Hematological responses of the grey carpet ( punctatum) and the epaulette shark ( ocellatum) to anoxia and re-oxygenation

Author Chapman, CA, Renshaw, GMC

Published 2009

Journal Title Journal of Experimental Zoology

DOI https://doi.org/10.1002/jez.539

Copyright Statement © 2009 Wiley-Blackwell Publishing. This is the author-manuscript version of the paper. Reproduced in accordance with the copyright policy of the publisher.The definitive version is available at www.interscience.wiley.com

Downloaded from http://hdl.handle.net/10072/29752

Link to published version https://onlinelibrary.wiley.com/doi/10.1002/jez.539

Griffith Research Online https://research-repository.griffith.edu.au JEZ Part A: Comparative Experimental Biology

Haematological Responses of the Grey (Chiloscyllium punctatum) and the Epaulette Shark (Hemiscyllium ocellatum) to Anoxia and Re-oxygenation For PeerExposure. Review

Journal of Experimental Zoology Part A: Ecological Genetics and Journal: Physiology

Manuscript ID: JEZ-A-2008-09-0125.R1

Wiley - Manuscript type: Research Paper

Date Submitted by the n/a Author:

Complete List of Authors: Chapman, Clint; Griffith University, Physiotherapy and Exercise Science Renshaw, Gillian; Griffith University, Physiotherapy and Exercise Science

Anoxia tolerance, Re-oxygenation, Haematology, Protective Keywords: mechanisms

John Wiley & Sons Page 1 of 49 JEZ Part A: Comparative Experimental Biology

1 1 2 3 4 Haematological Responses of the Grey Carpet Shark (Chiloscyllium 5 6 punctatum ) and the Epaulette Shark ( Hemiscyllium ocellatum ) to Anoxia 7 8 and Re-oxygenation Exposure. 9 10 11 12 Authors: Clint A. Chapman and Gillian M.C. Renshaw* 13 14 and Ischemia Research Unit, 15 16 School of Physiotherapy and Exercise Science, 17 18 19 Griffith University, Gold Coast Campus, 20 For Peer Review 21 Queensland, 4222. 22 23 24 Pages: 38 25 26 Figures: 6 27 28 Tables: 2 29 30 31 Abbreviated title: Haematological responses to anoxia and re-oxygenation. 32 33 34 35 *Correspondence to: Associate Professor Gillian Renshaw 36 37 38 Hypoxia and Ischemia Research Unit, School of Physiotherapy and 39 40 Exercise Science, Griffith University, Gold Coast Campus, 41 42 Queensland, Australia 4222. 43 44 45 Phone: 61755528392; 46 47 Fax: 61755528674; 48 49 50 Email: [email protected] 51 52 53 54 Key Words : Anoxia tolerance, Reoxygenation, Haematology, Epaulette shark, Grey carpet 55 56 shark, Haematocrit, Erythrocyte, Haemoglobin, , Lactate 57 58 59 60

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 2 of 49

2 1 2 3 ABSTRACT 4 5 6 7 8 We compared the haematological responses of wild and captive populations of two 9 10 11 closely related to a standardised anoxic challenge and during a 12 hour recovery 12 13 period in normoxia: the epaulette shark ( Hemiscyllium ocellatum , Bonnaterre, 1788) and the 14 15 grey carpet shark ( Chiloscyllium punctatum , Müller and Henle, 1838). Compared to 16 17 18 normoxic controls, a significant increase in haematocrit (captive 22.3%; wild 35.9%) coupled 19 20 with a decline in meanFor corpuscular Peer haemoglobin Review concentration occurred in epaulette sharks 21 22 indicating erythrocyte swelling in response to anoxia. However, the grey carpet shark had a 23 24 25 significantly increased haematocrit (captive 27.2%; wild 29.3%), erythrocyte count (captive 26 27 37.6%; wild 46.3%) and haemoglobin concentrations (captive 31.9%; wild 31.5%), 28 29 suggesting a release of erythrocytes into the circulation and/or haemoconcentration in 30 31 32 response to anoxia. Plasma glucose concentrations were maintained in both wild and captive 33 34 epaulette sharks and in wild grey carpet sharks during anoxia but increased significantly after 35 36 37 2 hours of reoxygenation (epaulette: captive 55.8%; wild 50.1%; grey carpet shark: wild 38 39 70.3%) and remained elevated for 12 hours. Captive grey carpet sharks had an immediate 40 41 increase in plasma glucose concentrations after anoxia (96.4%), which was sustained for 12 42 43 44 hours of reoxygenation. Lactate concentrations significantly increased in captive and wild 45 46 of both species after anoxia, reaching a peak at 2 hours of reoxygenation. Both 47 48 species showed significant, yet divergent, haematological changes in response to anoxia and 49 50 51 reoxygenation, which may not only prolong their survival and assist in recovery but also, 52 53 reflect their respective ecophysiological to the extreme environments that they 54 55 inhabit. 56 57 58 59 60

John Wiley & Sons Page 3 of 49 JEZ Part A: Comparative Experimental Biology

3 1 2 3 INTRODUCTION 4 5 6 7 8 While the tolerance of most to low conditions is negligible, 9 10 11 research into the comparative physiology of species encountering hypoxia and/or anoxia in 12 13 their natural environment has revealed that some vertebrates have developed a repertoire of 14 15 specialised adaptive strategies which prolong their tolerance to reduced oxygen conditions. In 16 17 18 elasmobranchs, the hypoxia tolerance of the epaulette shark ( Hemiscyllium ocellatum ; 19 20 Bonnaterre, 1788) hasFor been well Peer characterised Review(Mulvey and Renshaw, 2000); Routley et al., 21 22 2002; Soderstrom et al., ‘99; Wise et al., ‘98). During hypoxia, this species up regulates the 23 24 25 level of anaerobic while conserving ATP usage via adaptive mechanisms such as 26 27 ventilatory and metabolic depression (Routley et al., 2002), bradycardia (Soderstrom et al., 28 29 ‘99) and neuronal hypometabolism (Mulvey and Renshaw, 2000). More recently the 30 31 32 epaulette shark has been observed to tolerate significant periods of anoxia (Renshaw et al., 33 34 2002). It has been suggested that tolerance to low oxygen conditions in this species confers 35 36 37 an adaptive advantage which enables the epaulette shark to remain active on sheltered coral 38 39 reef platforms that are cyclically exposed to severe periods of hypoxia during nocturnal low 40 41 tides (Nilsson and Renshaw, 2004; Routley et al., 2002). 42 43 44 45 46 The grey carpet shark ( Chiloscyllium punctatum ; Müller and Henle, 1838) is a close 47 48 relative to the epaulette shark, residing in the family Hemiscyllidae. The grey carpet shark 49 50 51 has a more varied distribution and is commonly found in seagrass beds, mangrove swamps as 52 53 well as in tidal pools and on coral reefs (Last and Stevens, ‘94), however, it is not commonly 54 55 found in abundance on the intermittently hypoxic reef flats inhabited by the epaulette shark. 56 57 58 While the ability of the grey carpet shark to tolerate conditions of severely reduced oxygen 59 60 saturation has not previously been examined, our preliminary experiments indicated that this

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 4 of 49

4 1 2 3 species may also have the capacity to tolerate prolonged periods of anoxia. We examined the 4 5 6 haematological response of these two closely related sharks to anoxia followed by re 7 8 oxygenation in normoxia in order to determine whether changes in haematological profiles 9 10 11 reflect the ecophysiology of each population. 12 13 14 15 It is well known that haematological modifications can provide significant advantages 16 17 18 in reducing cellular stress and prolonging survival in a range of vertebrates (Affonso et al., 19 20 2002; Soldatov, ‘96; ForSoivio et al.,Peer ‘74; Richmond Review et al., 2005). Changes in haematological 21 22 measures such as haematocrit, erythrocyte (RBC) number and haemoglobin concentrations in 23 24 25 response to hypoxia have been well characterised in fishes (Affonso et al., 2002; 26 27 Baker et al., 2005; Pearson and Stevens, ‘91; Soivio et al., ‘74; Yamamoto, ‘87; Yamamoto et 28 29 al., ‘83), although a more severe stress, such as anoxia has had little attention. Such adaptive 30 31 32 haematological changes have been attributed to a number of different mechanisms, such as 33 34 fluid shifts out of the plasma (Kirk, ‘74; Hall et al., ‘26; Swift and Loyd, ‘74), cell 35 36 37 swelling (Soivio et al., ‘74a, b) and the release of RBCs into the circulating blood from 38 39 storage organs, such as the spleen (Lai et al., 2006; Pearson and Stevens, ‘91; Yamamoto et 40 41 al., ‘83; Yamamoto, ‘87). Interestingly, no significant differences in haematocrit and RBC 42 43 44 concentrations were observed following hypoxia in the epaulette shark (Routley et al., 2002), 45 46 the spotted dogfish ( Scyliorhinus canicula ) (Butler et al., ‘79) or the bonnethead shark 47 48 (Sphyrna tiburo ) (Carlson and Parson, 2003), however elasmobranch haematological 49 50 51 responses to an anoxic insult had not previously been examined. 52 53 54 55 Increases in plasma lactate during reduced oxygen conditions are a stereotypical 56 57 58 response to an increase in anaerobic metabolism. These increases have been observed 59 60 following cyclic, acute and progressive hypoxia in elasmobranchs such as the epaulette shark

John Wiley & Sons Page 5 of 49 JEZ Part A: Comparative Experimental Biology

5 1 2 3 (Mulvey and Renshaw, 2000; Routley et al., 2002; Wise et al., ‘98) and the electric ray 4 5 6 (Torpedo marmorata ) (Hughes and Johnston, ‘78), with the increased levels being similar to 7 8 those observed in hypoxia exposed intolerant teleost fishes (Virani and Rees, 2000; Wells and 9 10 11 Baldwin, 2006). 12 13 14 15 Upregulation of anaerobic metabolism during anoxia consumes significantly larger 16 17 18 quantities of plasma glucose. Elevated plasma glucose concentrations have been reported in 19 20 anoxiatolerant vertebratesFor such asPeer Chrysemys turtles,Review the crucian carp ( Carcassius carassius ) 21 22 and in anoxia intolerant , to fuel the increases in anaerobic metabolism (Lutz and 23 24 25 Nilsson, ‘97; McDonald and Milligan, ‘92). No changes in plasma glucose concentrations 26 27 have been observed during hypoxic exposure in either the spotted dogfish ( S. canicula ) 28 29 (Butler et al. ‘79) or the epaulette shark (Routley et al., 2002). 30 31 32 33 34 Results from previous studies concluded that elasmobranchs, in general, may not 35 36 37 possess the ability to alter their haematology or plasma glucose concentrations in response to 38 39 hypoxia or anoxia, a capacity which may be evolutionarily reserved to higher vertebrates 40 41 (Lutz and Nilsson, ‘97). However previous studies on hypoxiaintolerant elasmobranchs and 42 43 44 the anoxiatolerant epaulette shark have only investigated the changes in the blood 45 46 constituents in response to hypoxic challenges. This is the first study to examine the 47 48 haematological responses of two elasmobranch species to prolonged anoxic exposure 49 50 51 followed by reoxygenation in normoxia. 52 53 54 55 In general, studies of hypoxia and anoxiatolerance have only investigated animals 56 57 58 from either wild or captive populations. The hypoxic and anoxic response of the epaulette 59 60 shark has only been examined in animals caught from their natural environment on

