boost O2 The snapper used in this study are presumed to have never carrying capacity (Wells, 2009; Wells et al., 1989), restructure gill experienced low O2, raising the possibility that behavioural morphologies to enhance gas exchange (Sollid et al., 2003), increase responses and resulting physiological impacts could differ with Hb–O2 binding affinities to increase O2 uptake (Wood and Johansen, previous exposure as a result of a loss of naivety and/or acclimatory 1972; Wood et al., 1975) or even modify cardiac function to improve responses (i.e. physiological adaptation). low O2 performance (Petersen and Gamperl, 2010). When combined, To date, the effects of hypoxic acclimation on behaviour of fish cellular and tissue modifications should enhance the whole animal’s in low O2 conditions remain largely untested. Moreover, little direct capacity to tolerate hypoxia, and be evident as a lowering of the evidence identifies how adaptations associated with hypoxia critical O2 tension (Pcrit), which defines the partial pressure of O2 preconditioning influence fish behaviour. Therefore, the present (PO2) above which basal metabolic demand (maintenance) is study aimed to resolve whether the behavioural avoidance and satisfied (Timmerman and Chapman, 2004). swimming speed response of snapper would differ after acclimation Fish engage an array of physiological mechanisms to combat low to long-term hypoxia. The physiology of snapper was also oxygen, but may also use behavioural strategies to counter investigated in detail using both novel and commonly applied deleterious effects (Richards, 2009). This is particularly true for measures of aerobic and anaerobic physiology to gauge how species that have a limited capacity to adapt physiologically physiological changes (cellular, organ and whole animal) integrate (Pichavant et al., 2003). Reductions in locomotory activity and early with behavioural responses. We hypothesised that snapper would avoidance of low O2 represent two notable behaviours that would show one of two responses. Firstly, long-term hypoxia could

THE JOURNAL OF EXPERIMENTAL BIOLOGY 370 The Journal of Experimental Biology 216 (3) convey a degree of low-O2 experience to fish whereby they are actively move between these two flows via a square port (10×10cm) simply less naive and employ a more cautious avoidance strategy positioned centrally on the divider. Diffusers and baffles were used −1 well above Pcrit limits. In this scenario, snapper would adjust their to create rectilinear flow in the BA (4800lh combined flow rate) behavioural strategy and simply avoid low O2 earlier without any and each side of the BA received water from a large (400l) gassing major changes in physiology. Alternatively, long-term hypoxia could tower. PO2 on either side of the BA was therefore manipulated by provide greater low-O2 tolerance across a number of physiological purging nitrogen gas (BOC Gas Supplies), or compressed air, levels (e.g. improved Hb–O2 transport potential). In this scenario, through the towers. Oxygen set-points were controlled by two fish might remain in hypoxia and avoid even lower levels of O2, Oxyreg units (Loligo Systems, Tjele, Denmark) coupled to a DAQ feasibly without enlarged levels of low O2 stress. The present study device (miniLAB 1008, Measurement Computing, Norton, MA, therefore set out to resolve how physiological and behavioural USA) under the control of Labview software (v. 8.6, National changes might interact following chronic hypoxia. Instruments, Austin, TX, USA). The behavioural activity of fish in the choice box was quantified using a digital camera (Fire-I, MATERIALS AND METHODS Unibrain, San Ramon, CA, USA) that streamed video to a PC Fish handling and acclimation procedures running ‘Swistrack’ software (Correll et al., 2006). The movements Juvenile Pagrus auratus (Forster 1801) (Sparidae; common name of fish were sampled at a rate of 15Hz and were used to resolve snapper or red bream; 150–300g) were captured by line and avoidance behaviour and activity variables (including swimming barbless hooks from coastal waters around Leigh (36°19′S, speed and spontaneous turning rates). Water temperatures within 174°48′E, Northland, New Zealand). Following capture, fish were the choice chamber were actively maintained at 21.0±0.3°C (mean housed in one of two 500l tanks (maximum of 50 individuals per ± 95% CI). tank) at the Leigh Marine Laboratory. Fish were provided with a Fish were transferred to the BA at least 18h prior to continuous flow of high-quality, aerated seawater for at least experimentation. Following this period, behavioural variables were 6weeks before experimentation. After this period, one tank was determined over a 1h control period, during which any preference designated for hypoxia acclimation and the other as a normoxic for side and normoxic swimming activity was determined. During control. A reduced PO2 (10.2–12.1kPa) was maintained in the this 1h control period, all fish formed a strong preference for one hypoxia tank using an oxygen controller (Model PR514, PR particular water flow (i.e. >80% of time spent on one particular side Electronics, Rønde, Denmark) and an Oxyguard oxygen electrode of the divider) and excursions into the alternate channel were (Mini probe, Technolab, Mornington, Tasmania, Australia) infrequent and never in excess of 15s. This behavioural routine positioned in the centre of the holding tank. The PO2 level of formation presented investigators with the opportunity to ‘drive’ 10.2–12.1kPa was deemed a significant level of hypoxia for this fish from their preferred side and identify clear avoidance thresholds species because it reduces their aerobic scope by >50% at lower (Cook et al., 2011). Following the 1h control period, the preferred experimental temperatures (Cook et al., 2011). An on/off relay channel was deoxygenated progressively at a linear rate of −1 output from the controller actuated a solenoid-controlled flow of 1kPa5min . The PO2 at which snapper avoided their preferred compressed nitrogen (BOC Gas Supplies, Auckland, New Zealand) channel for 30s was taken as the avoidance threshold. The through a fine bubble diffuser fitted to the floor of the holding tank. experiment was terminated at this point; fish were immediately A second relay was programmed to actuate an auxiliary flow of captured and then rapidly euthanized by brain ablation before blood compressed air if PO2 values fell below the desired set-point of was sampled via caudal venepuncture (<30s post capture). 10.2kPa. This provided protection against any sudden drop in PO2 (e.g. as a result of post-prandial metabolism). Normoxic conditions Haematological and biochemical techniques in the control tank (>95% saturation) were maintained by continually A heparinised syringe and needle was used to draw caudal mixed passing compressed air through a fine bubble diffuser fitted in a blood at the point of avoidance in the behavioural trial above (N=8 similar fashion to the hypoxia tank. The initial stage of hypoxia per treatment), as well as a separate resting control group (N=10) acclimation involved decreasing the PO2 of the hypoxia tank at a for comparison. Whole blood was placed in 75mm capillaries and rate of ~1kPaday–1 over a 10day period until the desired minimum haematocrit (Hct; the percentage of red blood cells) was determined set-point range was established. Fish remained under these after 3min of centrifugation (Haemocentaur, MSE, London, UK). conditions for 6 to 12weeks, enabling investigators to complete the In addition, 10µl of whole blood was pipetted into 1ml of modified sampling and observational phase of the experiment. Normoxic Drabkin’s reagent and haemoglobin concentration ([Hb]) was controls were subject to the same period of experimental holding. calculated after reading absorbance at 540nm against a blank (Wells PO2 in both tanks was confirmed regularly with a Cell-Ox probe et al., 2007). Mean corpuscular haemoglobin concentration (MCHC) connected to a WTW 3310 meter (Wissenschaftlich-Technische was estimated from the ratio of [Hb]:(Hct×100). The remaining Werkstätten, Weilheim in Oberbayem, Germany). Water blood was then centrifuged (14,000g, 4°C), and the plasma was temperatures across the duration of holding and experimentation separated before being stored at −80°C for later analysis of plasma ranged from 20.1 to 21.8°C. Fish were fed a standard ration of squid glucose and lactate using standard enzymatic techniques (described and pilchard. All capture, holding and experimental techniques were in Cook et al., 2011). performed under approval of The University of Auckland Animal Muscle enzyme activity was determined according to methods Ethics Committee (approval: R711). detailed previously (Hickey and Clements, 2003). Pre-weighed samples of tissue (~50mg) were diluted in ice-cold homogenisation −1 The behavioural response of snapper to hypoxia buffer (inmmoll : 25 Tris-HCl pH7.8, 1 EDTA, 2 MgCl2, 50 KCl, Behavioural responses to an avoidable progressive hypoxia stimulus 0.5% Triton-X 100) to a ratio of 1:10 (w/v). Samples were (referred to as ‘escapable hypoxia’) were investigated in an oxygen homogenised (TissueLyser II, Qiagen, Auckland, New Zealand) and choice chamber described elsewhere (Cook and Herbert, 2012; Cook then centrifuged (20,000g, 4°C), with the supernatant retrieved for et al., 2011). Individual fish occupied a behavioural arena (BA) that analysis. Lactate dehydrogenase (LDH) activity was determined by consisted of two flows of water separated by a divider. Fish could adding 20µl of appropriately diluted tissue homogenate to 180µl

