Port Jackson Shark, Heterodontus Portusjacksoni, in Response to Lowered Salinity

J Comp Physiol B (2004) 174: 211–222 DOI 10.1007/s00360-003-0404-2

ORIGINAL PAPER

A. R. Cooper Æ S. Morris Osmotic, sodium, carbon dioxide and acid-base state of the Port Jackson shark, Heterodontus portusjacksoni, in response to lowered salinity

Accepted: 20 October 2003 / Published online: 19 December 2003 Springer-Verlag 2003

Abstract In marine elasmobranch fish the consequences content exhibited by the sharks, which caused a large for CO2 and acid–base state of moving into low salinity reduction in intracellular buffer. In water as dilute as water are not well described. Sub-adult Port Jackson 50% SW there was no evidence of specific effects on + sharks, Heterodontus portusjacksoni, occasionally enter the mechanisms of management of CO2 or H excretion brackish water and survive in 50% seawater (SW). The but rather significant and indirect effects of the severe unidirectional Na efflux and content, plasma volume, haemodilution. glomerular filtration rate (GFR), body mass, as well as CO2 and acid-base state in H. portusjacksoni were Keywords Acid-base Æ Na efflux Æ Shark Æ Hyposaline Æ investigated following transfer from 100% SW to 75% Heterodontus SW and then to 50% SW. A rapid water influx resulted in a doubling of the plasma volume within 24 h in sharks Abbreviations a–v arterial–venous Æ CA carbonic in 75% SW and an 11% increase in body weight. Os- anhydrase Æ CaCO2 content of CO2 in arterial 51 51 motic water influx was only partially offset by a dou- blood Æ CCO2 CO2 content Æ Cr-EDTA chromium- bling of the GFR. There was a 40% decrease in plasma ethylenediaminetetraactic acid Æ CvCO2 content of CO2 [Na] through a transiently elevated Na clearance and in venous blood Æ FW freshwater Æ GFR glomerular haemodilution. The result was a decrease in the inward filtration rate Æ Hct haematocrit Æ Jout Na flux gradient for Na+ together with reductions of nearly rate Æ MCHC mean cell haemoglobin 50% in CO2 and buffer capacity. The sharks remained concentration Æ OP osmotic pressure Æ PaCO2 partial hypo-natric to 50% SW by partially conforming to the pressure of CO2in arterial blood Æ PCO2 partial pressure decrease in external osmotic pressure and avoided the of CO2 Æ pHa arterial blood pH Æ pHer intra-erythrocyte + + need for active Na uptake. The gradient for Na efflux fluid Æ pHpl whole blood pH Æ pHv venous blood would by extrapolation approach zero at 27% SW pH Æ PvCO2 partial pressure of CO2in venous which may of itself prove a lethal internal dilution. In blood Æ SID strong ion difference Æ SW sharks transferred to 75% SW, a small transient seawater Æ TMAO trimethylamine-N-oxide Æ UFR hypercapnia and a later temporary metabolic alkalosis urinary flow rate were all largely explained through anaemia promoting loss of CO2 and buffer capacity. In sharks transferred to 50% SW the metabolic alkalosis persisted until the end Introduction of the 1-week trial. Within the erythrocytes, increased pH was consequent on the large decrease in haemoglobin Some elasmobranchs are able to move between seawater (SW) and freshwater (FW) (Thorson et al. 1973; Pier- marini and Evans 1998) but others, such as the Port Communicated by G. Heldmaier Jackson shark Heterodontus portusjacksoni, have a more A. R. Cooper limited tolerance of reduced salinity (Cooper and Morris School of Biological Sciences, University of Sydney, 1998a). Environmental salinity is an important factor in 2006 Sydney, NSW, Australia determining the distribution of a range of elasmo- S. Morris (&) branchs (e.g. Hopkins and Cech 2003). Morlab, School of Biological Sciences, University of Bristol, The movement of marine elasmobranchs into diluted Woodland Road, Bristol, BS8 1UG, UK E-mail: [email protected] SW or in a few cases into FW results in osmotic water Tel.: +44-117-9289181 influx facilitated by high branchial permeability (for Fax: +44-117-9288520 review see: Pang et al. 1977; Shuttleworth 1988; Evans 212