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 6 of 49

6 1 2 3 flats. Within these environments, animals are periodically exposed to increasingly severe 4 5 6 periods of hypoxia and it is possible that the anoxia tolerance of this species is enhanced by 7 8 preconditioning as a result of cyclic exposure to hypoxia in their natural environment 9 10 11 (Nilsson and Renshaw, 2004 review; Routley et al., 2002). Renshaw et al. (2002) observed a 12 13 preemptive induction of hypometabolism due to preconditioning 24 hours after an initial 14 15 anoxic challenge. Since animals raised in a captive environment with a constant oxygen 16 17 18 supply have not been preexposed to such preconditioning events, they provide a useful 19 20 comparison to the responsesFor of wildPeer sharks that Review are often exposed to cyclic preconditioning 21 22 provided by the epaulette shark’s natural reef environment. Therefore, the haematological 23 24 25 responses to anoxia were examined in animals caught from the wild and compared with those 26 27 of animals kept under constant environmental conditions in commercial aquaria. 28 29 30 31 32 The comparative investigation of hypoxia and anoxiatolerant animals offers an 33 34 enormous potential to provide an understanding of the underlying mechanisms involved in 35 36 37 the diverse array of strategies that specialised vertebrates have evolved to deal with oxygen 38 39 deprivation. This study examined the response of haematocrit, haemoglobin concentrations, 40 41 erythrocytes, plasma glucose and lactate levels of the epaulette shark and grey carpet shark to 42 43 44 a standardised anoxic challenge followed by reoxygenation in both captive and wild 45 46 populations. 47 48 49 50 51 52 53 54 55 56 57 58 59 60

John Wiley & Sons Page 7 of 49 JEZ Part A: Comparative Experimental Biology

7 1 2 3 MATERIALS AND METHODS 4 5 6 7 8 Study animals and locations 9 10 11 12 13 Captive epaulette sharks (n=9) and grey carpet sharks (n=9) were supplied by, 14 15 UnderWater World (Mooloolaba, Sunshine Coast) and Sea World (Main Beach, Gold Coast) 16 17 18 and the experiments were carried out on site. Epaulette sharks held in captivity were supplied 19 20 from an accredited commercialFor collectorPeer in the Review previous year had a mean length of 71.58 + 21 22 7.13 cm and weight of 1.43 + 0.36 kg. Grey carpet sharks were born in captivity 2 years 23 24 + + 25 earlier and had a mean length of 88.56 8.12 cm and weight of 2.94 0.84 kg. Both species 26 27 were kept in a constant seawater flowthrough aquarium with a normal photoperiod and no 28 29 additional light sources in their outside habitats. 30 31 32 33 34 Wild epaulette sharks (n=10) were collected from Heron Island in the Capricorn 35 36 37 bunker group (latitude 23° 27’ S, longitude 151° 55’ E) (GBRMPA Permit # GO4/12675.1) 38 39 and the experiments were conducted at the Heron Island Research Station. Wild epaulette 40 41 sharks had a mean length of 61.77 + 3.56 cm and weight of 0.55 + 0.12 kg. Wild grey carpet 42 43 44 sharks (n=6) were caught in Moreton Bay (latitude 27º 28’ S, longitude 153º 23’ E) (DPI and 45 46 Permit #PRM38182I) and experiments were conducted at the Moreton Bay 47 48 Research Station. Wild grey carpet sharks had a mean length of 92.62 + 5.51 cm and weight 49 50 + 51 of 2.73 0.47 kg. For both wild epaulette sharks and grey carpet sharks, experiments were 52 53 conducted 24 hours following capture, to allow sufficient time for the ventilatory rate, an 54 55 indication of stress levels, to return to normal without allowing time for potentially 56 57 58 significant changes in biochemical to captivity. Both captive and wild animals 59 60 were acclimated in ambient oceanic seawater at 24°C at their respective sites. Food was

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 8 of 49

8 1 2 3 withheld in both captive and wild animals, for 24 hours before the first blood samples were 4 5 6 taken at the commencement of the experiment. 7 8 9 10 11 Anoxic Challenge and Re-oxygenation 12 13 14 15 Individual animals were exposed to 1.5 hours of anoxia (experimental groups) or 16 17 18 normoxia (normoxic transfer1 for control groups) in 70 L glass tanks, followed by 12 hours 19 20 of reoxygenation afterFor transfer toPeer a separate 200Review L normoxic tank at 24°C, reoxygenation 21 22 for anoxic treated animals and normoxic transfer2 for controls. All sharks were maintained 23 24 25 in a fasted state. Each treatment tank was filled with normoxic seawater and sealed with 26 27 plastic wrap and a Perspex lid. Seawater was circulated around the tank using a submersible 28 29 228power head pump (CCC Pty. Ltd., Sydney, New South Wales, Australia) to ensure 30 31 32 uniform conditions throughout. In the anoxic tanks, nitrogen gas was bubbled through an air 33 34 stone to displace dissolved oxygen. The dissolved oxygen concentration [dO 2] levels were 35 36 37 measured and recorded each minute using TPS WP90 oxygen probes (TPS Pty. Ltd., 38 39 Brisbane, Queensland, Australia). In the anoxic tank the [dO 2] was maintained below 0.02 40 41 1 mg L . In the normoxic tanks, air was bubbled through an air stone to maintain the [dO 2] 42 43 1 44 above 7.0 mg L . Anoxic conditions were maintained within each tank by the moderate 45 46 adjustment of nitrogen injection into the tank water using brass gange valves. Animals were 47 48 closely pairmatched for length and one member from each pair was randomly selected for 49 50 51 either anoxic exposure or the normoxic control treatment. Animals were weighed and 52 53 measured at the conclusion of the experiment. 54 55 56 57 58 59 60

John Wiley & Sons Page 9 of 49 JEZ Part A: Comparative Experimental Biology

9 1 2 3 Blood Sampling and Haematology 4 5 6 7 8 Blood samples were taken rapidly from unanaesthetised fish immediately prior to the 9 10 11 beginning of the experiment (t 0) to provide a preexperimental measurement for each 12 13 haematological parameter. A 1.5 ml blood sample was taken from the dorsal aorta of each 14 15 using a heparinised syringe via the venipuncture method described in Stoskopf et al. 16 17 18 (‘84). Blood samples were then transferred to 3 ml lithium heparin vacutainers (Becton 19 20 Dickinson, North Ridge,For New SouthPeer Wales, Australia)Review and gently mixed. Additional blood 21 22 samples were collected immediately following control or experimental treatment (post/t 1.5 ) 23 24 25 then after 2 hours (t 2), 6 hours (t 6) and 12 hours (t 12 ) of reoxygenation in normoxia after 26 27 anoxic challenge or after normoxic transfer2 for controls. 28 29 30 31 32 Haematocrit was determined using heparinised haematocrit microtubes and 33 34 centrifuged at 2500 rpm for 5 minutes at 4°C in an Eppendorf 5415D centrifuge. Haematocrit 35 36 37 values were determined against a microhaematocrit grid (SigmaAldrich, Newcastle, New 38 39 South Wales, Australia). Red blood cell (RBC) concentrations [RBC] were calculated via 40 41 haemocytometry cell counts using a Neaubaur counting chamber. Blood samples were diluted 42 43 44 1:200 in chilled phosphate buffered saline (pH 7.2) and vortexed gently. The diluted solution 45 46 (5 µl) was loaded into the Neaubaur counting chamber and examined under 10 x 47 48 49 magnifications using a compound light microscope. Blood haemoglobin was determined in 50 51 triplicate, by a cyanmethaemoglobin technique using modified Drabkin’s Reagent (Sigma 52 53 Aldrich, Newcastle, New South Wales, Australia). One millilitre of Drabkin’s reagent and 4 54 55 56 µl of blood were placed into a 1.6 ml cuvette and vortexed gently. Samples were incubated in 57 58 the dark, at room temperature for 15 minutes before being analysed using a UV Mini 1240 59 60 spectrophotometer (Shimadzu, Suzhoa, Republic of China) at a wavelength of 540 nm.

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 10 of 49

10 1 2 3 Sample haemoglobin concentrations were compared against a haemoglobin standard curve 4 5 6 using the human haemoglobin standard supplied by SigmaAldrich (Newcastle, New South 7 8 Wales, Australia). 9 10 11 12 13 The RBC indices calculated were: the mean corpuscular volume (MCV), mean 14 15 corpuscular haemoglobin (MCH), mean corpuscular haemoglobin concentration (MCHC), 16 17 18 haematocrit levels and haemoglobin concentrations. The formulae of Stoskopf (‘93) were 19 20 used as follows: For Peer Review 21 22 MCV = (Hct / [RBC]) x 10, 23 24 25 MCH = (Hb x 10) / [RBC], and 26 27 MCHC = (Hb x 100) / Hct. 28 29 30 31 32 Blood plasma was separated from a 0.5 ml blood sample after it was centrifuged at 33 34 3000 rpm for 3 minutes at 4 °C then frozen at –80 °C for later use. Plasma glucose 35 36 37 concentrations were determined using a colorimetric glucose hexokinase assay kit (Sigma 38 39 Aldrich, Newcastle, New South Wales, Australia) following the manufacturer’s directions 40 41 and read at a wavelength of 340 nm. Briefly, cuvettes containing a 1:500 dilution of plasma 42 43 44 to glucose reagent were gently vortexed and incubated at room temperature for 15 minutes 45 46 before colorimetric analysis. Sample plasma glucose concentrations were compared against a 47 48 49 glucose standard curve using the glucose standard solution supplied by SigmaAldrich 50 51 (Newcastle, New South Wales, Australia). 52 53 54 55 56 Plasma lactate concentrations were determined using a colorimetric Lactate Assay kit 57 58 (Immuno Diagnostics, St. Peters, New South Wales, Australia) following the manufacturer’s 59 60 directions and read at a wavelength of 540 nm. Cuvettes containing a 1:100 dilution of

John Wiley & Sons Page 11 of 49 JEZ Part A: Comparative Experimental Biology

11 1 2 3 plasma to lactate reagent were gently vortexed and incubated at room temperature for 510 4 5 6 minutes before colorimetric analysis. Sample plasma lactate concentrations were compared 7 8 against a lactate standard curve, using the lactate standard solution supplied by Immuno 9 10 11 Diagnostics (St Peters, New South Wales, Australia). 12 13 14 15 Statistics 16 17 18 19 20 The data was analysedFor using a MixedPeer Model RepeatedReview Measures ANOVA so that we could 21 22 test the main effects and interaction effects of: sample time, population type (wild or captive) 23 24 25 and treatment (experiment or normoxic control), followed by a posthoc Bonferroni 26 27 adjustment. The alpha level was set at 0.05. The data and error bars are represented as mean + 28 29 standard deviation. 30 31 32 33 34 RESULTS 35 36 37 38 39 Haematocrit 40 41 42 43 44 For epaulette sharks, the mean haematocrit values prior to any treatment (pre 45 + + 46 experimental) were similar for captive (t 0 = 19 1.4%) and wild (t 0 = 22.8 2.8%) animals 47 48 irrespective of whether they were in control or experimental groups (Figure 1A and 1B). 49 50 51 However, there was a main effect of environment for the grey carpet shark, the mean 52 53 haematocrit values prior to any treatment were significantly higher (p<0.01) in the in wild 54 55 animal group (t = 27.3 + 2.2%) than the captive animal group (t = 20.9 + 2.5%), (Figure 1A). 56 0 0 57 58 No significant differences were observed in preexperimental mean haematocrit values of 59 60 either captive or wild grey carpet sharks allocated to the anoxic challenge group compared to