THE JOURNAL OF EXPERIMENTAL BIOLOGY Snapper responses to hypoxia acclimation 371 of LDH assay mix (in mmoll−1: 100 Tris-HCl pH7.0, 1 EDTA, 2 aspartate, 0.05 DL-carnitine, 1mgl−1 insulin (porcine), 1mgl−1 MgCl2, 1 dithiothreitol and 0.15 NADH). The oxidation of NADH thiamine pyrophosphate-co-carboxylase, pH7.6 (Forgan and Forster, was measured at 340nm in a plate reader (Spectramax 340, 2010). After 5–10min of perfusion, the heart was removed from Molecular Devices, Sunnyvale, CA, USA) after addition of 25µl the perfusion setup and prepared for mitochondrial respiration 1.5mmoll−1 pyruvate. Citrate synthase (CS) was measured by adding assays. Permeabilised heart fibres were prepared as follows. Spongy 30µl of dilute tissue homogenate to 170µl of CS assay mix [in myocardium excised from the perfused heart was placed into mmoll−1: 50 Tris-HCl pH8.0, 0.1 acetyl coenzyme A and 0.2 5,5′- modified, ice-cold relaxation medium (BIOPS, in mmol−1: 2.77 dithiobis-(2-nitrobenzoic acid) (DTNB)]. The reduction of DTNB CaK2EGTA, 7.23 K2EGTA, 5.77 Na2ATP, 6.56 MgCl2·6H2O, 20 was followed at 412nm after the addition of 25µl 5mmoll−1 taurine, 20 imidazole, 0.5 dithiothreitol, 50 K-MES, 15 Na-PCr and oxaloacetate. LDH and CS enzyme activity were expressed as 50 sucrose, pH7.1) and then teased apart into fibre bundles and Umg−1 but also in terms of their ratio to one another (LDH:CS). transferred to permeabilising solution (BIOPS + 50μgml−1 saponin). The use of the LDH:CS ratio accounted for individual variability After gentle shaking (30min, 0°C) the tissue was removed from the in enzyme activities and, importantly, enabled us to resolve permeabilising solution and washed three times in a modified differences in the relative anaerobic and aerobic capacities of fish mitochondrial respiratory medium [MiRO5, in mmoll−1 unless between the two treatments (Hochachka et al., 1983). All chemicals otherwise indicated: 0.5 EGTA, 3 MgCl2·6H2O, 60 K-lactobionate, −1 were sourced from Sigma-Aldrich (Auckland, New Zealand) 20 taurine, 10 KH2PO4, 20 Hepes, 160 sucrose and 1gl bovine As all blood components had been previously washed from serum albumin (Gnaiger et al., 2000)]. Fibres were removed from ventricular tissue within the modified Langendorff preparation (see MiRO5, blotted dry and weighed into bundles (~5–10mg) for below), ventricular myoglobin concentrations ([Mb]) could be respirometric determinations. Isolated mitochondria were prepared determined using the cyanmetmyoglobin (cyanmetMb) method by immersing 50–100mg of cardiac tissue into ice-cold MiRO5 (Drabkin, 1950). Two hundred microlitres of supernatant (also followed by manual mincing. After washing two to three times with utilised for enzyme activity analyses) was added to 800µl of MiRO5, the tissue was digested (MiRO5 + 5% trypsin) for 30min modified Drabkin’s reagent. Absorbance at 540nm was recorded and then filtered through fine muslin before re-suspension in in 1cm wide cuvettes (Multiskan Spectrum, Thermoscientific, 1000µl MiRO5. Following centrifugation (700g, 10min, 4°C), the Vaanta, Finland). Tissue [Mb] was calculated using a cyanmetMb supernatant was transferred to a new 1.7ml microcentrifuge tube extinction coefficient of 10.36mmolcm−1 according to the and spun (8000g, 10min, 4°C). The supernatant was discarded and methodology of Viriyarattanasak and colleagues (Viriyarattanasak 100µl of MiRO5 was added to resuspend the remaining pellet et al., 2011). containing isolated mitochondria. Protein concentration was determined from an aliquot of the mitochondrial suspension using Blood oxygen binding properties the BCA protein assay (ThermoFisher Scientific, Auckland, New Whole-blood binding properties and the fixed acid Bohr effect of Zealand) according to guidelines provided. All chemicals utilised Hb were investigated in venous mixed blood drawn from the caudal were obtained from Sigma-Aldrich. vein of rested fish in each treatment (N=7). Samples were prepared by suspending 50µl of whole blood in 4ml of 50mmol−1 Hepes Mitochondrial assays – permeabilised heart fibres salt solution (125mmoll−1 NaCl), buffered to a range of Fibres typically require super-saturation with oxygen in order to physiologically relevant pH levels (7.8, 7.6, 7.4, 7.2, 7.0 and 6.8). determine maximal flux capacities (Gnaiger, 2009). In general Upon dilution of whole blood, residual catecholamines that might settings, such as with mammalian cardiac or skeletal muscles, this influence blood O2 binding properties were expected to degrade may be problematic as muscle fibres are typically perfused by rapidly (Wells et al., 2003). Antifoam (2µl) was added to the solution capillaries, which are not functional in vitro. However, fish immediately before analysis. The binding properties of haemoglobin cardiomyocytes, in particular the spongy myocardium of the snapper, solutions at each pH were then investigated using a Haem-ox is perfused by luminal blood. This permitted us to test the oxygen analyser (TCS Research Products, New Hope, PA, USA). Oxygen dependence of flux (JO2) of permeabilised fibres isolated from equilibrium curves (OECs), Hill’s cooperativity coefficient (n) and snapper ventricle, without concerns of experimental diffusion the PO2 at which Hb was half O2 saturated (Hb-P50) were calculated limitation. using Haem-ox analytical software (version 2.14, TCS Research Using an Oroboros Oxygraph-2K respirometer (Oroboros Products). The Bohr coefficient (ф) was determined using the Instruments, Innsbruck, Austria), the mitochondrial function of heart standard equation, ф=∆logHb-P50/∆pH. The Root effect was tissue from the two acclimation treatments was evaluated by recorded as the decrease in maximal Hb saturation at each measured successive addition of tricarboxylic acid (TCA) cycle substrates pH level relative to a reference value for maximal Hb saturation (malate, glutamate, pyruvate then succinate) to stimulate (Hb-P100) at pH7.8 and 21°C (Regan and Brauner, 2010; Wells and Complex I, then Complex I+II, respiratory states (Gnaiger, 2009). Dunphy, 2009). Excess ADP (2.5mmoll−1) was added to stimulate oxidative phosphorylation (OXPHOS) (see Fig. 3). Following these Mitochondrial respiration determinations, chambers were equilibrated to ~21kPa. Once fueled Tissue preparation with further respiratory substrates (inmmoll−1: 5 malate, 10 The heart of resting, euthanised snapper was perfused in a modified glutamate, 10 pyruvate, 10 succinate and 2.5 ADP), oxygen was Langendorff preparation. The bulbous arteriosis was cannulated and allowed to deplete into anoxia with oxygen flux rates determined perfused with a gravity delivered flow of teleost Ringer’s solution, using Datlab 4.2.3.2.7 (Oroboros Instruments, Innsbrook, Austria). which, in combination with spontaneous ventricular contractions, The characteristics of mitochondrial respiration during progressive served to wash all blood from the tissue. The Ringer’s solution was O2 depletion [oxygen-dependent flux (ODF)] were quantified by composed of (in mmoll−1, unless otherwise indicated): 159.6 NaCl, transforming exported data according to the following equation: 2.1 KCl, 0.99 MgCl2, 1.30 CaCl2, 10 glucose, 3 Hepes acid, 6.99 ODF=(JO2/O2)/O2. When plotted graphically, this transformed Hepes sodium salt, 0.30 Na-glutamate, 0.40 L-glutamate, 0.02 Na- measure of respiration was observed to decrease in a linear fashion,