1993). In the most completely euryhaline species this anhydrase (Tufts and Perry 1998; Perry and Gilmour gain is managed so there is no resulting dilution anaemia 2002). The accompanying acidosis can be compensated (Piermarini and Evans 1998). In less euryhaline species for by a net increase in extracellular base (Maetz et al. net water gain occurs despite increases in glomerular 1976; Claiborne and Heisler 1984; Wood et al. 1984; filtration rates (GFRs) and urinary flow rates (UFRs) Claiborne et al. 2002; Choe and Evans 2003). Hyper- (Smith 1931a, 1931b; Goldstein and Forster 1971; capnic acidosis in fishes is compensated more slowly in ) Henderson et al. 1978) and the blood becomes diluted low salinity or in [HCO3 ]-deplete waters (Janssen and (for review see: Hickman and Trump 1969; Holmes and Randall 1975; Eddy et al. 1977; Heisler 1982, 1988; Donaldson 1969; Thorson et al. 1983; Shuttleworth Toews et al. 1983; Iwama and Heisler 1991; Choe and 1988; Evans 1993). The result can be lowered blood Evans 2003). Anaemia also reduces the non-bicarbonate haematocrit (Hct) and [Hb] of sharks (Goldstein and buffering capacity of the blood (Wood et al. 1979a, Forster 1971; Chan and Wong 1977b) including in H. 1979b) thus haemodilution and consequently lowered portusjacksoni (Cooper and Morris 1998a). Such anae- buffer capacity may have direct consequences for CO2 mia perturbs the acid–base and respiratory state of te- excretion. leosts (Wood et al. 1979b, 1982; Perry and Gilmour The acid–base response on transfer to a different 1993) but the acid–base responses of elasmobranchs are salinity is inextricably linked to both ion regulation, and not so well documented (e.g. Wood et al. 1994). ion regulation mechanisms (especially Na+ linked to There are almost no studies of the effect of salinity on H+ extrusion), and to maintenance of plasma volume. the blood acid–base status of marine elasmobranchs and Thus, the osmotic water influx and Na+ regulatory yet potentially this is an important aspect of their ability capacity are potentially major determinants of acid–base to penetrate into dilute water (e.g. Choe and Evans state. 2003). Teleosts moved from FW to higher salinity show Sub-adult Port Jackson sharks H. portusjacksoni are a relative metabolic acidosis (e.g. Maxime et al. 1990; distributed throughout the bays and lower estuaries of Claiborne et al. 1994), whereas those treated to the re- temperate coastal Australia (McLaughlin and OGower verse transfer to lower salinity show a relative metabolic 1970, 1971; Last and Stevens 1994), and are occasionally alkalosis (e.g. Madsen et al. 1996). A similar metabolic found in brackish waters up to 30 km from the sea. alkalosis correlated with living in FW has been described These sharks are not good regulators but nonetheless for the stingray Dasyatis sabina (Choe and Evans 2003). penetrate into estuaries. The osmotic and ionic status of The reasons for acid–base changes with salinity are H. portusjacksoni transferred to 50% SW appears to complex and unresolved but seem associated with come to a new equilibrium within 7 days (Cooper and changes in strong ion difference and differential ion Morris 1998a). H. portusjacksoni showed large declines uptake (Choe and Evans 2003). Marine elasmobranchs in plasma [Na+] and [Cl)], as well as [urea] and trim- are hypo-ionic with respect to SW and thus there is a ethylamine-N-oxide concentration ([TMAO]) and Hct, tendency for passive influx of ions. In contrast, elas- upon transfer to diluted SW (Cooper and Morris 1998a). mobranchs in FW must conserve and take up inorganic It seems possible that salinity-induced perturbations of ions from the water. Avoiding active ion uptake requires respiratory and acid–base state may contribute to lim- the fish to remain hypo-ionic to the water—effectively a iting the distribution of this shark into dilute SW. The strategy of conformation and with a lower limit that current study quantifies the changes in unidirectional does not allow penetration into FW. The strategy em- Na+ efflux and body Na+-status, together with changes ployed has implications for acid–base and respiratory in plasma fluid volume and volume management, and state. Of central importance is the management of Na+. examines these together with published data (Cooper There is debate as to the relative importance in elas- and Morris 1998a) as they may relate to maintenance of + + mobranchs of an apical Na /H exchanger (Piermarini CO2 transport and acid–base state in H. portusjacksoni and Evans 2001; Choe and Evans 2003) compared to an moving into 75% and 50% SW. apical V-ATPase-driven H+ excretion typical of teleosts in FW (Lin and Randall 1995; Wilson et al. 2000). + However, net H excretion promoting an internal Materials and methods alkalosis becomes more likely when the fish switch to + active Na uptake. Experimental design, blood sampling and acid–base measurements Lowered Hct of elasmobranch (Wood et al. 1994) and teleost fishes (Perry and Gilmour 1993; Gilmour and The acute ( £ 24 h) and chronic (up to 168 h) acid–base and CO2 Perry 1996) can reduce the rate of CO excretion (review: status of H. portusjacksoni transferred to diluted SW was deter- 2 mined using the experimental protocol described in Cooper and Tufts and Perry 1998). The mechanisms by which the Morris (1998a). Male and female Port Jackson sharks (0.7–2.0 kg) hypercapnia is compensated remain unclear. Elasmo- were caught by long-line off Bermagui, NSW and transported to branchs are unable to enhance CO2 excretion by in- Sydney Aquarium, Darling Harbour. Experimental sharks were transferred to the recirculating aquaria at the University of Sydney creased Hb-oxygenation since they lack a Haldane )1 effect but may be able to increase cardiac output without and acclimated in full-strength SW (33–35 g l ) maintained at 19.0±0.5 C for 1 week prior to experimentation. The PO2 of the compromising CO2 movement into the gill epithe- aquaria SW was maintained above 18.5 kPa and the sharks fasted lia since they have abundant cytoplasmic carbonic for 3 days prior to experimentation. Subsequently, the sharks were 213