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 12 of 49

12 1 2 3 their respective control groups (Figure 1B). Furthermore, in the normoxic control groups, 4 5 6 there were no significant changes in mean haematocrit values in captive or wild animals for 7 8 either epaulette sharks or grey carpet sharks at any time points compared to their pre 9 10 11 experimental values, irrespective of sample time (Figure 1A). 12 13 14 15 Immediately following 1.5 hours of anoxic exposure there was a main effect of 16 17 18 sample time. Captive and wild epaulette sharks had a significant increase in the mean 19 + + 20 haematocrit value (captive,For t 1.5 Peer= 23.2 1.6%; Review wild, t 1.5 = 30.9 2.4%) above the pre 21 22 experimental mean measure (p<0.01), which remained at this elevated plateau for up to 2 23 24 25 hours following reoxygenation (Figure 1B). Similarly in the grey carpet shark, mean 26 27 haematocrit values increased significantly from preexperimental mean values (p<0.001), 28 29 immediately following 1.5 hours of anoxia in both captive (t = 26.6 + 1.5%) and wild (t = 30 1.5 1.5 31 + 32 35.3 2.1%) animals and remained at this elevated plateau for up to 2 hours following re 33 34 oxygenation (Figure 1B). After 6 hours normoxic reoxygenation, mean haematocrit values in 35 36 37 wild and captive epaulette sharks as well as grey carpet sharks returned to preexperimental 38 39 values and remained there for an additional 6 hours (Figure 1B). There were also interactions 40 41 with the environmental source of the sharks sampled: wild epaulette sharks had significantly 42 43 44 higher haematocrit values at the postanoxia sample point than did captive epaulette sharks 45 46 (p<0.001) (Figure 1B). However after 2 hours of reoxygenation, there were no significant 47 48 differences in mean haematocrit values between captive and wild epaulette sharks. Wild grey 49 50 51 carpet sharks also had significantly higher haematocrit values from captive animals after 52 53 anoxic exposure and after 2 hours of reoxygenation. 54 55 56 57 58 59 60

John Wiley & Sons Page 13 of 49 JEZ Part A: Comparative Experimental Biology

13 1 2 3 Red blood cell count 4 5 6 7 8 The mean [RBC] of control animals in either species did not change significantly 9 10 11 during the sample time course, data not shown. In the experimental groups, mean [RBC] 12 13 prior to treatment revealed no significant differences in captive and wild epaulette shark 14 15 + 6 1 + 6 1 (captive, t 0 = 0.34 0.06 x 10 uL ; wild, t 0 = 0.34 0.04 x 10 uL ) or grey carpet sharks 16 17 + 6 1 + 6 1 18 (captive, t 0 = 0.31 0.007 x 10 uL ; wild, t 0 = 0.3 0.03 x 10 uL ) (Figure 2). There was 19 20 an interaction of speciesFor by sample Peer time in responseReview to anoxic challenge followed by re 21 22 oxygenation. In epaulette sharks, there were no significant changes in mean [RBC] in captive 23 24 25 or wild epaulette sharks immediately following 1.5 hours of anoxia or throughout the 12 26 27 hours of normoxic reoxygenation (Figure 2) when compared to preexperimental mean 28 29 concentrations. Conversely grey carpet sharks responded to anoxia with a significant increase 30 31 + 6 1 32 in [RBC] in captive animals and wild animals (captive, t 1.5 = 0.43 0.03 x 10 uL ; wild, t 1.5 33 34 = 0.3 + 0.03 x 10 6 uL 1) from their respective preexperimental mean concentrations 35 36 37 (p<0.001). Mean [RBC] remained elevated at this plateau for 2 hours following re 38 39 oxygenation and then returned back to preexperimental concentrations by 6 hours of 40 41 normoxic reoxygenation (Figure 2). Mean [RBC] remained at preexperimental 42 43 44 concentrations throughout the remainder of the experiment (12 hours of normoxic re 45 46 oxygenation). These results paralleled the significance patterns observed in the haematocrit 47 48 measures described above and in haemoglobin concentration described below. 49 50 51 52 53 Haemoglobin 54 55 56 57 58 Prior to the commencement of the experiment, epaulette sharks from the wild 59 + 1 60 population had significantly higher haemoglobin concentrations (t0 = 6.24 0.4 g dL ) than

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 14 of 49

14 1 2 3 captive animals (t = 5.34 + 0.4 g dL 1) (Figure 3A). Similarly, the wild grey carpet sharks had 4 0 5 + 1 6 significantly higher mean haemoglobin concentrations (t0 = 7.5 0.5 g dL ) than captive grey 7 + 1 8 carpet sharks (t0 = 6.27 0.5 g dL ) prior to the commencement of the experiment (p<0.05) 9 10 11 (Figure 3A). 12 13 14 15 Throughout the remainder of the normoxic (control) sampling regime, neither 16 17 18 epaulette sharks nor grey carpet sharks from either wild or captive populations showed any 19 20 significant differencesFor in mean haemoglobinPeer concentrations Review from their respective initial mean 21 22 concentrations (Figure 3A). With regard to haemoglobin concentration there was a two way 23 24 25 interaction of species by time. The mean haemoglobin concentrations of wild and captive 26 27 epaulette sharks exposed to an anoxic challenge followed by reoxygenation, did not 28 29 significantly change from preexperimental mean concentrations at either the postanoxia 30 31 32 sample point or during the 12 hours of normoxic reoxygenation (Figure 3B). In contrast, 33 34 captive and wild grey carpet sharks had a significant increase in haemoglobin concentrations 35 36 + 1 + 1 + 1 + 37 (captive, t0 = 6.27 0.5 g dL ; t 1.5 = 8.27 0.6 g dL ; wild, t 0 = 7.5 0.5 g dL ; t 1.5 = 9.86 38 1 39 0.4 g dL ) immediately following anoxia treatment (p<0.001), which remained elevated at 40 41 this plateau for at least 2 hours of normoxic reoxygenation (Figure 3B). Furthermore, wild 42 43 44 grey carpet sharks had significantly higher haemoglobin concentrations than captive animals, 45 46 immediately after anoxia and following 2 hours of normoxic reoxygenation (p<0.001). In 47 48 both captive and wild grey carpet sharks, mean haemoglobin concentrations had returned to 49 50 51 preexperimental concentrations by 6 hours of normoxic reoxygenation, and remained at 52 53 these concentrations until the last sample point at 12 hours of reoxygenation. 54 55 56 57 58 59 60

John Wiley & Sons Page 15 of 49 JEZ Part A: Comparative Experimental Biology

15 1 2 3 Calculated erythrocytic indices 4 5 6 7 8 There were no significant differences in calculated RBC indices between untreated 9 10 11 captive and wild epaulette sharks at preexperimental mean levels (Table 1). However wild 12 13 and captive epaulette sharks showed a significant increase in mean corpuscular volume 14 15 (MCV) immediately following anoxic exposure (p<0.001), which remained elevated for the 16 17 18 first 2 hours of normoxic reoxygenation before returning back to preexperimental levels by 19 20 6 hours of normoxic reoxygenation.For Peer Conversely, Review a significant decrease in mean corpuscular 21 22 haemoglobin concentration (MCHC) occurred immediately following anoxia and for 2 hours 23 24 25 of normoxic reoxygenation in both wild (p<0.001) and captive (p<0.01) epaulette sharks, 26 27 before returning back to preexperimental levels. These significant differences were not 28 29 evident when [RBC] rather than haematocrit was used in the calculations to obtain mean 30 31 32 corpuscular haemoglobin (MCH) (Table 1). 33 34 35 36 37 No significant differences in any of the RBC indices were observed in untreated 38 39 captive or wild grey carpet sharks (Table 2) nor in wild and captive grey carpet sharks after 40 41 1.5 hours of anoxia and up to 12 hours of normoxic reoxygenation (Table 2). 42 43 44 45 46 Glucose 47 48 49 50 51 There was a significant main effect of species. Postanoxic challenge, the grey carpet 52 53 sharks had higher plasma glucose concentrations than did epaulette sharks. This was clarified 54 55 further by examining the interaction of species, sample time and population type. Throughout 56 57 58 the control regime, epaulette sharks (wild and captive) and grey carpet sharks (wild and 59 60 captive) showed no significant differences in mean plasma glucose concentrations from their

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 16 of 49

16 1 2 3 respective preexperimental mean concentrations (data not shown). The mean plasma glucose 4 5 6 concentrations in epaulette sharks and the grey carpet shark prior to treatment were not 7 + 1 8 significantly different between captive (epaulette shark, t 0 = 0.55 0.04 mg dL ; grey carpet 9 10 + 1 + 1 11 shark, t 0 = 0.68 0.04 mg dL ) or wild sharks (epaulette shark, t 0 = 0.6 0.08 mg dL ; grey 12 + 1 13 carpet sharks, t 0 = 0.67 0.06 mg dL ) nor in the preexperimental levels of sharks allocated 14 15 to anoxic or normoxic groups (Figure 4). 16 17 18 19 20 While the responseFor pattern Peer was similar Review for both species, the grey carpet sharks 21 22 responded with a plasma glucose level of a higher magnitude. Immediately following anoxic 23 24 25 exposure, no significant differences were observed in mean plasma glucose concentrations in 26 + 1 + 1 27 captive (t 1.5 = 0.64 0.09 mg dL ) and wild (t 1.5 = 0.72 0.1 mg dL ) epaulette sharks or in 28 29 the wild grey carpet sharks (t = 0.77 + 0.09 mg dL 1) compared to their preexperimental 30 1.5 31 32 mean concentrations (Figure 4). However, after 2 hours of reoxygenation, mean plasma 33 + 1 34 glucose concentrations significantly increased in both captive (t 2 = 0.85 0.09 mg dL ) and 35 36 + 1 37 wild (t 2 = 0.9 0.13 mg dL ) epaulette sharks (p<0.01) compared to preexperimental 38 39 concentrations and remained at this elevated plateau throughout the 12 hours of re 40 41 oxygenation. However in captive grey carpet sharks had significant increase mean plasma 42 43 + 1 44 glucose above preexperimental levels immediately postanoxia (t 1.5 = 1.36 0.09 mg dL ) 45 46 (p<0.001). The mean plasma glucose concentrations of captive grey carpet sharks remained at 47 48 this elevated plateau for up to 12 hours of reoxygenation. 49 50 51 52 53 A further interaction occurred between species, sample time and population revealing 54 55 that only the captive grey carpet sharks had an immediate elevation in plasma glucose levels 56 57 58 postanoxia (Figure 4). In wild grey carpet sharks, the mean plasma glucose concentrations 59 60