THE JOURNAL OF EXPERIMENTAL BIOLOGY 372 The Journal of Experimental Biology 216 (3) enabling determination of slope characteristics – the gradient of this oxygen partial pressure (Pcrit) were determined using intermittent line is here termed the oxygen-dependent flux index (ODFI). ODFI closed-phase respirometry with an automated protocol (Cook et al., values are greatest in tissues showing high oxygen dependence and 2011). Fish were transferred from their holding tanks to a 5l lowest in tissues showing low oxygen dependence. respirometry chamber supplied with fully saturated (normoxic) water at constant temperature (21±0.1°C). A computer running custom Mitochondrial assays – isolated mitochondria software controlled the flush, wait and measure cycles of the Isolated mitochondria from the two acclimation treatments were respirometry protocol chamber (4, 1 and 5min intervals, initially evaluated by successive addition of TCA cycle substrates respectively) and recorded the closed-phase decline in chamber O2 (as above). As isolated mitochondria are not diffusion limited, the saturation via a WTW CellOx probe connected to a 3310 meter Mito-P50 (i.e. the PO2 at which mitochondrial respiration is 50% of (Wissenschaftlich-Technische Werkstätten, Weilheim, Germany). maximum recorded levels) of isolated mitochondria was determined The mass-specific rate of oxygen consumption (MO2) was calculated following respiratory depletion of oxygen from saturating conditions from the decline in closed-phase O2 (i.e. one measure of every (~21kPa) to anoxia. 10min) according to standard formulae, detailed previously (Cook et al., 2011). SMR was thus resolved over 40h of normoxia using Microscopy techniques 240 measures of MO2 and the 15% quantile method (Chabot and The mitochondrial ultrastructure of cardiac tissues (ventricle) was Claireaux, 2008; Cook et al., 2011; Dupont-Prinet et al., 2010). RMR investigated to identify whether hypoxic preconditioning changed was defined as the mean MO2 throughout the duration of recording. the structural morphology of the heart. Excised, perfused sections MMR was defined as the peak measurement of MO2 observed during of spongy myocardium were immersed in fixative buffer the first three recorded measurement cycles. Once SMR, RMR and −1 −1 (10mmoll Hepes, 250mmoll sucrose, pH7.1, 2.5% MMR were resolved, the PO2 of water supplying the respirometry glutaraldehyde) and then stored at 4°C. Preparation involved chamber was reduced progressively to the following levels: 15, 12, −1 washing samples three times (10minwash ) with Sorensen’s buffer 9, 7.5, 6 and 4kPa. These steps were used to identify the Pcrit where −1 −1 (0.1moll ) before being fixed (1% osmium tetroxide in 0.1moll MO2 was reduced below SMR and fish were in an oxy-conforming Sorensen’s buffer). This was followed by a series of alcohol washes state (Behrens and Steffensen, 2007). Defined PO2 set-points were (30–100% ethanol, dry ethanol, and then two washes in 100% achieved by circulating water through a 40l gassing tower that was acetone). Samples were infiltrated with epoxy resin (1:1, 812 intermittently purged with compressed nitrogen for deoxygenation. epoxy:acteone), later replaced with 100% epoxy resin and left Three MO2 measures were recorded at each PO2 level. A one-way overnight. Resin was removed the following morning and replaced ANOVA and a Tukey’s post hoc test were used to identify levels with fresh 100% epoxy. After a further 6h, tissue was transferred of PO2 at which fish could not maintain SMR (i.e. resting to moulds and kept at 60°C for 48h. Following trimming, ultra-thin MO2

Table1. Behavioural and haematological parameters from snapper acclimated to long-term normoxia (~21kPa) or chronic moderate hypoxia (10.2–12.1kPa) for >6weeks Normoxia acclimation Moderate hypoxia acclimation Resting Post-hypoxia avoidance Resting Post-hypoxia avoidance Behavioural parameters Routine swimming speed (BLs–1) 0.63±0.07 0.55±0.09 Hypoxia avoidance threshold (kPa) 5.3±1.1 3.3±0.7† N 88 Haematological parameters Haematocrit (%) 24.8±1.7 27.0±2.1 22.7±2.1 27.1±1.5‡ Haemoglobin (gdl–1) 5.6±0.3 6.4±0.5‡ 4.8±0.4* 6.0±0.2‡ MCHC (gdl–1) 22.6±1.7 23.6±1.1 21.1±1.9 22.1±0.8† Plasma lactate (mmoll–1) 1.3±0.4 3.7±0.6‡ 0.9±0.5 4.0±0.5‡ Plasma glucose (mmoll–1) 1.7±0.5 5.2±0.7‡ 1.7±0.3 5.1±0.9‡ N 88 8 8 Mean values are presented ±95% CI. Significant differences (P<0.05) between resting and post-hypoxia avoidance values are depicted: *comparisons between normoxia- and hypoxia-acclimated resting controls, †comparisons between normoxia- and hypoxia-acclimated groups post hypoxia exposure; ‡comparisons between resting control and post-avoidance fish from respective treatments (normoxia or hypoxia acclimated). BL, body lengths; MCHC, mean corpuscular haemoglobin concentration.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Snapper responses to hypoxia acclimation 373