) either held for a further 168 h (1 week) in either full-strength SW ferred to diluted SW was 150 mmol l 1 (see also Cooper and (100% SW) or acclimated for 1 week to 75% SW before being Morris 1998a) and resulted in a 3% increase in the aCO2 and an transferred to one of the following groups: increase in pK of 0.1 units. i. Control (acclimated in 100% SW and transferred to 100% SW) ii. 75% SW (acclimated in 100% SW and transferred to 75% SW) Determination of plasma volume and GFR iii. 50% SW (acclimated in 75% SW and transferred to 50% SW)A fourth treatment was initially included in which sharks The GFR and Na efflux rate (J ) were determined for H. por- were transferred directly from 100% to 50% SW but this out tusjacksoni (n=6 for each salinity) acclimated in 100% SW and was discontinued when the sharks became refractory to stimuli transferred to either 100% SW (control) or 75% SW (experimental) and showed discolouration of the skin within 72 h (Cooper for 72 h. The GFR was determined by the rate of clearance from 1999). the blood of the marker 51chromium-ethylenediaminetetraacetic 51 )1 )1 Sharks were anaesthetised in 100 mg l)1 MS222 and transferred acid ( Cr-EDTA; 1 llg wet weight; activity 3.7 kB ll ; sup- 51 to an operating table and the gills irrigated with aerated water plied CJ-13P, Amersham). The Cr-EDTA marker is a physio- containing 50 mg l)1 MS222. Cannulation of the caudal artery and logically inert biomedical marker not bound by plasma proteins or vein was as previously described (Cooper and Morris 1998b) and taken up by the erythrocytes (Chantler et al. 1969). The accuracy of consisted of inserting a 10-cm cannula around a stainless steel molecular filtration markers, especially inulin, has been questioned mandrel, which was then withdrawn (PE tubing, ID 0.36 mm; OD with regard to fish (e.g. Beyenbach and Kirschner 1976) but more 0.96 mm). This cannula could be connected to a 30-cm length (PE contemporary labels have been successfully employed (e.g. Curtis tubing, ID 0.58 mm, OD 0.96 mm) 24 h prior to sampling. The and Wood 1992; McKim et al. 1999). The marked EDTA space cannulae were flushed with heparinised (100 units ml lithium hep- thus provides a useful estimate of plasma volume (Holmes and arin)1, Sigma) shark Ringers (Rall and Sheldon 1961) and sealed Donaldson 1969). The sharks were allowed at least 24 h to recover by a 3-way valve. The sharks were placed in metabolism cages before initiating the experiment with the injection of undiluted 51 made from PVC pipe of 1 m length and 300 mm diameter, closed at Cr-EDTA. each end with 1 cm2 plastic mesh and supplied with an air-equili- Blood samples (200 ll) were taken at 2, 8, 14 and 26 h after the 51 brated water flow. A channel was cut in the top of the pipes to first injection. Further Cr-EDTA injections were performed di- provide an exit point for the catheter tubing. The procedure al- rectly after blood sampling at 26 h and at 50 h. Blood samples lowed the shark to move in an otherwise unrestrained environment removed after 28, 50, 52 and 74 h; to determine the extent of any whilst allowing blood samples to be taken without disturbing the progressive haemodilution. The sharks were weighed at the com- shark. The respiratory gas status and pH of the water in the three pletion of the experiment (74 h). The initial blood samples (0 h) 51 exposure treatments were not different and the mean values (n=32) were taken 2 h after the injection of Cr-EDTA to allow for a were; PO (kPa) 19.51±0.23, PCO (kPa) 0.25±0.02, and homogeneous distribution of the isotope throughout the plasma 2 2 51 pH 7.49±0.03. space. For calculations with respect to multiple Cr-EDTA injec- Blood samples were removed using pre-chilled heparinised 1-ml tions, the differences in activity prior and subsequent to re-injection Hamilton gas-tight syringes from the cannulated caudal artery and were used. The volume of blood removed during the sampling 51 vein of quiescent sharks at either 0, 6, 12, 24, 72 or 168 h after procedure for Cr-EDTA was replaced with an equivalent volume transfer (n=6 for each time period). The samples were measured of elasmobranch Ringers solution. The whole blood osmotic immediately for whole blood CO2 partial pressure (PCO2) and CO2 pressure (OP) was measured using a vapour pressure osmometer content (CCO ) and whole blood pH (pH ) and intra-erythrocyte (Wescor 5100C, Logan, USA) prior to the blood being centrifuged 2 pl ) fluid pH (pHer). Whole blood PCO2 was measured with a BMS and the plasma decanted and stored at 15 C for subsequent Mk2 blood micro system (Radiometer) at 19 C and connected to a analysis. 51 PHM73 meter. The CO2 electrode was calibrated with analytical The radioactivity of Cr-EDTA in 50-ll plasma samples was grade humidified gases containing 0.34% and 2.15% CO . The measured using a gamma counter (1272 CliniGamma ‘‘3-inch’’, 2 51 blood CCO2 was measured in a 25-ll sample injected into a CO2 LKB-Wallac). The rates of clearance of Cr-EDTA were calcu- )1 analyser (Ciba-Corning 965) calibrated daily with a 15 mmol l lated from the equation; EDTA space (ml)=Q/A0, where Q is the NaHCO3 standard. Both pHpl and pHer were measured with a total injected radioactivity (t=0) (cpm), and A0=plasma radioac- capillary electrode (G229a) mounted within the BMS Mk2 and tivity extrapolated back to the injection time (t) (cpm) as the )1 51 calibrated with Radiometer precision buffers. ln(value) h after accounting for decay of Cr by reference to the Blood samples that were taken immediately after the 1-week cpm of the stock solution (Greenaway et al. 1990). The GFR was 51 acclimation period are referred to as 0 h values. Samples taken assumed to be equal to Cr-EDTA clearance and was determined )1 from sharks held in 75% SW for 1 week were used as pre-treatment from the equation; GFR (ml h )=bÆEDTA space, where b=the values for 50% SW groups. The experimental design was inde- slope of the natural log of radioactivity (decay corrected) in the pendent and thus with six sharks at each sample time, more than 96 plasma when plotted against time. sharks were sampled. ) The whole blood [HCO3 ] was calculated using the Henderson– Hasselbalch equation: ÂÃ Determination of Na space, turnover and efflux HCO3 pH ¼ pK1 þ log Unidirectional sodium efflux measurements were initiated by the aCO PCO 2 2 injection of 22Na in an aqueous sodium chloride solution )1 22 where pK1=the first dissociation constant for CO2 and H2O and (0.5 llg wet weight, Cat. No. SKS. 1, Amersham; specific Na )1 )1 )1 aCO2=the solubility coefficient of CO2 (mmol l kPa ). Re- activity 7.4 kB ll ) diluted 1:50 with elasmobranch Ringers quired values for pK and aCO2 were derived from those of Scy- solution. The blood sampling protocol was the same as described liorhinus stellaris blood. The pK (pKapp=6.029 at pH 7.8 and above for the determination of the GFR (n=6). 19 C) was obtained from Albers and Pleschka (1967) and the Plasma samples (100 ll) were added to 2 ml scintillant (Packard )1 22 aCO2 (aCO2=0.3435 lmol l kPa) calculated after Boutilier et al. Ultima) and the radioactivity of the Na measured for 10 min (1984) from the data of Pleschka and Wittenbrock (1971). A de- using the liquid scintillation analyser (Tri-Carb 1600TR, Packard). crease in plasma ionic strength is accompanied by an increase in the The Na space is equal to that fluid volume within the animal at the aCO2 and pK1 (Albers 1970; Maas et al. 1984; Burton 1987). This same [Na] as the blood (Holmes and Donaldson 1969) and allows effect on aCO2 and pK was estimated as outlined by Maas et al. the calculation of total body Na content as the Na space (ml) (1984), which incorporated the change in plasma [Na+]. The multiplied by the plasma [Na] (mmol l)1). The Na space was cal- maximal decrease in plasma [Na+] for H. portusjacksoni trans- culated as per Shaw (1963). The flux constant (K), or fraction of the 214 total Na pool being exchanged per unit time, was from the equa- Table 1). In contrast, the body weight of sharks trans- tion; K=(lnA0)lnAt)/t (see Haywood 1974), where Ao=specific 22 22 ferred to 75% SW for 72 h increased by 11% (Table 1). activity of Na at time zero, At=specific activity of Na at time t, t=time (h). The calculation of plasma Na efflux, when 22Na is The changes in whole blood OP and plasma Na con- injected into the blood and measured in the plasma, was derived centration (below) reflected the changes in body weight from the equation: (Fig. 1). The whole blood OP of sharks held in 100% y0 ¼ y0 ðÞexpðÞ M=A t SW for up to 72 h did not differ from pre-treatment 0 2 values (Fig. 1A). Upon transfer to 75% SW, the OP where y¢=concentration of radioactive ions in the SW (counts - )1 0 decreased by 18% after 12 h and continued to decline min ), y0=concentration of radioactive ions in the plasma )1 )1 (counts min ), A2=plasma [Na] (mmol l ) and M=flux rate, which can be simplified as Na efflux=Na poolÆ(K/M) lmol g)1 h)1 (see Shaw 1963). The plasma [Na] was measured on an atomic absorption spectrophotometer (GBC 906AA, Melbourne) using the Table 1 The change in body weight (DBW), EDTA space and the procedures described in Cooper and Morris (1998a). glomerular filtration rate (GFR as 51Cr-EDTA clearance) of Het- erodontus portusjacksoni transferred to either 100% or 75% SW for up to 72 h. Values are given as means±SEM. Hyphen denotes data unavailable. Theasterisk denotes significantly different from sharks CO dissociation curves 2 in 100% SW. (n=6 in each treatment)