John Wiley & Sons Page 17 of 49 JEZ Part A: Comparative Experimental Biology

17 1 2 3 increased by a significantly greater magnitude than the epaulette shark at 2 hours of re 4 5 + 1 6 oxygenation (t 2 = 1.14 0.02 mg mL ) (p<0.001). 7 8 9 10 11 Lactate 12 13 14 15 In control animals, captive and wild epaulette sharks, along with captive grey carpet 16 17 18 sharks, showed a significant increase in mean plasma lactate concentrations with a main 19 + 1 20 effect of species (captiveFor epaulette Peer shark, t 1.5 = Review121.3 31.9 mg dL ; wild epaulette shark, t 1.5 21 22 + 1 + 1 = 145.8 37.3 mg dL ; captive grey carpet shark, t 1.5 = 71.9 52.2 mg dL ) immediately 23 24 25 following 1.5 hours of confinement in the control aquaria (p<0.001) (Figure 5A). Control 26 27 groups of epaulette sharks had significantly higher plasma lactate levels than grey carpet 28 29 sharks until 12 hours posttransfer. In both wild and captive epaulette sharks, mean plasma 30 31 32 lactate concentrations were sustained at an elevated plateau for 2 hours after transfer2, to a 33 34 normoxic tank, as a control for normoxic reoxygenation. Mean plasma lactate concentrations 35 36 37 had returned back to mean preexperimental concentrations 2 hours after transfer2, in the 38 39 captive grey carpet shark or 6 hours after transfer 2 in the captive and wild epaulette sharks. 40 41 42 43 44 The preexperimental mean plasma lactate concentrations in epaulette sharks was 45 + 1 46 significantly higher in captive animals (t 0 = 31.2 13.4 mg dL ) than in wild animals (t 0 = 2.3 47 48 + 0.5 mg dL 1) (p<0.001) (Figure 5A). However, a main effect of sample time was evident in 49 50 51 response to 1.5 hours of anoxia: epaulette sharks (captive and wild) had significant increases 52 + 53 in mean plasma lactate concentrations immediately following anoxia (captive, t 1.5 = 254.1 54 55 22.8 mg dL 1; wild, t = 242.1 + 16.1 mg dL 1) (p<0.001) (Figure 5B). By 2 hours of 56 1.5 57 58 normoxic reoxygenation, mean plasma lactate concentrations continued to increase from 59 + 1 + 60 preexperimental concentrations (captive, t 2 = 472.9 37.3 mg dL ; wild, t 2 = 359.6 43.2 mg

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 18 of 49

18 1 2 3 dL 1) (p<0.001). Epaulette sharks maintained these elevated concentrations for at least 6 4 5 6 hours of normoxic reoxygenation. By 12 hours of normoxic reoxygenation plasma lactate 7 8 concentrations had returned to 225.1 + 28.5 mg dL 1 in captive animals and 59 + 43.3 mg dL 1 9 10 11 in wild animals. 12 13 14 15 Preexperimental mean plasma lactate concentrations in grey carpet sharks were not 16 17 + 1 + 1 18 significantly different between captive (t 0 = 2.5 0.9 mg dL ) and wild (t 0 = 3.5 2.3 mg dL ) 19 20 animals irrespective ofFor whether theyPeer had been allocatedReview to normoxic control groups or anoxia 21 22 treatment groups (Figure 5A and B). The main effect of sample time was evident 23 24 25 immediately following 1.5 hours of anoxia, both captive and wild grey carpet sharks had 26 27 significant increases in mean plasma lactate concentrations post anoxic challenge (captive, t 1.5 28 29 = 166.9 + 41 mg dL 1; wild, t = 139.3 + 34.8 mg dL 1) (p<0.001). In wild grey carpet sharks, 30 1.5 31 32 the mean plasma lactate concentrations were sustained at this elevated plateau for up to 6 33 + 1 34 hours of reoxygenation (wild, t 6 = 197.2 29 mg dL ), while captive grey carpet sharks had 35 36 37 an additional significant increase in plasma lactate levels after 2 hours of reoxygenation 38 + 1 39 (captive, t 2 = 299.1 86.1 mg dL ) and remained at this elevated concentration for up to 6 40 41 hours of reoxygenation. By 12 hours of reoxygenation, mean plasma lactate concentrations 42 43 44 in the captive and wild grey carpet shark had significantly declined (p<0.001), yet they were 45 + 46 still significantly elevated above preexperimental concentrations (captive, t 12 = 59.1 mg 47 48 16.9 dL 1; wild, 92.1 + 40.7 mg dL 1) (p<0.001). 49 50 51 52 53 54 55 56 57 58 59 60

John Wiley & Sons Page 19 of 49 JEZ Part A: Comparative Experimental Biology

19 1 2 3 DISCUSSION 4 5 6 7 8 Haematological differences in control animals from captive and wild populations of two 9 10 11 species of shark. 12 13 14 15 Wild epaulette sharks as well as grey carpet sharks are intermittently exposed to 16 17 18 hypoxia in their natural environment while captive sharks at Underwater World or Sea World 19 20 do not experience hypoxia.For The meanPeer haematocrit Review values and [RBC] were similar for captive 21 22 and wild epaulette sharks in control groups prior to any treatment. However the haemoglobin 23 24 25 concentration in the wild epaulette sharks was significantly higher. Perhaps wild epaulette 26 27 sharks maintain higher haemoglobin levels in response to intermittent hypoxic challenge. 28 29 30 31 32 In grey carpet sharks, both the mean percentage haematocrit and haemoglobin 33 34 concentration were significantly higher in animals caught in the wild, it is possible that this 35 36 37 haematologically responsive species could have been exposed to the preconditioning effects 38 39 of low oxygen in their natural environment prior to capture. Last and Stephens (‘94) reported 40 41 that this species can survive out of water for extended periods of time in habitats associated 42 43 44 with tidal pools and coral reefs. We have also observed this species in the intertidal zone 45 46 associated with mangroves. The physiological compensation for such extreme challenges has 47 48 direct ecological relevance. If, for example the physiological response of the grey carpet 49 50 51 shark to a sudden stranding or decrease in ambient oxygen involves releasing a pool of 52 53 oxygenated red blood cells then oxygen delivery would receive a temporary boost. 54 55 56 57 58 59 60

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 20 of 49

20 1 2 3 Haematological responses to anoxic challenge and re-oxygenation 4 5 6 7 8 While [RBC] for epaulette sharks were within the range previously reported by 9 10 11 Baldwin and Wells (‘90), both the percentage haematocrit and haemoglobin concentrations 12 13 were higher. Such variations may be due to seasonality, animal location and/or the effect of 14 15 confinement and handling stress. Routley et al. (2002) reported no changes in haematocrit 16 17 18 levels in response to progressive hypoxia in the epaulette shark, however anoxic exposure 19 20 and at 2 hours reoxygenationFor Peer in the epaulette Review shark in the present study resulted in a 21 22 significant increased in haematocrit. The RBC size (calculated by MCV), significantly 23 24 25 increased in both captive and wild epaulette sharks, while the calculated MCHC decreased 26 27 and no changes in [RBC] or haemoglobin concentrations were observed. Taken together 28 29 these data indicate that, in the epaulette shark, an increase in RBC volume occurred 30 31 32 immediately following anoxia and was still elevated at 2 hours of reoxygenation. It is 33 34 generally accepted that such cell swelling occurs in response to reduced oxygen conditions is 35 36 + + 37 predominantly attributed to either a loss of Na /K ATPase channel function, as ATP levels 38 39 are depleted and/or a fall in extracellular pH, leading to the accumulation of intracellular 40 41 sodium. In teleost fishes, reduced extracellular pH and PO 2 cause a catecholamineinduced 42 43 + + 44 activation of βadrenergic Na /H exchangers in the RBC membrane (Fievet et al., ‘87). The 45 46 activation of Na +/H + exchangers removes H + from the RBC in exchange for extracellular 47 48 sodium. This influx of sodium is followed by chloride, which causes an osmotic influx of 49 50 51 water, effectively alkalizing the intracellular space and resulting in cell swelling (Chiocchia 52 53 and Motais, ‘89; Fievet et al., ‘87; Salama and Nikinmaa, ‘90). 54 55 56 57 58 Such increases in cell volume have been well characterised in response to reduced 59 60 oxygen conditions in teleost fishes (Chiocchia and Motais, ‘89; Fievet et al., ‘87; Salama and

John Wiley & Sons Page 21 of 49 JEZ Part A: Comparative Experimental Biology

21 1 2 3 Nikinmaa, ‘90). Although, both Tufts and Randall (‘89) and Wood et al. (‘94) reported the 4 5 + + 6 absence of a detectable βadrenergic Na /H exchanger in the anoxiaintolerant dogfish 7 8 (Squalus canicula ). However the presence and activation of H +/Na + exchangers in an anoxia 9 10 11 tolerant elasmobranch such as the epaulette shark has not been examined. 12 13 14 15 During hypoxia and anoxia, the accumulation of anaerobic metabolites such as lactate 16 17 + 18 and H can also be responsible for a decrease in extracellular pH (Jensen, 2004). In the 19 20 present study, increasesFor in RBC Peer volume in theReview epaulette shark, remained elevated at peak 21 22 levels for 2 hours following anoxia and returned to normal by 6 hours of reoxygenation in 23 24 25 normoxia. Blood lactate concentrations were significantly higher in the epaulette shark than 26 27 the lactate levels responsible for inducing RBC swelling in teleosts such as rainbow trout 28 29 (Oncorhynchus mykiss ), ( Salmo gairdneri ) and sea bass ( Morone labrax ) (Soivio et al., ‘80; 30 31 32 Thomas and Hughes, ‘82a and ‘82b; Thomas et al., ‘86). Therefore this significant increase in 33 34 lactate concentrations in the epaulette shark may reflect an extracellular acidosis, which in 35 36 37 turn could result in increases in RBC volume. Interestingly however a reduction in RBC 38 39 volume occurred during reoxygenation, when ambient oxygen levels were restored to 40 41 normal, even though plasma lactate concentrations remained elevated. These data suggest 42 43 44 that once oxygen is restored, oxygen dependent mechanisms counteract RBC swelling 45 46 (Renshaw and Nikinmaa, 2007, review). 47 48 49 50 51 Cell swelling also occurs in response to ATP depletion, resulting in a loss of ion 52 53 homeostasis (Tetens and Lykkeboe, ‘81). Erythrocyte ATP depletion causes a loss of Na +/K + 54 55 channel function, leading to the accumulation of intracellular sodium and subsequent cell 56 57 58 swelling. While many authors have discussed the interaction between organophosphates and 59 60 haemoglobinoxygen binding affinity (Smit and Hattingh, 2003; Wells et al., 2005),