Comparisons between avoidance levels and the physiological state 1.0 of treatments at the point of avoidance were performed using a ) A –1 Student’s t-test. Between-treatment comparisons of descriptors of 0.8 mitochondrial ultrastructure, enzyme activities and heart [Mb] were all performed using Student’s t-tests, with data appropriately (e.g. 0.6 log) transformed when variances were not equal. OEC data were compared using two-way repeated-measures ANOVA, with Tukey’s 0.4 post hoc tests incorporated where applicable. The Schrier–Ray–Hare 0.2 test was performed in SPSS (v. 18, IBM, www.ibm.com/software/ analytics/spss/), but all other analyses were performed in SigmaPlot s Swimming speed (BL 0 (v. 11, Systat Software, www.systat.com). All results are presented as means ± 95% CI. Significance was accepted at P<0.05. 60 B )

–1 50 RESULTS 40 Behavioural and physiological responses to progressive hypoxia 30 Individuals exposed to moderate hypoxia for 6–12weeks avoided 20 low-O2 conditions at significantly lower PO2 (3.3±0.7kPa) than normoxia-acclimated individuals (5.3±1.1kPa, F=12.20, P<0.05; 10

 rate (deg s Turning Table1), but at all levels of PO2, the two treatments behaved similarly 0 in terms of swimming speed, turning rate, percentage of time in 100 lowest PO and the frequency of hypoxic excursions (Fig.1, refer C

2 2 O to Table2 for statistical summary). No treatment differences or  90 interaction effects existed in any behavioural variable investigated. The haematological profiles of both normoxia- and hypoxia- 80 acclimated snapper were different to those of rested (unexposed) controls immediately after avoidance (see Table1). Specifically, 70 exposing normoxia- and hypoxia-acclimated snapper to hypoxia 60 yielded a significant increase in [Hb] (t=–2.14, P<0.05 and t=–3.32, % of time in lowest P P<0.01, respectively), plasma lactate (t=–2.219, P<0.01 and t=–7.12, 50 P<0.01) and plasma glucose (t=–3.32, P<0.01 and t=–8.90, P<0.01). A significant increase in blood Hct was also observed in hypoxia- 3.0 D Normoxia conditioned acclimated individuals (t=–3.96, P<0.01), but normoxia-acclimated 2.5 Hypoxia conditioned ) fish only showed a near-significant increase (t=–1.95, P=0.08). –1 2.0 Despite hypoxia-acclimated fish being exposed to lower P (see O2 1.5 above) there was no significant difference in [Hb] (t=1.29, P=0.22), Hct (t=–0.27, P=0.79), MCHC (t=–1.49, P=0.15), plasma lactate 1.0 (t=–1.72, P=0.49) or plasma glucose (t=–0.30, P=0.77) between the 0.5 (excursions min two groups following avoidance. This lack of difference in Hypoxic excursions 0 physiology between the two treatments is particularly important and –0.5 is discussed further below. Control 21–16 16–12 12–8 8–4 4–0

Blood and muscle physiology and biochemistry PO2 (kPa) The haemoglobin of snapper exhibits typical responses to pH. For example, pH had a strong negative effect on Hb-P50 (F=21.38, Fig.1. Behavioural characterisation of snapper to an escapable progressive P<0.01, Fig.2A), whilst exerting a positive effect on binding hypoxia protocol following 6weeks exposure to normoxic (closed symbols) cooperativity (n-value, F=182.42, P<0.01; Fig.2B). Snapper blood or hypoxic (open symbols) conditions. (A) Average swimming speed, (B) average spontaneous turning rate, (C) percentage time spent on the also revealed a modest Root effect according to the reduction in preferred (and therefore progressively hypoxic) side of the BA, and (D) maximal Hb–O2 saturation at pH<7.4 (F=32.94, P<0.01; Fig.2C). frequency of movements from preferred side of the chamber into the In terms of treatment differences, the Hb-P50 of hypoxia-acclimated normoxic refuge. Error bars represent 95% confidence intervals. BL, body snapper was on average 12% higher than normoxia-acclimated lengths. individuals at any level of pH (F=7.84, P<0.01) but the Bohr coefficient of normoxia-acclimated fish (ф=–1.02±0.12) was similar to that of hypoxia-acclimated individuals (ф=1.08±0.12; t=0.66, individuals (t=–2.14, P<0.05). Furthermore, only LDH activity in P=0.524). The sigmoidal shape of the oxygen binding curve (n- the red skeletal muscle was different between hypoxia- and value), at any level of pH, was also not affected by treatment normoxia-acclimated fish; all other enzyme activities were similar conditions (F=1.13, P=0.29). (Table3). It is important to note that when LDH was expressed Exposure to moderate hypoxia across a 6week period led to other relative to CS, the LDH:CS ratios were higher in cardiac and red minor modifications in the resting physiology of snapper (for a skeletal muscle of hypoxia-treated fish. LDH activity and LDH:CS summary of results, see Table1). With respect to haematology, only in white muscle tissue did not differ significantly. Snapper [Hb] differed between treatments under resting conditions; the [Hb] ventricular myoglobin concentrations were not affected by long- of hypoxia-exposed individuals was 8% lower than that of normoxic term hypoxia (~2.7mgg−1, pooled values).

THE JOURNAL OF EXPERIMENTAL BIOLOGY 374 The Journal of Experimental Biology 216 (3)

Table2. Statistical results from the Schrier–Ray–Hare test 10.0 identifying the combined effects of acclimation treatment, level of A 67.5 PO2 and potential interactions upon exposure to a progressive low- 8.0 O stimulus 2 52.5 Effect SS H P d.f. 6.0 Swimming speed 37.5 (kPa)

Treatment 906.5 0.33 0.571 1 4.0 (mmHg) 50 50 PO level 1365.8 4.94 0.295 4 P 2 22.5 P Interaction 1307.7 0.47 0.986 4 Turning rate 2.0 Treatment 1665.3 0.76 0.921 1 7.5

PO2 level 5870.5 2.70 >0.999 4 0 Interaction 3399.7 1.56 >0.999 4

% time in lowest PO2 2.4 Treatment 647.9 0.31 0.950 1 B

PO2 level 3409.9 1.63 >0.999 4 2.2

Interaction 5217.3 2.49 >0.999 4 n Excursion duration 2.0 Treatment 64.8 0.03 0.983 1 1.8 PO2 level 2407.4 1.15 >0.999 4 Interaction 5691.8 2.66 >0.999 4 1.6

1.4 Hill’s coefficient, coefficient, Hill’s 1.2 Mitochondrial respiration and ultrastructure Respiration flux of isolated cardiac mitochondria was not 1.0 significantly affected by the hypoxia acclimation treatment (Table4). 100 Complex I- and II-supported oxidative phosphorylation (Complex C I, OXPHOS and Complex I+II, OXPHOS) and ‘leak’ rates 95 (synonymous with Complex I respiration) were all statistically similar between treatments. Despite these similarities, permeabilised cardiac fibres showed different oxygen-uptake dynamics: the 90 respiration of fibres from the heart of hypoxia-acclimated fish

saturation (%) 85 appeared to be less affected by the diffusive oxygen gradient between 2 the respiratory medium and active mitochondria. This was reflected