Whole blood CO2 dissociation curves were determined from oxy- Time (h) Treatment DBW (%) EDTA space GFR ) ) genated and deoxygenated blood samples pooled from sharks (% body weight) (Dl 100 g 1 h 1) transferred to either full-strength or diluted SW for 72 h (n=6 for each salinity; see below). Blood samples from each treatment were pooled in a 10-ml heparinised plastic vial and placed on ice. The 2–24 100% SW - 2.54±0.93 245±66 Hct was measured immediately and throughout the subsequent 75% SW 5.31±0.86* 511±104* sampling procedure to detect any cell swelling or lysis. The Hct 26–48 100% SW - 3.59±0.98 190±84 differed between control (100% SW=18% Hct) and experimental 75% SW 3.59±0.58 205±51 groups (75% SW=15% Hct; 50% SW=14% Hct) but did not 50–72 100% SW +1±1 - - vary within each treatment throughout the sampling procedure. 72 (body 75% SW +11±1* 4.55±1.42 199±180 mass) The whole blood CO2 dissociation curves were generated from 80-ll blood samples tonometered with humidified gas mixtures using BMS MkII (Radiometer). The blood was equilibrated for 25 min at 19 C with either an oxygenated or a deoxygenated (N2) gas mixture supplied from gas mixing pumps (Wo¨ stoff, Bochum, Germany). The PCO2 in the mix was varied between 0.10 kPa and 5.0 kPa, using instrument grade CO2. Duplicate whole blood CCO2 and pH mea- surements were carried out for each pre-determined PCO2, and the ) [HCO3 ] calculated using procedures described above.