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 22 of 49

22 1 2 3 controversy over the erythrocyte organophosphate concentrations in response to low oxygen 4 5 6 conditions exists among different vertebrates. While changes in erythrocyte ATP levels in 7 8 response to anoxia have not been examined in this species, depleting erythrocytic ATP 9 10 11 concentrations may be responsible for increased RBC volume as a result of a slow loss in ion 12 13 homeostasis of the RBC, which is reversed by reoxygenation. 14 15 16 17 18 In contrast to the epaulette shark, no evidence of RBC swelling occurred in the grey 19 20 carpet shark. Plasma lactateFor levels Peer rose significantly Review in grey carpet sharks from both captive 21 22 and wild populations but did not reach the high levels attained in the epaulette shark. 23 24 25 However, there were significant increases in haematocrit, as well as in [RBC] and 26 27 haemoglobin concentrations in both wild and captive grey carpet sharks immediately 28 29 following anoxia and at 2 hours of reoxygenation. These rapid increases in haematocrit, 30 31 32 haemoglobin and [RBC] could result from the action of one or more compensatory 33 34 mechanisms: a) haemoconcentration due to fluid shifts out of the plasma; b) the accelerated 35 36 37 division and maturation of blood cells already in the circulation; c) an increase in oxygenated 38 39 [RBC] due to their release from a storage organ such as the spleen, haemopoetic organs such 40 41 as the epigonal organ and Leydig organ; d) an increase in oxygenated [RBC] due to the 42 43 44 diversion of blood from muscles and/or the gastrointestinal tract. 45 46 47 48 Plasma volume shifts per se would not be an evolutionary advantage in low oxygen 49 50 51 environments and since they usually occur when the ability to maintain electrolytes 52 53 homeostasis is lost, one could expect that it heralds slow death rather than anoxia tolerance. 54 55 The grey carpet sharks used in this study remained alive for longer than one year. While fluid 56 57 58 shifts in response to severe hypoxia or anoxia have not been reported in elasmobranchs, 59 60 Cross et al. (‘69) reported that there was no change in renal function and H + excretion in

John Wiley & Sons Page 23 of 49 JEZ Part A: Comparative Experimental Biology

23 1 2 3 response to hypercapnia in the dogfish ( Squalus acanthias ). We are currently examining 4 5 6 electrolyte homeostasis in both the epaulette and the grey carpet shark postanoxic challenge. 7 8 9 10 11 Stokes and Firkin (1971) used tritiated thymidine to show that mitosis in the systemic 12 13 circulation gives rise to new thrombocytes and RBCs in the Port Jackson shark (Heterodontus 14 15 portusjacksoni ). It could be argued that the formation and/or accelerated maturation of RBCs 16 17 18 could confer an ecophysiological advantage, providing that the remained moist. Since 19 20 we observed that theFor [RBC] didPeer return to preexperimentalReview values in grey carpet sharks 21 22 during reoxygenation, it seems unlikely that the elevated [RBC] observed postanoxic 23 24 25 challenge, occurred via mitosis in the periphery and disappeared rapidly during re 26 27 oxygenation. However, if stores of oxygenated red blood cells were sequestered in storage 28 29 organs and then released or if oxygenated blood was diverted from nonessential organs such 30 31 32 as muscles and/or the gastrointestinal tract, it is conceivable that these mechanisms would 33 34 have an ecophysiological relevance for grey carpet sharks experiencing diminished oxygen 35 36 37 levels in tidal pools and estuaries. Evidence from teleost fish and some elasmobranchs 38 39 suggests that the release of red blood cells via splenic contraction does occur in response to 40 41 elevated catecholamines (Nilsson et al. ’75). 42 43 44 45 46 Splenic contraction has been well characterised in teleost fishes in response to 47 48 hypoxia (Lai et al., 2006; Pearson and Stevens, ‘91; Yamamoto et al., ‘83; Yamamoto, ‘87). 49 50 51 Butler et al. (‘79) reported no changes in RBC concentrations during hypoxia in the dogfish 52 53 (S. canicula ). However Nilsson et al. (‘75) observed a splenic contraction in response to 54 55 catecholamine and artificial nerve stimulation in both S. acanthias and S. canicula in 56 57 58 perfused and isolated spleens. In contrast, Opdyke and Opdyke (‘71) observed no release of 59 60 RBC from the spleen in S. acanthias in response to electrical stimulation of the splenic

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 24 of 49

24 1 2 3 pedicle or catecholamine infusion. Therefore, the function of the elasmobranch spleen to act 4 5 6 as a RBC reservoir is currently controversial (Nilsson et al., ‘75; Opdyke and Opdyke, ‘71). 7 8 However, the ability of the spleen to act as a RBC reservoir in an anoxiatolerant 9 10 11 elasmobranch has not been examined. 12 13 14 15 The elevated [RBC] in the grey carpet shark in response to anoxia could have arisen 16 17 18 as an adaptive ecophysiologically relevant response to the low oxygen levels periodically 19 20 encountered in mangroveFor swamps, Peer estuaries, coralReview reefs and tidal pools which are part of the 21 22 habitat that this species occupies. The increased [RBC] could confer increase oxygen 23 24 25 absorption and delivery to vital organs during bouts of naturally occurring hypoxic exposure. 26 27 This ability to increase [RBC] may also transfer an adaptive advantage during anoxia, if the 28 29 haemoglobin in the RBCs is already saturated with oxygen. 30 31 32 33 34 Glucose 35 36 37 38 39 Pre and postexperimental blood glucose concentrations in the epaulette shark and 40 41 wild grey carpet sharks were not significantly higher than other shark species such as the 42 43 44 dusky shark ( Carcharhinus obscurus ) and the nurse hound shark ( Scyliorhinus stellaris ) 45 46 (Cliff and Thurman, ‘84; Piiper et al., ‘72). However, after 2 hours of reoxygenation, plasma 47 48 glucose increased significantly by 55.6% above the anoxic level in both wild and captive 49 50 51 epaulette sharks and by 63.9% in wild grey carpet sharks, which remained elevated at these 52 53 levels for at least 12 hours of reoxygenation. Since all animals were fasted prior to and 54 55 during the experiment, such increases in plasma glucose concentrations following re 56 57 58 oxygenation suggests that glycogen stores are involved in the recovery phase, following 59 60

John Wiley & Sons Page 25 of 49 JEZ Part A: Comparative Experimental Biology

25 1 2 3 anoxic exposure, in both species. A prolonged elevation in glucose levels could potentially 4 5 6 replenish an energy debt once reoxygenation occurred via increased aerobic metabolism. 7 8 9 10 11 Elevated plasma glucose concentrations are characteristic of the anoxic response in 12 13 teleost fishes and higher vertebrates (Hardisty et al., ‘76; MacCormack et al., 2006; Wright et 14 15 al., ‘89). However, this strategy relies on glycogen stores during reduced oxygen availability 16 17 18 when anaerobic metabolism is the predominant source of ATP (ChippariGomes et al., 2005; 19 20 Baker et al., 2005). For In contrast, Peer plasma glucose Review concentrations were maintained but not 21 22 significantly elevated immediately following anoxic exposure in the wild and captive 23 24 25 epaulette sharks or in wild grey carpet sharks. 26 27 28 29 Routley et al. (2002) observed that plasma glucose levels in wild epaulette sharks 30 31 32 remained stable during exposure to progressive hypoxia. Current data confirmed that no 33 34 changes in plasma glucose concentrations occurred in response to anoxia in either wild or 35 36 37 captive epaulette sharks. Similarly, wild grey carpet sharks also showed no significant 38 39 changes in plasma glucose concentrations immediately after anoxia. Interestingly, captive 40 41 grey carpet sharks responded with a significant increase in plasma glucose concentrations 42 43 44 immediately following anoxic exposure. Even though the preexperimental levels of glucose 45 46 were not significantly different in wild and captive grey carpet sharks it was clear that the 47 48 magnitude and timing of the glucose response in captive grey carpet sharks was different than 49 50 51 wild animals. The magnitude of the significant elevation in plasma glucose levels in response 52 53 to anoxic challenge differed between the two grey carpet shark populations: after 2 hours of 54 55 reoxygenation, wild grey carpet sharks had a significant 63.9% increase while captive grey 56 57 58 carpet sharks had a 93.5% increase from preexperimental levels. Both populations remained 59 60 at this concentration for at least 12 hours of reoxygenation. Since normoxic control groups

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 26 of 49

26 1 2 3 had no increases in plasma glucose concentrations due to capture or confinement, the 4 5 6 significant increase in glucose in response to anoxia followed by reoxygenation suggest not 7 8 only that diminished oxygen may act to prime the conversion of glycogen stores to glucose in 9 10 11 both captive and wild grey carpet sharks but also that the captive group may have had a 12 13 greater response because there was a lack of natural periods of hypoxia in captive 14 15 environments and/or captive grey carpet sharks may have had larger glycogen stores due to 16 17 18 increased food availability. 19 20 For Peer Review 21 22 Wild and captive grey carpet sharks also showed a much more pronounced increase in 23 24 25 plasma glucose concentrations compared to epaulette sharks. These differences may be due to 26 27 the magnitude of the energy debt accrued in the grey carpet shark following anoxic challenge 28 29 resulting in the greater utilisation of glucose following reoxygenation than that observed in 30 31 32 the epaulette shark. 33 34 35 36 37 Lactate 38 39 40 41 In elasmobranchs, lactate concentrations rise quickly when they are exposed to acute 42 43 44 stress such as capture and handling (Cliff and Thurman, ‘84; Hoffmayer and Parsons, 2001; 45 46 Piiper and Baumgarten, ‘69; Piiper et al., ‘72), confinement (Martini, ‘74), transport (Cliff 47 48 and Thurman, ‘84) and diminished oxygen levels (Hughes and Johnston, ‘78; Mulvey and 49 50 51 Renshaw, 2000; Routley et al., 2002; Wise et al., ‘98). In the present study, control epaulette 52 53 sharks and captive grey carpet sharks showed a significant increase in plasma lactate 54 55 concentrations in response to capture, handling and confinement in a normoxic tank. These 56 57 58 increases were most likely in response to handling and confinement as reported by Wise et al. 59 60 (‘98). The preexperimental plasma lactate concentrations in captive epaulette sharks were

John Wiley & Sons Page 27 of 49 JEZ Part A: Comparative Experimental Biology

27 1 2 3 significantly higher than those of wild caught epaulette sharks, in a captive environment these 4 5 6 benthic reef sharks may have a lower activity level for two reasons: food is provided and they 7 8 don’t need to swim to ventilate their gills, such reduced aerobic activity could lead to lactate 9 10 11 accumulation. This phenomenon was not observed in captive grey carpet sharks, which 12 13 occupy both pelagic and benthic niches in their natural environment. 14 15 16 17 18 Postanoxic challenge, epaulette shark plasma lactate concentrations peaked after 2 19 20 hours of reoxygenation.For Blood Peer lactate concentration Review reached 52 mmol 1 or 40 mmol 1 in 21 22 captive and wild epaulette sharks respectively. These values are higher than those previously 23 24 25 reported for any other elasmobranchs. The peak lactate concentrations in grey carpet sharks 26 27 were not as high (33 mmol 1 in captive and 20 mmol 1 in wild sharks) as those reached in the 28 29 epaulette shark and were comparable to the peak lactate concentrations, reported for other 30 31 32 elasmobranch species in response to hypoxia (Cliff and Thurman, ‘84; Hoffmayer and 33 34 Parsons, 2001, Piiper et al., ‘72). The difference between the plasma lactate response in wild 35 36 37 and captive grey carpet sharks may indicate that anaerobic pathways were more responsive in 38 39 the captive grey carpet shark and/or that the wild population had more active pathways for 40 41 lactate clearance. 42 43 44 45 46 The high plasma lactate levels in the epaulette shark reveal that there was a significant 47 48 upregulation of anaerobic metabolism during anoxia which could be expected to prolong 49 50 51 survival. Interestingly, both captive epaulette sharks and captive grey carpet sharks had 52 53 significantly higher plasma lactate concentrations than wild animals following anoxia, 54 55 revealing that anaerobic pathways were more responsive to anoxia in captive animals than in 56 57 58 wild animals which may have been subject to natural hypoxic preconditioning in their own 59 60 environment. Such differences between wild and captive populations may be explained by