Hb–O 80 by a lower conformance to PO2 (i.e. low ODFI scores) in the hypoxia- Normoxia conditioned acclimated fish (t=–2.560, P=0.03; Fig.3). Snapper heart Hypoxia conditioned mitochondria occupied ~19% of myocyte volume with mean cristae 75 densities of ~22cristaeµm−1 (Fig.4), and were not altered by    6.6 6.8 7.0 7.2 7.4 7.6 7.8 8.0 6weeks exposure to moderate hypoxia (see Table4 for statistical summary). pH Fig.2. Haemoglobin–oxygen binding properties of in vitro snapper blood Whole-animal respirometry following 6weeks exposure to normoxic (closed symbols) or hypoxic (open Long-term exposure to moderate hypoxia had no significant effect symbols) conditions. (A) Bohr shift demonstrating the effect of pH on the on the rates of O2 consumption (Fig.5). The SMR of the normoxia Hb-P50 value. (B) Hillʼs coefficient (n) identifying the cooperativity of Hb–O2 −1 −1 treatment group was 152.1±13.4mgO2kg h compared with binding. (C) Maximal saturation of Hb relative to that observed at pH7.8, −1 −1 indicative of the Root effect. 155.4±10.4mgO2kg h in the hypoxia treatment (F=0.46, −1 −1 P=0.51). RMR was 196.69±10.83 and 209.2±20.4mgO2kg h for the normoxic and hypoxic treatment groups, respectively (F=1.99, present study. It therefore seems likely that long-term hypoxia does P=0.18). MMR in the normoxia treatment group was 442.38±28.64 not encourage snapper to adopt a more cautious avoidance response −1 −1 and 420.1±21.5mgO2kg h in the hypoxia treatment group by leaving low O2 conditions at higher PO2. In fact, quite the opposite (F=1.29, P=0.28). Pcrit values from the normoxic fish was observed. Given that each treatment was exposed to a (6.93±0.61kPa) were no different to those of the hypoxia-acclimated standardised and linear drop in ambient PO2, hypoxia-acclimated fish (7.01±0.55kPa; t=–0.19, P=0.85). individuals spent a longer period of time exposed to critically low O2 that increased in severity (i.e. 2.4kPa lower PO2 for an extra DISCUSSION 12min). Yet surprisingly, hypoxia-acclimated fish did not experience Long-term hypoxia improves tolerance of low O2 higher levels of low-O2 stress because levels of plasma lactate were Snapper exposed to moderate long-term hypoxia (10.2–12.1kPa) comparable between treatments. The ability of fish to tolerate more avoid low O2 at a significantly lower level of PO2 than normoxic severe hypoxia for greater periods of time, but show similar levels controls (3.1±0.7 vs 5.5±1.1kPa; Table1), signifying that hypoxia of physiological perturbation at the point of avoidance, would acclimation does not sensitise the behavioural response of this strongly suggest that hypoxia-acclimated fish gained a significant species as a result of low-O2 experience. It has previously been advantage in terms of improved low-O2 tolerance. This was reported that snapper avoid hypoxia below Pcrit thresholds (Cook presumably the result of a change in physiology at low O2 because et al., 2011) and this was again observed with both groups in the fish did not show any downregulation of swimming speed or turning

THE JOURNAL OF EXPERIMENTAL BIOLOGY Snapper responses to hypoxia acclimation 375

Table3. Muscle properties of snapper acclimated to either long- studies. Indeed, both the concentration and O2 affinity of Hb term normoxia (~21kPa) or chronic moderate hypoxia decreased by 8 and 12%, respectively, following hypoxia exposure, (10.2–12.1kPa) for 6weeks changes that would not typically be deemed adaptive. The drop in Normoxia-exposed Hypoxia-exposed resting [Hb] should be detrimental, unless this change accompanied group group tPa decreased Hct, which could lower cardiac energetic requirements by decreasing blood viscosity (Gallaugher et al., 1995). However, Heart LDH 101.2±37.3 130.0±33.3 –0.939 0.336 no changes in Hct were observed, suggesting that the reduction in CS 21.8±5.3 22.7±6.2 0.155 0.880 [Hb] would not have benefitted cardiac function. Because Hct, [Hb] LDH:CS 4.1±0.4 5.9±0.3* –7.105 <0.01 and MCHC levels were equivalent in normoxia- and hypoxia- [Mb] 2.69±0.5 2.70±0.7 –0.043 0.966 acclimated fish after acute low O2 (Table1), the differences Red muscle observed in resting fish could have been due to altered red blood LDH 0.63±0.07 0.55±0.09 –2.211 0.047 cell distribution (i.e. splenic storage). With respect to the O2-binding CS 28.2±4.8 23.5±3.7 1.177 0.262 characteristics of Hb, the observed drop in Hb–O affinity may LDH:CS 0.28±0.05 0.39±0.05* –3.354 0.006 2 White muscle have provided an adaptive benefit as it would favour O2 offloading LDH 91.5±9.0 101.1±22.0 –0.265 0.796 at the tissues. Indeed, this condition could aid aerobic tissue function CS 5.3±1.1 3.3±0.7 0.805 0.436 by elevating O2 diffusion gradients while also potentially increasing LDH:CS 62.9±15.01 70.2±21.9 –1.175 0.265 the O2 content of the venous return (Brauner and Wang, 1997; Wang N 88 and Malte, 2011) and the availability of O2 to the snapper heart. Mean values are presented ±95% CI. All enzyme activity values are Although generally considered atypical for fish in hypoxia, the –1 expressed as Umg tissue (wet mass). CS, citrate synthase; LDH, lactate observed drop in Hb–O2 affinity of snapper is consistent with the dehydrogenase; LDH:CS, LDH activity standardised to CS activity. small decrease in Hb–O affinity of lowland mammals during –1 2 Myoglobin concentration ([Mb]) is expressed as mgg tissue (wet mass). moderate hypoxia (Lenfant, 1973). It could therefore be argued that exposing snapper to more extreme levels of hypoxia might have induced more ‘typical’ haematological responses (e.g. behaviour that could otherwise have reduced energetic expenditure increased Hb–O2 affinity). However, 10.2–12.1kPa does not and allowed fish to reside in hypoxia for longer periods (Fig.1A,B). actually represent a moderate level of PO2 for snapper. Instead, with a >50% reduction of metabolic scope at this level (Cook et Is there evidence of aerobic adjustments improving hypoxia al., 2011), it is unlikely that our fish would have retained adequate tolerance? fitness characteristics at lower PO2. This level of hypoxia was Exposing fish to long-term hypoxia is typically considered to therefore chosen as challenging but ethically acceptable for long- improve hypoxia tolerance through alterations in Hb. For example, term acclimation. Interestingly, the haematological response elevated [Hb] enhances the blood O2 carrying capacity (Lai et al., observed in snapper could potentially limit O2 uptake potential, 2006; Petersen and Gamperl, 2011; Wells et al., 1989). Similarly, but no measures of whole-animal O2 uptake (including Pcrit) increased Hb–O2 affinities are commonly reported to improve the actually differed between the two groups (Fig.5). No evidence exists O2-uptake capacity of Hb during environmental hypoxia (Weber to suggest that the drop in [Hb] and Hb–O2 affinity hindered O2 et al., 1976; Wood and Johansen, 1972). Both features can be uptake during hypoxia. It therefore seems plausible that snapper viewed as beneficial (adaptive) to the survival of fishes in hypoxic possessed adequate capacities for O2 uptake during long-term waters. However, following 6weeks of chronic hypoxia, snapper hypoxia, and that a physiological advantage was possibly gained did not show many of the responses commonly observed in other with a prioritisation of O2 delivery to tissues.