Statistics

Data from sharks transferred to either 100%, 75% or 50% SW were compared by two-way ANOVA using SYSTAT. Within a given parameter arterial–venous (a–v) differences were tested using a three-factor ANOVA. All values are expressed as means±SEM, unless otherwise stated, with a probability (P) value <0.05 con- sidered significant. For those symbols without error bars, SEM was smaller than the size of the symbol. Homogeneity of variances were assured using Bartletts v2 test. In cases where variances were het- erogeneous, log or square root transformations were performed prior to further analysis. Post-hoc testing was performed using ei- ther contrast analysis or Tukeys multiple means comparison analysis. Data collected from serially sampled sharks for the measurement of whole blood OP, GFR, plasma [Na] and Na efflux rates were analysed using repeated measures ANOVA. Regression analysis was carried out on the relationship between whole blood ) [HCO3 ] and pH of sharks held in either 100%, 75% or 50% SW. Analysis of co-variance (ANCOVA) was used to detect any differences between the regression lines.

Results

Body weight, whole blood osmotic pressure, plasma space and clearance Fig. 1A–B A The whole blood osmotic pressure (OP) and B plasma [Na] of serially sampled Heterodontus portusjacksoni transferred to either 100% or 75% SW for up to 72 h. Values The body weight of control sharks held in 100% SW for shown as means±SEM. The asterisk denotes significantly different 72 h did not differ from pre-treatment values (0 h; from sharks in 100% SW (n=6) 215