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 28 of 49

28 1 2 3 differences between the basal/standard aerobic metabolic rate and the maximum aerobic rate, 4 5 6 the metabolic scope (Fry, ‘47), under specific conditions. In response to anoxia, captive 7 8 epaulette sharks appeared to activate anaerobic pathways and accumulate lactate much earlier 9 10 11 and at a much higher concentration than wild animals that may have already been primed by 12 13 environmental exposure. In contrast, both species of captive sharks would have had 14 15 prolonged acclimatisation to a stable captive environment with a constant oxygen supply. In 16 17 18 addition, wild animals exposed to continual changes in oxygen saturation within their natural 19 20 environment may requireFor a larger Peer metabolic scope Review to delay the accumulation of anaerobic end 21 22 products. 23 24 25 26 27 Delayed peaks in plasma lactate concentrations in both the epaulette shark and the 28 29 grey carpet shark following anoxia, were similar to those observed in the sharpnose shark 30 31 32 (Rhizoprionodon terraenovae ) (Hoffmayer and Parsons, 2001), and the electric ray (T . 33 34 marmorata ) (Hughes and Johnston, ‘78) in response to hypoxia, and the nurse hound shark 35 36 37 (Scyliorhinus stellaris ) (Piiper et al., ‘72) and the dusky shark ( Carcharhinus obscurus ) (Cliff 38 39 and Thurman, ‘84) after exhaustive exercise. These delayed peaks in lactate concentrations 40 41 are attributed to slow lactate removal of poorly perfused tissues, such as white muscle fibres, 42 43 44 due to vascular shunts and the reduction of circulation to the gut and peripheral tissue (Wells 45 46 and Baldwin, ‘06). 47 48 49 50 51 Conclusions 52 53 54 55 In its natural environment, on coral reef flats the anoxiatolerant epaulette shark is 56 57 58 periodically and cyclically exposed to diminished oxygen. Previous studies have shown that 59 60 the epaulette shark initially increases anaerobic metabolism on exposure to hypoxic

John Wiley & Sons Page 29 of 49 JEZ Part A: Comparative Experimental Biology

29 1 2 3 preconditioning (Routley et al., 2002; Wise et al., ‘98) and subsequently enters into a state of 4 5 6 neuronal metabolic depression (Mulvey and Renshaw, 2000; Renshaw et al., 2002). 7 8 9 10 11 In response to a standardised anoxic challenge, increases in haematocrit and plasma 12 13 lactate occurred which paralleled increases in erythrocyte volume and anaerobic metabolism 14 15 respectively. Significant increases in plasma lactate is a typical response to anoxic exposure 16 17 18 in both teleosts and elasmobranch species, however lactate concentrations in the epaulette 19 20 shark were the highestFor ever to bePeer reported in Review an elasmobranch, indicating not only a high 21 22 anaerobic capacity but also a high tolerance to the accumulation of metabolic end products 23 24 25 such as lactate and associated acidosis. This study is the first report of a change in 26 27 haematocrit due to erythrocyte swelling in response to anoxia followed by reoxygenation in 28 29 an elasmobranch species, the epaulette shark, which is of potential ecophysiological 30 31 32 relevance because erythrocyte swelling assists oxygen offloading. 33 34 35 36 37 The grey carpet shark showed significant haematological responses to anoxic 38 39 exposure that differed from the epaulette shark. This study is the first to demonstrate that the 40 41 grey carpet shark can tolerate an extended anoxic challenge and to report a significant 42 43 44 increase in [RBC] in any elasmobranch, immediately following anoxia. It is suggested that 45 46 this increase in [RBC] may be due to a release of erythrocytes into the circulation, due to 47 48 splenic contraction or a reduction of the blood supply to non vital organs and the 49 50 51 musculature. Alternatively, it is possible that haemoconcentration alone or in combination 52 53 with RBC release results in the significant increase in RBC concentration in grey carpet 54 55 sharks in response to anoxic challenge. 56 57 58 59 60

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 30 of 49

30 1 2 3 During anoxia, plasma lactate concentrations significantly increased in both species, 4 5 6 indicating the upregulation of anaerobic metabolism. Additionally, both species maintained 7 8 plasma glucose concentrations, which indicate a very sensitive regulation of blood glucose 9 10 11 levels. Following reoxygenation, this study clearly shows for the first time, two 12 13 elasmobranch species entering a phase of hyperglycaemia during normoxic reoxygenation 14 15 following anoxia. The hyperglycaemic state during reoxygenation could potentially reflect 16 17 18 an upregulation of aerobic metabolism, which could serve to rapidly replenish ATP levels. 19 20 For Peer Review 21 22 While these elasmobranch species are closely related, they demonstrate different 23 24 25 ecophysiological adaptations that would confer an advantage during severe declines in 26 27 oxygen availability. The epaulette shark significantly up regulates the level of anaerobic 28 29 metabolism to generate ATP. In addition there was an increased RBC volume, which may 30 31 32 increase oxygen unloading from the erythrocytes. Similarly, the increase in plasma lactate in 33 34 the grey carpet shark indicates an upregulation of anaerobic metabolism. In contrast to the 35 36 37 epaulette shark, the grey carpet shark rapidly increased erythrocyte number in response to 38 39 anoxic challenge, indicating a redirection of oxygen stores during periods of oxygen deficit. 40 41 Both species demonstrate ecophysiologically relevant adaptations which may prolong 42 43 44 survival in the intermittently hypoxic or even briefly anoxic environment encountered in tidal 45 46 pools, mangrove swamps, and in some estuaries and coral reef environments. 47 48 49 50 51 52 53 54 55 56 57 58 59 60

John Wiley & Sons Page 31 of 49 JEZ Part A: Comparative Experimental Biology

31 1 2 3 ACKNOWLEDGEMENTS 4 5 6 7 8 We would like to thank Andreas Fischer and UnderWater World (Sunshine Coast, Australia), 9 10 11 Trevor Long and Marine Horton at Sea World (Gold Coast, Australia) for supplying animals. 12 13 Acknowledge the help of the staff at Heron Island Research Station for facilitating the 14 15 fieldwork and Blake Harahush for her assistance with experiments. 16 17 18 19 20 For Peer Review 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 32 of 49

32 1 2 3 LITERATURE CITED 4 5 6 7 Affonso EG, Polez VLP, Correa CF, Mazon AF, Araujo MRR, Moraes G, Rantin FT. 2002. 8 9 Blood parameters and metabolites in the teleost fish Colossoma macropomum 10 11 exposed to sulfide or hypoxia. Comp Biochem Physiol C 133:375382. 12 Baker DW, Wood AM, Kieffer JD. 2005. Juvenile Atlantic and shortnose sturgeons (Family 13 14 Acipenseridae) have different hematological responses to acute environmental 15 16 hypoxia. Physiol Biochem Zool 78:916925. 17 18 Baldwin J, Wells RMG. 1990. Oxygen transport potential in tropical elasmobranchs from the 19 : Relationship between haematology and blood viscosity. J Exp 20 For Peer Review 21 Mar Bio and Ecol 144:145155. 22 23 Bourne PK. 1986. Changes in haematological parameters associated with capture and 24 25 captivity of the marine teleost, Pleuronectes platessa L. Comp Biochem Physiol A 26 85:43543. 27 28 Bracewell P, Cowx IG, Uglow RF. 2004. Effects of handling and electrofishing on plasma 29 30 glucose and whole blood lactate of Leuciscus cephalus . J Fish Biol 64:6571. 31 32 Butler P J, Taylor EW, Davison W. 1979. The effect of long term, moderate hypoxia on acid 33 34 base balance, plasma catecholamines and possible anaerobic end products in the 35 unrestrained dogfish Scyliorhinus canicula . J Comp Physiol B 132:297303. 36 37 Carlson JK, Parsons GR. 2003. Respiratory and haematological responses of the bonnetthead 38 39 shark, Sphyrna tiburo , to acute changes in dissolved oxygen. J Exp Mar Biol Ecol 40 41 294:1526. 42 Chiocchia G, Motais R. 1989. Effect of catecholamine on deformability of red cells from 43 44 trout: relative roles of cyclic AMP and cell volume. J Physiol (Paris) 412:321332. 45 46 ChippariGomes AR, Gomes LC, Lopes NP, Val AL, AlmeidaVal VMF. 2005. Metabolic 47 48 adjustments in two Amazonian cichlids exposed to hypoxia and anoxia. Comp 49 50 Biochem Physiol B 141:347355. 51 Cliff G, Thurman GD. 1984. Pathological and physiological effects of stress during capture 52 53 and transport in juvenile dusky shark, Carcharhinus obscurus . Comp Biochem. 54 55 Physiol A 78:167173. 56 57 Dick PT, Dixon DG. 1985. Changes in circulating blood cell levels of rainbow trout, Salmo 58 gairdneri (Richardson), following acute and chronic exposure to copper. J Fish Biol 59 60 26:475481.

John Wiley & Sons Page 33 of 49 JEZ Part A: Comparative Experimental Biology

33 1 2 3 Fange R, JohanssonSjobeck ML. 1975. The effect of splenectomy on the hematology and on 4 5 the activity of deltaaminolevulinic acid dehydratase (ALAD) in hemopoietic tissues 6 7 of the dogfish, Scyliorhinus canicula (). Comp Biochem Physiol A 8 9 52:57780. 10 + 11 Fievet B, Motais R, Thomas S. 1987. Role of adrenergicdependent H release from red cells 12 in acidosis induced by hypoxia in trout. Am J Physiol 252:R269275. 13 14 Fry FEJ. (1947). Effects of the environment on animal activity. University of Toronto Studies 15 16 Biological Series No 55, Publication of the Ontario Fish Research Laboratory no 17 18 68:162. 19 Hall FG, Gray IE, Lepkovsky S. 1926. The influence of asphyxiation on the blood 20 For Peer Review 21 constituents of marine fishes. J Biol Chem 67:549554. 22 23 Hardisty MW, Zelnik PR, Wright VC. 1976. The effects of hypoxia on blood sugar levels and 24 25 on the endocrine pancreas, interrenal, and chromaffin tissues of the lamprey, 26 Lampetry fluviatilis (L.). Gen Comp Endo 28:184204. 27 28 Hoffmayer ER, Parsons GR. 2001. The physiological response to capture and handling stress 29 30 in Atlantic sharpnose shark, Rhizoprionodon terraenovae . Fish Physiol Biochem 31 32 25:277285. 33 34 Hughes GM, Johnston IA. 1978. Some responses of the electric ray ( Torpedo marmorata ) to 35 low ambient oxygen tensions. J Exp Biol 73:107117. 36 37 Jensen FB. 2004. Red blood cell pH, the Bohr effect, and other oxygenationlinked 38 39 phenomena in blood O 2 and CO 2 transport. Act Physiol Scand 182:215227. 40 41 Lai JC, Kakuta I, Mok HO, Rummer JL, Randall D. 2006. Effects of moderate and 42 substantial hypoxia on erythropoietin levels in rainbow trout kidney and spleen. J Exp 43 44 Biol 209:27348. 45 46 Last PR, Stevens JD. 1994. Sharks and Rays of Australia. CSIRO Division of Fisheries. 47 48 Melbourne, Australia. 49 50 Lutz PL, Nilsson GE. 1997. Contrasting strategies for anoxic brain survivalglycolysis up or 51 down. J Exp Biol 200:411419. 52 53 Maccormack TJ, Lewis JM, AlmeidaVal VM, Val AL, Driedzic WR. 2006. Carbohydrate 54 55 management, anaerobic metabolism, and adenosine levels in the armoured catfish, 56 57 Liposarcus pardalis (Castelnau), during hypoxia. J Exp Zool 305:36375. 58 Martini, FF. 1974. Effects of capture and fasting confinement on an elasmobranch, Squalus 59 60 acanthias. Unpublished Ph.D. Thesis, New York: Cornell University.