Table4. Mitochondrial characteristics of snapper acclimated to either long-term normoxia (~21kPa) or chronic moderate hypoxia (10.2–12.1kPa) for 6weeks Normoxia-exposed group Hypoxia-exposed group tP Isolated mitochondria

Complex I (leak) activity (JO2, GMP) 341.90±138.45 357.14±164.72 0.139 0.894 Complex I, OXPHOS (JO2, GMP-ADP) 611.87±162.33 758.02±172.46 0.882 0.412 Complex I+II, OXPHOS (JO2, GMPS-ADP) 3284.41±603.97 3762.90±874.78 1.209 0.272 Mt-P50 (kPa) 0.011±0.001 0.012±0.002 –0.579 0.583 N 44 Permeabilised cardiac fibres

Complex I (leak) activity (JO2, GMP) 11.67±2.29 12.69±3.06 0.542 0.599 Complex I, OXPHOS (JO2, GMP-ADP) 39.76±4.24 47.92±7.57 1.853 0.092 Complex I+II, OXPHOS (JO2, GMPS-ADP) 21.11±2.26 24.96±2.96 2.059 0.066 ODFI 0.78±0.06 0.66±0.07 –2.56 0.028 N 66 Cardiac ultrastructure Mitochondrial density (%) 19.8±0.8 19.2±0.6 1.285 0.246 Cristae density (cristaeμm–1) 22.0±2.4 22.3±0.5 –0.219 0.834 N 44

–1 –1 Mean values are presented ±95% CI. Respiration within permeabilised fibres is expressed as weight-specific oxygen flux (JO2) (pmolO2s mg wetmass) –1 –1 respiration in isolated mitochondria was measured as oxygen flux (pmolO2s mg protein). G, glutamate; M, malate; P, pyruvate; S, succinate. Mt-P50, PO2 at which mitochondrial respiration is 50% of maximal levels; ODFI, oxygen-dependent flux index; OXPHOS, oxidative phosphorylation.

THE JOURNAL OF EXPERIMENTAL BIOLOGY 376 The Journal of Experimental Biology 216 (3)

25 80 of mitochondria within cardiomyocytes (Gnaiger et al., 1998). Although no ultrastructural differences were immediately apparent S within hearts, analyses could not determine the relative placement 20 60 and/or arrangements of mitochondria within cardiomyocytes, which would alter diffusion distances. wet mass)

15 –1 The observed alterations in blood function and mitochondrial

(kPa) 40 2 mg function following long-term hypoxia did not correspond with O –1 P 10 ADP changes in whole-animal O uptake (i.e. SMR, RMR and MMR), s 2 2

Respiratory flux and neither did whole-animal Pcrit shift with treatment. This response 20 5 P contrasts that of a lowered Pcrit for the sailfin molly (Poecilia

G,M (pmol O latipinna) following more severe low O2 over comparable time periods (Timmerman and Chapman, 2004). However, because 0 0 0 10 20 30 40 snapper did not increase the level of circulating erythrocytes or boost Time (min) O2-uptake capacity with an increase in Hb–O2 affinity, it is perhaps unrealistic to expect a lowering of Pcrit thresholds as a result of long- Fig.3. Representative trace of the mitochondrial respirational flux assay in term hypoxia in this species. In fact, whole-animal O2-uptake saponin-permeabilised fibres that was used to determine the apparent measures and Pcrit may not actually provide the best estimate of oxygen binding affinity of mitochondria in situ using high-resolution low-O2 tolerance because O2 within the respiratory cascade is respirometry. A substrate titration protocol was employed that involved the governed by the rate of O uptake, O supply and O utilisation at −1 2 2 2 addition of the Complex I substrates glutamate (G; 10mmoll ), malate (M; the tissues. Indeed, improved hypoxia tolerance could feasibly occur 5mmoll−1) and pyruvate (P; 10mmoll−1), followed by ADP (1.25mmoll−1) without a change in P if, for example, O supply was enhanced and then succinate (S; 10mmoll−1). Fibres were permitted to respire into crit 2 through the lowering of Hb–O2 affinity and/or improved efficiencies anoxia. The black line represents the change in PO2 throughout the assay procedure (primary y-axis), with the grey line depicting the respiratory flux in the handling of O2 by mitochondria, as seen in the present study (secondary y-axis) (Fig.2, Table4). Although not investigated in the present study, changes in tissue capillarisation may have also contributed to improved low-O2 tolerance in hypoxia-acclimated snapper. Such Measures of in situ O2 utilisation identified improved low-O2 adjustments could partly explain why hypoxia-acclimated snapper performance in the cardiac mitochondria of hypoxia-acclimated occupied lower O2 levels without a change in Pcrit thresholds. snapper. Respiratory function in most teleost hearts is oxygen dependent, with a drop in O2 decreasing respiratory function in a Are anaerobic adjustments responsible for improved hypoxia near-linear manner (Forgan and Forster, 2010). Although this is tolerance? consistent with our findings, the respirational flux of cardiac tissue The red tissue groups of skeletal and cardiac muscle showed an from the hypoxia-acclimated snapper was less influenced by increased capacity for anaerobic metabolism, indicated by relatively decreasing [O2] than that of normoxia-acclimated snapper. We higher LDH:CS ratios (Table3). Elevation of these ratios suggests therefore speculate that hearts from hypoxia-acclimated snapper may that when snapper are chronically challenged by low PO2 have been able to maintain ATP synthesis at lower O2 levels. (10.2–12.1kPa), anaerobic systems may compensate for constrained Although no statistical differences in mitochondrial flux capacities aerobic ATP synthesis. Similar increases in anaerobic potential have were observed between treatments, hypoxia-acclimated snapper been observed in other species of fish exposed to periods of chronic trended higher in terms of maximal oxidative phosphorylation hypoxia (Greaney et al., 1980; Johnston and Bernard, 1982; Ton et capacity (P=0.06; Table4). This apparent increase in flux may act al., 2003; Yang et al., 1992; Zhou et al., 2000), but not in others to increase oxygen gradients and enhance flux into cardiac muscles, (Driedzic et al., 1985; Martínez et al., 2006; Zhou et al., 2000). and this effect is perhaps better observed in terms of the ODFI. Whilst these adjustments in anaerobic enzyme activity were possibly Differences in mitochondrial oxygen-uptake kinetics could also be important to snapper during moderate long-term hypoxia explained by alterations in the electron transport systems of (10.2–12.1kPa), hypoxia-acclimated fish in the behavioural trials mitochondria, including changes in the expression (amounts or also showed signs of improved low-O2 tolerance at sub-critical PO2 isoforms) of cytochrome c oxidase, and/or changes in the positioning levels (<7kPa), where fish would likely struggle to meet basal

Fig.4. Mitochondrial ultrastructure of cardiac tissue from normoxia- (A) and hypoxia-acclimated (B) treatment groups. No differences in mitochondrial surface density (predominant images) or cristae density (insets) were seen in either treatment.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Snapper responses to hypoxia acclimation 377

low-O tolerance by boosting tissue O delivery without any loss A 2 2 200 in whole-animal O2 uptake. This study therefore presents novel insight into how snapper might cope with low-O2 challenge after 150 long-term acclimation. Although the most common indicators of aerobic and anaerobic performance were screened in the present study, other physiological strategies may exist that improved the 100 hypoxia tolerance of snapper. Despite this possibility, snapper are

) undoubtedly limited in their ability to adapt to low-O2 conditions –1 50

h and do not demonstrate the same level of physiological plasticity –1 as other more hypoxia-tolerant species (Timmerman and Chapman, kg 2 0 2004; Wood and Johansen, 1972; Yang et al., 1992; Zhou et al., 200 B 2000). This presents a cause for concern, should O2 conditions within

(mg O the habitat of snapper deteriorate. 2 O .