) such that after 72 h, values were 35% lower than those PCO2, CCO2, pH and [HCO3 ] of control sharks (Fig. 1A). The EDTA space (plasma space) of H. portusjacksoni The whole blood PaCO2 of H. portusjacksoni moved to held in full-strength SW remained constant between diluted SW did not differ from control values while the 2.5% and 3.6% of the total body weight (Table 1). In PvCO2 increased by 50% after 6 h and 12 h of transfer contrast, the plasma space of sharks transferred to 75% to either 75% or 50% SW, respectively, but was restored SW quickly doubled, within 24 h, to over 5% body to pre-treatment values (0 h) within 24 h. (Table 3). weight despite a concomitant doubling of the glomerular There were transitory increases in blood CCO2 between filtration rate (GFR as EDTA clearance; Table 1). 6 h and 24 h after transfer to 75% SW, the largest being Within 48 h, however, neither the plasma space nor the a 49% elevation of CaCO2 after 12 h. Compared to GFR of sharks held in 75% SW differed significantly those in 75% SW, the sharks moved to 50% SW from control values (Table 1). exhibited a short-term decrease in CaCO2 (Table 3). The CvCO2 of sharks transferred to either 75% or 50% SW increased by 37% and 25%, respectively, after 12 h but Plasma and body [Na] and Na efflux returned to control values after 24 h (Table 3). Signifi- cant a–v differences were generally confirmed by The plasma [Na] of the control sharks did not differ ANOVA but were small enough not to be detected at from pre-treatment values throughout the sampling some specific exposure times in post hoc tests (Table 3). procedure (Fig. 1B). In contrast, the plasma [Na] of The further analysis of acid–base status in response to sharks transferred to 75% SW declined by 30% within salinity was carried out using combined arterial and 6 h and after 72 h was 40% lower than control values venous values. (Fig. 1B). The percentage turnover of exchangeable Na The whole blood arterial pH (pHa) of sharks trans- (K=100) showed a 2.7-fold increase over the first 2 h in ferred to diluted SW did not differ from control values 75% SW from 8.2% to 22% (Table 2). Thereafter, the (Table 3) although the whole blood venous pH (pHv)of rate in 75% SW-exposed sharks became more similar to sharks transferred to diluted SW showed an eventual the rate in 100% SW sharks and varied between 5% and alkalosis of approximately 0.1 pH units. Sharks in 75% just over 9%. Acute differences were also observed in a SW achieved a maximum pH 7.88 after 24 h which was decreased total Na pool and increased unidirectional Jout then corrected whereas those transferred to 50% SW in experimental sharks (Table 2). Thus, in 75% SW-ex- reached a pH 7.88 after 1 week (168 h) (Table 3). There posed sharks, the mean Jout for Na reached were indications of an intra-erythrocytic alkalosis in 577 lmol 100 g)1 h-1 within the first 2 h exposures both the 75% and 50% SW treatments (Table 3) but compared to 418 lmol 100 g)1 h-1 in sharks kept in these could not be confirmed by post hoc testing and the 100% SW. At the same time the total Na pool was arterial and venous values were pooled for further significantly decreased to 2.83 mmol 100 g)1 compared analysis of salinity effects. with 6.84 mmol 100 g)1 (Table 2). The sharks in 75% SW did not re-establish the Na pool (as with plasma [Na]) although Jout slowed, concomitant with the de- CO2 dissociation curves creased outward diffusion gradient and lowered GFR (Table 2). The whole blood CO2 dissociation curves were steep- est in the physiological PCO2 range (<0.4 kPa) (Fig. 2A). Oxygenated and deoxygenated blood at a fixed PCO2 showed no differences in pH and thus no Table 2 The changes in the unidirectional Na efflux (Jout), total Na Haldane effect (consistent with very small Bohr fac- and the relative exchange coefficient (K·100) of H. portusjacksoni tors; Cooper and Morris 2003). As a consequence, transferred to either 100% or 75% SW for up to 72 h. Flux rates and 22Na dilutions were determined over the initial 2 h of each there was no significant difference between the non- period. The periodicity of sampling allowed the events between 8– bicarbonate buffering capacities of oxygenated and 10 h to be calculated separately. All values are given as mean- deoxygenated blood (ANCOVA). The non-bicarbonate s±SEM. The asterisk denotes significantly different from sharks in ) buffering capacity (b) was calculated as; b=D[HCO3 ]/ 100% SW. (n=6 in each treatment) D pH (see Truchot 1987), the slope derived by )1 Time (h) Treatment K (% h ) Total Na pool Jout (lmol regression analysis. The data from oxygenated and (mmol 100 g)1) 100 g)1 h)1) deoxygenated blood for each treatment were thereafter pooled for subsequent analysis (Fig. 2B). There was a 0–2 100% SW 8.2±1.7 6.84±1.52 418±97 significant decrease in the non-bicarbonate buffering 75% SW 22.0±3.7* 2.83±0.63* 577±141* 8–10 100% SW 8.2±1.6 5.80±0.2.53 438±172 capacity of sharks transferred to either 75% SW 75% SW 9.4±1.2 4.59±1.15 435±129 (b=)4.5) or 50% SW (b=)4.0) compared with con- 24–26 100% SW 5.0±0.4 9.23±2.34 368±30 trol values (b=)7.7; ANCOVA; Fig. 2B). The lowest 75% SW 6.7±0.4* 2.46±0.34* 164±23* buffering was clearly at the lowest exposure salinity 48–50 100% SW 6.1±1.3 9.08±3.04 438±137 75% SW 5.6±0.4 4.22±1.27* 220±58* (50% SW) and was only 52% of that in sharks in 100% SW. 216