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 34 of 49

34 1 2 3 McDonald DG, Milligan CL. 1992. Chemical properties of the blood. In: Hoar W.S., Randall 4 5 DJ, Farrell AP, editors. Fish Physiology, XIIB. London: Academic Press, p 56–133. 6 7 Milligan CL, Wood CM. 1987. Regulation of blood oxygen transport and red cell pHi after 8 9 exhaustive activity in rainbow trout ( Salmo gairdneri ) and starry flounder ( Platichthys 10 11 stellatus ). J Exp Biol 133:263282. 12 Montero D, Tort L, Robaina L, Vergara JM, Izquierdo MS. 2001. Low vitamin E in diet 13 14 reduces stress resistance of gilthead seabream ( Sparus aurata ) juveniles. Fish & 15 16 Shellfish Immuno 11:473490. 17 18 Mulvey JM, Renshaw GMC. 2000. Neuronal oxidative hypometabolism in the brainstem of 19 the epaulette shark ( Hemiscyllium ocellatum ) in response to hypoxic preconditioning. 20 For Peer Review 21 Neuroscience Letters 290:14. 22 23 Nilsson S, Holmgren S, Grove DJ. 1975. Effects of drugs and nerve stimulation on the spleen 24 25 and arteries of two species of dogfish, Scyliorhinus canicula and Squalus acanthias . 26 Act Physiol Scand 95:21930. 27 28 Opdyke DF, Opdyke NE. 1971. Splenic responses to stimulation in Squalus acanthias . Am J 29 30 Physiol 221:623625. 31 32 Pearson MP, Stevens ED. 1991. Size and hematological impact of the splenic erythrocyte 33 34 reservoir in rainbow trout, Oncorhynchus mykiss . Fish Physiol Biochem 9:3950. 35 Perry S, Gilmour K. 1996. Consequences of catecholamine release on ventilation and blood 36 37 oxygen transport during hypoxia and hypercapnia in an elasmobranch Squalus 38 39 acanthias and a teleost Oncorhynchus mykiss. J Exp Biol 199:21052118. 40 41 Piiper J, Baumgarten D. 1968. Analysis of brachial gaseous exchange in an elasmobranch 42 fish: Scyliorhinus stellaris . J Physiol (Paris) 60:518. 43 44 Renshaw GMC, Dyson SE. 1999. Increased nitric oxide sythase in the vasculature of the 45 46 epaulette shark brain following hypoxia. NeuroReport 10:17071712. 47 48 Renshaw, GMC, Kerrisk, CB, Nilsson, GE. 2002. The role of adenosine in the anoxic 49 50 survival of the Epaulette shark, Hemiscyllium ocellatum . Comp Biochem Physiol B 51 131:133141. 52 53 Renshaw GMC, Nikinmaa M. Oxygen sensors of the peripheral and central nervous system. 54 55 2007. In: D. Johnson, editor. Handbook of Neurochemistry and Molecular 56 57 Neurobiology, 3rd edition, 20, Sensory Neurochemisty, Springer, New York, p 272 58 296. 59 60

John Wiley & Sons Page 35 of 49 JEZ Part A: Comparative Experimental Biology

35 1 2 3 Richmond JP, Burns JM, Rea LD, Mashburn, KL. 2005. Postnatal ontogeny of erythropoietin 4 5 and hematology in freeranging Steller sea lions ( Eumetopias jubatus ). Gen Comp 6 7 Endo 141:240247. 8 9 Routley MH, Nilsson GE, Renshaw GMC. 2002. Exposure to hypoxia primes the respiratory 10 11 and metabolic responses of the Epaulette shark to progressive hypoxia. Comp 12 Biochem Physiol A 131:313321. 13 14 Salama A, Nikinmaa, M. 1990. Effect of oxygen tension on catecholamineinduced formation 15 16 of cAMP and on swelling of carp red blood cells. Am J Physiol 259:C723726. 17 18 Smith HW. 1931. The absorption and excretion of water and salts by the elasmobranch 19 fishes. II Marine elasmobranchs. Am J Physiol 98:296310. 20 For Peer Review 21 Soderstrom V, Renshaw GMC, Nilsson GE. 1999. Brain blood flow and blood pressure 22 23 during hypoxia in the Epaulette shark Hemiscyllium ocellatum , a Hypoxiatolerant 24 25 Elasmobranch. J Exp Biol 202:829835. 26 Soivio A, Nikinmaa M, Westman K. 1980. The blood oxygen binding properties of hypoxic 27 28 Salmo gairderi . J Comp Physiol 136:8387. 29 30 Soivio A, Westman K, Nyholm K. 1974a. Changes in haematocrit values in blood samples 31 32 treated with and without oxygen; a comparative study with four salmonid species. J 33 34 Fish Biol 6:763. 35 Soivio A, Westman K, Nyholm K. 1974b. The influence of changes in oxygen tension on the 36 37 haematocrit value of blood samples from asphyxic rainbow trout ( Salmo gairdneri ). 38 39 Aqua 3:395401. 40 41 Soldatov AA. 1996. The effect of hypoxia on red blood cells of flounder: a morphologic and 42 autoadiographic study. J Fish Biol 48:321328. 43 44 Stokes EE, Firkin BG. 2008. Studies of the Peripheral Blood of the Port Jackson Shark 45 46 (Heterodontus portusjacksoni ) with Particular Reference to the Thrombocyte. Br J 47 48 Haematol. 1971 Apr;20(4):42735 49 50 Stoskopf MK. 1993. Clinical pathology. In fish medicine. W.B. Saunders, Philadelphia 51 882pp . 52 53 Stoskopf MK, Smith B, Klay G. 1984. Clinical note: Blood sampling of captive sharks. J 54 55 Zool Anat Med 15:116117. 56 57 Swift DJ, Lloyd R. 1974. Changes in urine flow rate and haematocrit value of rainbow trout 58 Salmo gairdneri (Richardson) exposed to hypoxia. J. Fish Biol. 6:379387. 59 60

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 36 of 49

36 1 2 3 Tetens V, Lykkeboe G. 1981. Blood respiratory properties of rainbow trout, Salmo gairdneri : 4 5 Responses to hypoxia acclimation and anoxic incubation of blood in vitro. J Comp 6 7 Physiol B 145:117125. 8 9 Thomas S, Fievet B, Motais R. 1986. Effect of deep hypoxia on acidbase balance in trout: 10 11 role of ion transfer processes. Am J Physiol Regul Integr Comp Physiol 250:R319 12 R327 13 14 Thomas S, Hughes GM. 1982a. Effects of hypoxia on blood gas and acidbase parameters of 15 16 sea bass. J Appl Physiol 53:13361341. 17 18 Thomas S, Hughes, GM. 1982b. A study of the effects of hypoxia on acidbase status of 19 rainbow trout blood using an extracorporeal blood circulation. Resp Physiol 49:371 20 For Peer Review 21 82. 22 23 Thorson T B, Gerst JW. 1973. Comparison of some parameters of serum and uterine fluid of 24 25 pregnant, vivparous sharks ( Carcharhinus leucas ) and serum of their nearterm 26 young. Comp Biochem Physiol A 42:3340. 27 28 Tufts BL, Randall DJ. 1989. The functional significance of adrenergic pH regulation in fish 29 30 erythrocytes. Can J Zool 67:235238. 31 32 Virani NA, Rees BB. 2000. Oxygen consumption, blood lactate and interindividual variation 33 34 in the gulf killifish, Fundulus grandis , during hypoxia and recovery. Comp Biochem 35 Physiol A 126:397405. 36 37 Waring CP, Stag RM, Poxton MG. 1992. The effects of handling on flounder ( Platichthys 38 39 flesus L ) and Atlantic Salmon ( Salmo salar L ). J Fish Biol 41:131144. 40 41 Wells RMG, Baldwin J. 2006. Plasma lactate and glucose flushes following burst swimming 42 in silver trevally ( Pseudocaranx dentex : Carangidae) support the "releaser" 43 44 hypothesis. Comp Biochem Physiol A 143:347352. 45 46 Wise G, Mulvey JM, Renshaw GMC. 1998. Hypoxia Tolerance in the Epaulette Shark 47 48 (Hemiscyllium ocellatum ). J Exp Zool 281, 15. 49 3 50 Wood CM, Perry SF, Walsh PJ, Thomas S. 1994. HCO dehydration by the blood of an 51 elasmobranch in the absence of a Haldane effect. Resp Physiol 98:31937. 52 53 Wright PA, Perry SF, Moon TW. 1989. Regulation of hepatic glucogenesis and 54 55 glycogenolysis by catecholamines in rainbow trout during environmental hypoxia. J. 56 57 Exp. Biol. 147:169188. 58 Yamamoto K. 1987. Contraction of spleen in exercised cyprinid. Comp Biochem Physiol A 59 60 87:10877.