M 150 LIST OF SYMBOLS AND ABBREVIATIONS BA behavioural arena 100 CS citrate synthase Hb haemoglobin Hct haematocrit 50 JO2 mitochondrial O2 flux LDH lactate dehydrogenase 0 MCHC mean corpuscular haemoglobin concentration 063 9 12 15 18 21 MMR maximum metabolic rate metabolic oxygen consumption P (kPa) MO2 O2 n Hill’s cooperativity coefficient RMR routine metabolic rate Fig.5. Respiratory characterisation of snapper from the normoxia- (A) and ODF oxygen-dependent flux hypoxia-acclimated (B) treatment groups at 21°C. Square symbols indicate ODFI oxygen-dependent flux index the standard metabolic rate (SMR) of the treatment determined in OEC oxygen equilibrium curve normoxia. Circular symbols represent the routine metabolic rate (RMR) of OXPHOS oxidative phosphorylation the treatment during progressive stepwise decline in ambient P . The solid O2 Hb-P P at which haemoglobin is 50% oxygen saturated horizontal line represents the SMR of the treatment group extrapolated 50 O2 Hb-P P at which haemoglobin is 100% oxygen saturated across the relevant P range while dashed horizontal lines represent the 100 O2 O2 Mt-P P at which mitochondrial respiration is 50% of maximal 95% confidence intervals of the SMR. The dotted line represents the linear 50 O2 levels regression through data points significantly below SMR, and is considered P critical oxygen partial pressure to represent points of oxy-conformance. Error bars represent 95% crit confidence intervals. PO2 partial pressure of oxygen 2 SMR standard metabolic rate TCA tricarboxylic acid metabolic demands using aerobic pathways alone. It is therefore ф Bohr coefficient plausible that the anaerobic adjustments provided a dual benefit during long-term hypoxia. For example, anaerobic ATP supply and ACKNOWLEDGEMENTS perhaps oxidative lactate metabolism would be enhanced during Technical assistance and support provided by Adrian Turner during microscopic analyses is gratefully acknowledged. Technical staff Murray Birch and Peter moderate hypoxia, but ATP generation at sub-critical PO2 might Browne fabricated the experimental choice chamber and respirometry setups. also have been improved. Unfortunately, our study provides no real John Atkins designed and built all respirometric software and hardware insight into the role of anaerobic function during either moderate components. or acute hypoxia because greater focus was placed on identifying aerobic rather than anaerobic mechanisms of adaptation. Further FUNDING investigations are therefore required to gain a deeper understanding D.G.C. would like to acknowledge support from the University of Auckland Scholarship Office for doctoral funding. of the metabolic adaptations that improve both the chronic and acute low-O tolerance of snapper and other fish species. 2 REFERENCES Behrens, J. W. and Steffensen, J. F. (2007). The effect of hypoxia on behavioural Conclusions and physiological aspects of lesser sandeel, Ammodytes tobianus (Linnaeus, 1785). Snapper exposed to moderate levels of hypoxia for 6 to 12weeks Mar. Biol. 150, 1365-1377. Brauner, C. J. and Wang, T. (1997). The optimal oxygen equilibrium curve: a showed a significant shift in their hypoxia avoidance threshold from comparison between environmental hypoxia and anemia. Am. Zool. 37, 101-108. 5.3±1.1 to 3.3±0.7kPa, a major alteration in behaviour that did not Chabot, D. and Claireaux, G. (2008). Quantification of SMR and SDA in aquatic animals using quantiles and non-linear quantile regression. Comp. Biochem. Physiol. result in greater levels of physiological perturbation (stress) after 150, S99. avoidance. The ability to endure hypoxia for longer was not Cook, D. G. and Herbert, N. A. (2012). The physiological and behavioural response of achieved through differences in swimming activity, but was likely juvenile kingfish (Seriola lalandi) differs between escapable and inescapable progressive hypoxia. J. Exp. Mar. Biol. Ecol. 413, 138-144. a result of subtle adjustments in mitochondrial uptake efficiency Cook, D. G., Wells, R. M. G. and Herbert, N. A. (2011). Anaemia adjusts the aerobic and anaerobic capacity within the cardiac and skeletal red muscle, physiology of snapper (Pagrus auratus) and modulates hypoxia avoidance behaviour during oxygen choice presentations. J. Exp. Biol. 214, 2927-2934. or other factors unidentified at this time. Although our findings are Correll, N., Sempo, G., de Meneses, Y. L., Halloy, J., Deneubourg, J. L., Martinoli, contrary to general hypotheses, we suggest that the decrease in [Hb] A. and IEEE (2006). SwisTrack: A Tracking Tool for Multi-Unit Robotic and Biological Systems. New York: IEEE. and Hb–O2 binding affinity, in combination with improved Diaz, R. J. and Rosenberg, R. (2008). Spreading dead zones and consequences for mitochondrial uptake efficiencies, might have helped to improve marine ecosystems. Science 321, 926-929.

THE JOURNAL OF EXPERIMENTAL BIOLOGY 378 The Journal of Experimental Biology 216 (3)