Table 3 The whole blood and intra-erythrocyte fluid pH, PCO2 An a–v difference in intra-erythrocyte fluid pH (pHer) was generally and CCO2 of H. portusjacksoni transferred to either 100%, 75% or significant only in the control sharks in 100% SW (P<0.01). 50% SW for up to 1 week (168 h). All values are given as mean- ADenotes a significant a–v difference at a specific exposure time as s±SEM. The asterisk denotes significantly different from sharks in detected by post hoc contrast analysis. Generally there was sig- either 100% SW and the hash mark different from those in 75% nificant a–v difference in the PCO2 in sharks in 100% SW SW. Two-factor ANOVA was applied to each parameter to test for (P=0.049), 75% SW (P<0.01) and 50% SW (P=0.001). An a–v arterial-venous (a–v) differences. Three-factor ANOVA was ap- difference in CCO2 was generally significant only in those sharks plied to test for differences between treatments in each of pH, exposed to 50% SW (P<0.01) and could not be confirmed for ) ) PCO2, CCO2 and [HCO3 ]. Generally there was significant a–v [HCO3 ]. In view of the few time-specific a–v differences, further difference in the whole blood pH (pHpl) in sharks in 100% SW analysis (Fig. 3) combined arterial and venous values of CCO2 and (P<0.001), 75% SW (P<0.01) and 50% SW (P=0.15), although pH. (n=6 at each time in each treatment; totaln=96 sharks) time-specific differences could not be detected in post hoc analysis.

Time Treatment PaCO2 (kPa) PvCO2 (kPa) CaCO2 CvCO2 Blood pHa Erythrocyte Blood pHv Erythrocyte )1 )1 (h) (mmol l ) (mmol l ) pHa pHv

0 100% SW 0.28±0.01 0.31±0.02 6.3±0.3 6.3±0.1 7.82±0.01 7.06±0.01 7.76±0.01 A 7.10±0.03 50% SW 0.27±0.02 0.30±0.02 6.0±0.7 6.3±0.7 7.80±0.03 7.02±0.05 7.76±0.03 7.03±0.04 6 100% SW 0.28±0.05 0.29±0.03 5.6±0.5 5.7±0.6 7.79±0.02 7.04±0.03 7.74±0.03 6.98±0.02 75% SW 0.36±0.02 0.43±0.04*A 7.1±0.7 5.5±0.3 7.80±0.01 7.05±0.03 7.71±0.05 A 7.11±0.03* 50% SW 0.32±0.03 0.34±0.02 7.2±0.8 6.3±0.6 7.84±0.03 7.18±0.05*# 7.81±0.03*# 7.11±0.05* 12 100% SW 0.23±0.03 0.25±0.02 5.7±0.5 6.2±0.4 7.82±0.02 7.03±0.03 7.75±0.02 A 7.09±0.03 75% SW 0.38±0.04 0.37±0.02* 8.5±0.5 8.5±0.3* 7.85±0.02 7.04±0.01 7.81±0.02 7.04±0.02 50% SW 0.30±0.03 0.38±0.01*A 5.8±0.7# 7.7±0.4*A 7.84±0.02 7.07±0.06 7.82±0.03 7.17±0.02*# 24 100% SW 0.26±0.04 0.33±0.01A 6.4±0.7 6.7±1.1 7.82±0.03 7.05±0.06 7.79±0.02 7.13±0.03 75% SW 0.33±0.01 0.34±0.01 9.0±0.5* 8.0±0.7 7.84±0.04 7.16±0.05 7.88±0.05*# 7.09±0.06 50% SW 0.31±0.02 0.33±0.03 4.8±0.6# 6.9±0.5A 7.83±0.01 7.17±0.04 7.81±0.03 7.10±0.02 72 100% SW 0.25±0.04 0.31±0.02A 6.1±0.4 6.3±0.5 7.83±0.01 7.06±0.05 7.78±0.01 7.10±0.04 75% SW 0.31±0.06 0.39±0.06A 6.7±0.7 8.2±0.7 7.88±0.02 7.18±0.04* 7.85±0.04 7.20±0.04*# 50% SW 0.30±0.03 0.35±0.02 6.0±0.6 7.9±0.5A 7.83±0.02 7.10±0.04 7.81±0.03 7.09±0.02 168 100% SW 0.30±0.02 0.32±0.02 7.0±0.5 6.4±0.5 7.84±0.01 7.12±0.04 7.79±0.02 A 7.21±0.06 75% SW 0.27±0.02 0.30±0.02 6.0±0.7 6.3±0.7 7.80±0.03 7.02±0.05 7.76±0.03 7.03±0.04*# 50% SW 0.28±0.03 0.33±0.02 5.3±0.2 7.1±0.3A 7.88±0.02 7.14±0.04 7.88±0.03*# 7.16±0.01