John Wiley & Sons Page 37 of 49 JEZ Part A: Comparative Experimental Biology

37 1 2 3 Yamamoto K, Itazawa Y, Kobayashi H. 1983. Erythrocyte supply from the spleen and 4 5 hemoconcentration in hypoxic yellowtail. Mar Biol 73:221226. 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 For Peer Review 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 38 of 49

38 1 2 3 Table 1. Erythrocyte indices calculated for the Epaulette shark ( Hemiscyllium 4 5 6 ocellatum). 7 8 Captive Wild 9 10 Pre Post 2 Hrs 6 Hrs 12 Hrs Pre Post 2 Hrs 6 Hrs 12 Hrs 11 MCV 556.5 669.4 663.7 563.8 525.7 620.6 772 799.4 688.3 675.4 Anoxia † 12 (ƒ L) (37) (84) * (62) (46) (67) (60) (48) (80) (55) (43) 155 157.9 155.2 159.9 157.9 171.7 183.6 181.7 173 13 MCH 167 (9) 14 (15) (10) (19) (25) (13) (30) (24) (19) (15) MCHC 27.9 23.8 23.4 27.1 26.4 15 1 † 28.4 (2) 21.7 (2) 22.9 (2) 27.5 (2) 28.0 (2) (g dL ) (2) (1) (3) 16 (2) * (2) 583.7 524.5 632.3 630.5 636.5 722.7 649 632.7 688.1 645.6 17 Control MCV (ƒ L) (109) (41) (41) (50) (74) (85) (61) (49) (8) (92) 18 166.7 148.4 171.2 167 163.3 197.3 180.3 174.6 214.8 188.1 MCH 19 (33) (19) (15) (11) (22) (21) (12) (30) (13) (6) 20 MCHC 28.5 28.2 27.1 26.5 25.7 1 For Peer Review 27.4 (3) 27.8 (1) 27.5 (4) 31.2 (2) 29.5 (3) (g dL ) 21 (1) (2) (1) (2) (3) 22 Erythrocytic indices: mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH) and mean 23 corpuscular haemoglobin concentration (MCHC) calculated for captive (n=9) and wild (n=10) epaulette sharks, 24 25 pre and postanoxic challenge and during 12 hours of normoxic reoxygenation. The figure in brackets () is the 26 standard deviation. 27 28 29 Table 2. Erythrocyte indices calculated for the Grey carpet shark ( Chiloscyllium 30 31 32 punctatum ). 33 34 Captive Wild 35 36 Pre Post 2 Hrs 6 Hrs 12 Hrs Pre Post 2 Hrs 6 Hrs 12 Hrs MCV 666.3 626 691.8 651.8 699.6 910.8 809.5 877.2 930 941.9 37 Anoxia 38 (ƒ L) (103) (52) (78) (97) (91) (118) (109) (56) (60) (50) 200.8 195.1 196.1 202.2 211.2 246.2 225.6 241 255.6 242.9 MCH 39 (29) (25) (14) (28) (28) (23) (25) (12) (33) (22) 40 MCHC 30.3 31.2 28.5 31.1 30.2 27.1 (1) 27.9 (1) 27.5 (2) 27.7 (2) 25.7 (3) 41 (g L 1) (3) (3) (2) (2) (2) MCV 686.2 687.9 669.3 687.6 643 917.4 923.5 997 941.3 953.5 42 Control (ƒ L) (82) (96) (74) (57) (59) (40) (50) (143) (49) (122) 43 204.7 196.3 190.9 202.1 203.4 254.9 245.2 274.6 252.5 251.2 MCH 44 (10) (13) (13) (21) (25) (13) (21) (33) (7) (37) 45 MCHC 28.8 28.7 29.4 31.6 1 30 (2) 27.8 (1) 26.5 (2) 27.6 (1) 26.9 (1) 26.3 (1) 46 (g L ) (3) (3) (1) (1) 47 Erythrocytic indices: mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH) and mean 48 corpuscular haemoglobin concentration (MCHC) calculated for captive (n=9) and wild (n=6) grey carpet sharks, 49 50 pre and postanoxic challenge and during 12 hours of normoxic reoxygenation. The figure in brackets () is the 51 standard deviation 52 53 54 55 56 57 58 59 60

John Wiley & Sons Page 39 of 49 JEZ Part A: Comparative Experimental Biology

39 1 2 3 Figure legends. 4 5 6 7 8 Figure 1A. Mean haematocrit values for epaulette sharks and grey carpet sharks in two 9 10 11 different environments, held in normoxia (controls). Transfer 1 corresponds to the time 12 13 period that experimental animals were in anoxia and transfer 2 corresponds to the time course 14 15 of reoxygenation in normoxia for experimental animals. The symbol * indicates a 16 17 18 significantly higher initial value compared to captive sharks. No significant changes were 19 20 observed within controlFor groups atPeer any of the sample Review times. 21 22 23 24 25 Figure 1B. Mean haematocrit values for the epaulette shark and the grey carpet shark before 26 27 and immediately following 1.5 hours of anoxia and at 2 hours, 6 hours and 12 hours of re 28 29 oxygenation in normoxia. No significant changes were observed in any of the control groups. 30 31 32 The symbols indicate significant differences compared to the mean * preexperiment or † at 2 33 34 hours following anoxic challenge. 35 36 37 38 39 Figure 2. Mean red blood cell counts for the epaulette shark and the grey carpet shark before 40 41 and immediately following 1.5 hours of anoxia and at 2 hours, 6 hours and 12 hours of re 42 43 44 oxygenation in normoxia. No significant changes were observed in any of the control groups. 45 46 Symbols indicate significant differences compared to the mean * preexperiment or † at 2 47 48 hours following anoxic challenge. 49 50 51 52 53 Figure 3A. 54 55 Mean haemoglobin for epaulette sharks and grey carpet sharks in two different environments, 56 57 58 held in normoxia (controls). Transfer 1 corresponds to the time period that experimental 59 60 animals were in anoxia and transfer 2 corresponds to the time course of reoxygenation in

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 40 of 49

40 1 2 3 normoxia for experimental animals. The symbol * indicates a significantly higher initial 4 5 6 value compared to captive sharks. No significant changes were observed within control 7 8 groups at any of the sample times. 9 10 11 12 13 Figure 3B. Mean haemoglobin for the epaulette shark and the grey carpet shark before and 14 15 immediately following 1.5 hours of anoxia and at 2 hours, 6 hours and 12 hours of re 16 17 18 oxygenation in normoxia. No significant changes were observed in any of the control groups. 19 20 Symbols indicate significantFor differences Peer compared Review to the mean * preexperiment or † at 2 21 22 hours following anoxic challenge. 23 24 25 26 27 Figure 4. Mean plasma glucose for the epaulette shark and the grey carpet shark before and 28 29 immediately following 1.5 hours of anoxia and at 2 hours, 6 hours and 12 hours of re 30 31 32 oxygenation in normoxia. No significant changes were observed in any of the control groups. 33 34 Symbols indicate significant differences compared to the mean * preexperiment, † at 2 35 36 37 hours, #6 hours, or × 12 hours following anoxic challenge. 38 39 40 41 Figure 5A. Mean plasma lactate for epaulette sharks and grey carpet sharks in two different 42 43 44 environments, held in normoxia (controls). Transfer 1 corresponds to the time period that 45 46 experimental animals were in anoxia and transfer 2 corresponds to the time course of 47 48 reoxygenation in normoxia for experimental animals. Symbols indicate significant 49 50 51 differences compared to the mean * prenormoxic transfer 2 and † at 2 hours after transfer 2. 52 53 54 55 Figure 5B. Mean plasma lactate of the epaulette shark and the grey carpet shark. Mean 56 57 58 plasma lactate concentrations before and immediately following 1.5 hours of anoxia and at 2 59 60 hours, 6 hours and 12 hours of reoxygenation in normoxia. Symbols indicate significant

John Wiley & Sons Page 41 of 49 JEZ Part A: Comparative Experimental Biology

41 1 2 3 differences compared to the mean * preexperiment, † at 2 hours, # 6 hours, or × 12 hours 4 5 6 following anoxic challenge. 7 8 9 10 11 12 13 14 15 16 17 18 19 20 For Peer Review 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 42 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 Figure 1A. Mean haematocrit values for epaulette sharks and grey carpet sharks in two different 32 environments, held in normoxia (controls). Transfer 1 corresponds to the time period that experimental animals were in anoxia and transfer 2 corresponds to the time course of 33 reoxygenation in normoxia for experimental animals. The symbol * indicates a significantly higher 34 initial value compared to captive sharks. No significant changes were observed within control groups 35 at any of the sample times. 36 297x209mm (300 x 300 DPI) 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 John Wiley & Sons Page 43 of 49 JEZ Part A: Comparative Experimental Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 1B. Mean haematocrit values for (a) the epaulette shark and (b) the grey carpet shark b efore 33 and immediately following 1.5 hours of anoxia and at 2 hours, 6 hours and 12 hours of re- 34 oxygenation in normoxia. No significant changes were observed in any of the control groups. The 35 symbols indicate significant differences compared to the mean * pre-experiment or at 2 hours 36 following anoxic challenge. 297x209mm (300 x 300 DPI) 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 44 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 2. Mean red blood cell counts for the epaulette shark and the grey carpet shark before and 33 immediately following 1.5 hours of anoxia and at 2 hours, 6 hours and 12 hours of re-oxygenation 34 in normoxia. No significant changes were observed in any of the control groups. Symbols indicate 35 significant differences compared to the mean * pre-experiment or at 2 hours following anoxic 36 challenge. 297x209mm (300 x 300 DPI) 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 John Wiley & Sons Page 45 of 49 JEZ Part A: Comparative Experimental Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 3A. Mean haemoglobin for epaulette sharks and grey carpet sharks in two different 33 environments, held in normoxia (controls). Transfer 1 corresponds to the time period that 34 experimental animals were in anoxia and transfer 2 corresponds to the time course of 35 reoxygenation in normoxia for experimental animals. The symbol * indicates a significantly higher initial value compared to captive sharks. No significant changes were observed within control groups 36 at any of the sample times. 37 297x209mm (300 x 300 DPI) 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 46 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 3B. Mean haemoglobin for (a) the epaulette shark and (b) the grey carpet shark before and 33 immediately following 1.5 hours of anoxia and at 2 hours, 6 hours and 12 hours of re-oxygenation 34 in normoxia. No significant changes were observed in any of the control groups. Symbols indicate 35 significant differences compared to the mean * pre-experiment or at 2 hours following anoxic 36 challenge. 297x209mm (300 x 300 DPI) 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 John Wiley & Sons Page 47 of 49 JEZ Part A: Comparative Experimental Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 4. Mean plasma glucose for the epaulette shark and the grey carpet shark before and 33 immediately following 1.5 hours of anoxia and at 2 hours, 6 hours and 12 hours of re-oxygenation 34 in normoxia. No significant changes were observed in any of the control groups. Symbols indicate 35 significant differences compared to the mean * pre-experiment, at 2 hours, #6 hours, or × 12 36 hours following anoxic challenge. 297x209mm (300 x 300 DPI) 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 John Wiley & Sons JEZ Part A: Comparative Experimental Biology Page 48 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 5A. Mean plasma lactate for epaulette sharks and grey carpet sharks in two different 33 environments, held in normoxia (controls). Transfer 1 corresponds to the time period that 34 experimental animals were in anoxia and transfer 2 corresponds to the time course of 35 reoxygenation in normoxia for experimental animals. Symbols indicate significant differences 36 compared to the mean * pre-normoxic transfer 2 and at 2 hours after transfer 2. 297x209mm (300 x 300 DPI) 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 John Wiley & Sons Page 49 of 49 JEZ Part A: Comparative Experimental Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 5B. Mean plasma lactate of the (a) epaulette shark and the (b) grey carpet shark. Mean 33 plasma lactate concentrations before and immediately following 1.5 hours of anoxia and at 2 hours, 34 6 hours and 12 hours of re-oxygenation in normoxia. Symbols indicate significant differences 35 compared to the mean * pre-experiment, at 2 hours, # 6 hours, or × 12 hours following anoxic 36 challenge. 297x209mm (300 x 300 DPI) 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 John Wiley & Sons