Drabkin, D. L. (1950). The distribution of the chromoproteins, hemoglobin, myoglobin, Pichavant, K., Maxime, V., Soulier, P., Boeuf, G. and Nonnotte, G. (2003). A and cytochrome-c, in the tissues of different species, and the relationship of the total comparative study of blood oxygen transport in turbot and sea bass: effect of chronic content of each chromoprotien to body mass. J. Biol. Chem. 182, 317-333. hypoxia. J. Fish Biol. 62, 928-937. Driedzic, W. R., Gesser, H. and Johansen, K. (1985). Effects of hypoxic adaptation Poulsen, S. B., Jensen, L. F., Nielsen, K. S., Malte, H., Aarestrup, K. and on myocardial performance and metabolism of Zoarces viviparous. Can. J. Zool. 63, Svendsen, J. C. (2011). Behaviour of rainbow trout Oncorhynchus mykiss presented 821-823. with a choice of normoxia and stepwise progressive hypoxia. J. Fish Biol. 79, 969- Dupont-Prinet, A., Chatain, B., Grima, L., Vandeputte, M., Claireaux, G. and 979. McKenzie, D. J. (2010). Physiological mechanisms underlying a trade-off between Regan, M. D. and Brauner, C. J. (2010). The evolution of Root effect hemoglobins in growth rate and tolerance of feed deprivation in the European sea bass the absence of intracellular pHprotection of the red blood cell: insights from primitive (Dicentrarchus labrax). J. Exp. Biol. 213, 1143-1152. fishes. J. Comp. Physiol. B 180, 695-706. Forgan, L. G. and Forster, M. E. (2010). Oxygen-dependence of metabolic rate in the Richards, J. G. (2009). Metabolic and molecular responses of fish to hypoxia. In muscles of craniates. J. Comp. Physiol. B 180, 715-729. Hypoxia in Fishes, Vol. 27 (ed. J. G. Richards, C. J. Brauner and A. P. Farrell), pp. Gallaugher, P., Thorarensen, H. and Farrell, A. P. (1995). Hematocrit in oxygen 443-485. London, UK: Elsevier. transport and swimming in rainbow trout (Oncorhynchus mykiss). Respir. Physiol. Sollid, J., De Angelis, P., Gundersen, K. and Nilsson, G. E. (2003). Hypoxia 102, 279-292. induces adaptive and reversible gross morphological changes in crucian carp gills. J. Gnaiger, E. (2009). Capacity of oxidative phosphorylation in human skeletal muscle: Exp. Biol. 206, 3667-3673. new perspectives of mitochondrial physiology. Int. J. Biochem. Cell Biol. 41, 1837- Timmerman, C. M. and Chapman, L. J. (2004). Behavioral and physiological 1845. compensation for chronic hypoxia in the sailfin molly (Poecilia latipinna). Physiol. Gnaiger, E., Lassnig, B., Kuznetsov, A. V., Rieger, G. and Margreiter, R. (1998). Biochem. Zool. 77, 601-610. Mitochondrial oxygen affinity, respiratory flux control and excess capacity of Ton, C., Stamatiou, D. and Liew, C.-C. (2003). Gene expression profile of zebrafish cytochrome c oxidase. J. Exp. Biol. 201, 1129-1139. exposed to hypoxia during development. Physiol. Genomics 13, 97-106. Gnaiger, E., Kuznetsov, A. V., Schneeberger, S., Seiler, R., Brandacher, G., Viriyarattanasak, C., Hamada-Sato, N., Watanabe, M., Kajiwara, K. and Suzuki, T. Steurer, W. and Margreiter, R. (2000). Mitochondria in the cold. In Life in the Cold (2011). Equations for spectrophotometric determination of relative concentrations of (ed. G. Heldmaier and M. Klingenspor), pp. 431-442. Heidelberg, Germany: myoglobin derivatives in aqueous tuna meat extracts. Food Chem. 127, 656-661. Springer. Wang, T. and Malte, H. (2011). O2 uptake and transport: the optimal P50. In Greaney, G. S., Place, A. R., Cashon, R. E., Smith, G. and Powers, D. A. (1980). Encyclopedia of Fish Physiology (ed. A. P. Farrell), pp. 893-898. San Diego, CA: Time course of changes in enzyme activities and blood respiratory properties of Academic Press. killifish during long-term acclimation to hypoxia. Physiol. Zool. 53, 136-144. Weber, R. E., Wood, S. C. and Lomholt, J. P. (1976). Temperature acclimation and Herbert, N. A. and Steffensen, J. F. (2005). The response of Atlantic cod, Gadus oxygen-binding properties of blood and multiple haemoglobins of rainbow trout. J. morhua, to progressive hypoxia: fish swimming speed and physiological stress. Mar. Exp. Biol. 65, 333-345. Biol. 147, 1403-1412. Wells, R. M. G. (2009). Blood-gas transport and hemoglobin function: adaptations for Herbert, N. A., Skjæraasen, J. E., Nilsen, T., Salvanes, A. G. V. and Steffensen, J. functional and environmental hypoxia. In Hypoxia, Vol. 27 (ed. J. G. Richards, A. P. F. (2011). The hypoxia avoidance behaviour of juvenile Atlantic cod (Gadus morhua Farrell and C. J. Brauner), pp. 256-301. London, UK: Academic Press. L.) depends on the provision and pressure level of an O2 refuge. Mar. Biol. 158, 737-746. Wells, R. M. G. and Dunphy, B. J. (2009). Potential impact of metabolic acidosis on Hickey, A. J. R. and Clements, K. D. (2003). Key metabolic enzymes and muscle the fixed-acid Bohr effect in snapper (Pagrus auratus) following angling stress. structure in triplefin fishes (Tripterygiidae): a phylogenetic comparison. J. Comp. Comp. Biochem. Physiol. 154A, 56-60. Physiol. B 173, 113-123. Wells, R. M. G., Grigg, G. C., Beard, L. A. and Summers, G. (1989). Hypoxic Hickey, A. J. R., Chai, C. C., Choong, S. Y., de Freitas Costa, S., Skea, G. L., responses in a fish from a stable environment: Blood oxygen transport in the Phillips, A. R. J. and Cooper, G. J. S. (2009). Impaired ATP turnover and ADP Antarctic fish Pagothenia borchgrevinki. J. Exp. Biol. 141, 97-111. supply depress cardiac mitochondrial respiration and elevate superoxide in nonfailing Wells, R. M. G., Baldwin, J., Seymour, R. S., Baudinette, R. V., Christian, K. and spontaneously hypertensive rat hearts. Am. J. Physiol. Cell Physiol. 297, C766-C774. Bennett, M. B. (2003). Oxygen transport capacity in the air-breathing fish, Megalops Hochachka, P. W., Stanley, C., Merkt, J. and Sumar-Kalinowski, J. (1983). cyprinoides: compensations for strenuous exercise. Comp. Biochem. Physiol. 134A, Metabolic meaning of elevated levels of oxidative enzymes in high altitude adapted 45-53. animals: an interpretive hypothesis. Respir. Physiol. 52, 303-313. Wells, R. M. G., Baldwin, J., Seymour, R. S., Christian, K. A. and Farrell, A. P. Johnston, I. A. and Bernard, L. M. (1982). Ultrastructure and metabolism of skeletal (2007). Air breathing minimizes post-exercise lactate load in the tropical Pacific muscle fibres in the tench: effects of long-term acclimation to hypoxia. Cell Tissue tarpon, Megalops cyprinoides Broussonet 1782 but oxygen debt is repaid by aquatic Res. 227, 179-199. breathing. J. Fish Biol. 71, 1649-1661. Lai, J. C. C., Kakuta, I., Mok, H. O. L., Rummer, J. L. and Randall, D. (2006). Wood, S. C. and Johansen, K. (1972). Adaptation to hypoxia by increased HbO2 Effects of moderate and substantial hypoxia on erythropoietin levels in rainbow trout affinity and decreased red cell ATP concentration. Nat. New Biol. 237, 278-279. kidney and spleen. J. Exp. Biol. 209, 2734-2738. Wood, S. C., Johansen, K. and Weber, R. E. (1975). Effects of ambient PO2 on Lenfant, C. (1973). High-altitude adaptation in mammals. Am. Zool. 13, 447-456. hemoglobin-oxygen affinity and red cell ATP concentrations in a benthic fish, Martínez, M. L., Landry, C., Boehm, R., Manning, S., Cheek, A. O. and Rees, B. B. Pleuronectes platessa. Respir. Physiol. 25, 259-267. (2006). Effects of long-term hypoxia on enzymes of carbohydrate metabolism in the Yang, T. H., Lai, N. C., Graham, J. B. and Somero, G. N. (1992). Respiratory, blood, Gulf killifish, Fundulus grandis. J. Exp. Biol. 209, 3851-3861. and heart enzymatic adaptations of Sebastolobus alascanus (Scorpaenidae; Petersen, L. H. and Gamperl, A. K. (2010). In situ cardiac function in Atlantic cod Teleostei) to the oxygen minimum zone: a comparative study. Biol. Bull. 183, 490- (Gadus morhua): effects of acute and chronic hypoxia. J. Exp. Biol. 213, 820-830. 499. Petersen, L. H. and Gamperl, A. K. (2011). Cod (Gadus morhua) cardiorespiratory Zhou, B. S., Wu, R. S. S., Randall, D. J., Lam, P. K. S., Ip, Y. K. and Chew, S. F. physiology and hypoxia tolerance following acclimation to low-oxygen conditions. (2000). Metabolic adjustments in the common carp during prolonged hypoxia. J. Fish Physiol. Biochem. Zool. 84, 18-31. Biol. 57, 1160-1171.

THE JOURNAL OF EXPERIMENTAL BIOLOGY