blood OP, but maintain a larger osmotic gradient Discussion inwards (Smith 1931a, 1931b; Thorson 1958, 1961; Piermarini and Evans 1998). Osmotic and sodium response to SW dilution The ability of truly euryhaline elasmobranchs to maintain fluid volume while in diluted SW is largely H. portusjacksoni is a marginal osmoregulator and when through elevation in GFR and UFR (Smith 1931a; De exposed to dilute SW exhibited rapid lowering in the Vlaming and Sage 1973; Piermarini and Evans 1998; concentrations of the major inorganic and organic Choe and Evans 2003). The transient two-fold increase plasma osmolytes with concomitant effects on intra- in the GFR of H. portusjacksoni transferred to 75% erythrocytic concentration (Cooper and Morris 1998a). SW was similar to increases shown by other predomi- H. portusjacksoni in 75% SW showed a 16% decline in nantly marine skates and sharks (e.g. Burger 1965; plasma [Na+], and after a further week in 50% SW, the Goldstein and Forster 1971; Schmidt-Nielsen et al. decrease was 27% with concomitant decreases in the 1972; Forster and Goldstein 1976; Chan and Wong erythrocytes of (18% and 78%). It was previously un- 1977a, 1977b). However, this doubling of GFR in H. clear as to what proportion of the decrease was due to portusjacksoni was less than 25% of the GFR increases ion loss and what to haemodilution (Cooper and Morris shown by species entering very dilute waters (e.g. Smith 1998a). In the current study, the decline in blood OP 1931a; Janech and Piermarini 1997). The acutely in- (35%) and accompanying increase in body weight (11% creased GFR in H. portusjacksoni re-established plasma chronic) of H. portusjacksoni transferred to 75% SW volume but was concomitant with, and probably as- were due to the increase in osmotic water gain, which sisted by, increased Na efflux (below). The generally transiently doubled the plasma volume within 24 h. The lowered osmolyte concentrations ameliorate water in- inability of most marine elasmobranchs to prevent os- flux. A strategy of partial conformation (or partial motic water loading from diluted SW increases body regulation) differs from that employed by regulating weight by between 10% and 25% (Goldstein and For- euryhaline elasmobranchs. In H. portusjacksoni the ster 1971; Forster and Goldstein 1976; Chan and Wong persistent water load after 1 week (11%) was con- 1977a, 1977b; Cooper and Morris 1998a; Sulikowski sistent with an inability to raise GFR to match influx and Maginniss 2001). In contrast, strongly euryhaline and with a new equilibrium at lower body osmolyte elasmobranchs such as the bull shark, Carcharhinus concentration (Cooper and Morris 1998a). The acutely leucas, the sawfish, Pristis microdon, and the Atlantic elevated GFR, without matching increases in osmolyte stingray regulate a constant extracellular space and uptake, amounts to active excretion by the sharks to 217

66 mmol l)1 in those in 50% SW (Cooper and Morris 1998a), these sharks remained hypo-natric. The rela- tionship between D[Na+] (plasma) and water salinity was a good linear fit and could be expressed as D[Na+]=2.75(%SW))70.7; r2=0.99. Solving this for D[Na+]=0 provided a salinity of 25.7% SW which is the theoretical salinity at which the inward gradient for Na+ will disappear. However, at a salinity of 25.7% SW, the plasma [Na+] will have declined to only 120 mmol l)1 and may of itself prove lethal. Some sharks adjust flux through altered branchial Na permeability, e.g. the py- jama shark, Poroderma africanum (Haywood 1974), but the apparent Na permeability coefficient (K) (deter- mined from maximum efflux measurements, e.g. Greenaway 1981), showed little change comparing H. portusjacksoni in 75% (K=0.0061) and 100% SW (K=0.0051). Circumstances for H. portusjacksoni in 75% and 50% SW are quite different compared, for example, to those of Atlantic stingrays maintaining Na+ uptake from fresh- water (Choe and Evans 2003). While in freshwater elas- mobranchs the uptake of Na+ may limit H+ extrusion it seems that the Port Jackson sharks in the diluted SW rapidly cleared and diluted body Na+ to maintain the plasma hypo-natric with an inward [Na+] gradient of at least 65 mmol l)1. Reducing body Na+ along with the organic plasma osmolytes in addition to slowing osmotic inflow may also maintain some functionality of the Na+/ H+ exchange system, albeit with a lower Na+ influx, and thereby minimise acid–base perturbations.

Carbon dioxide and regulation of acid–base state

The whole blood acid–base status of H. portusjacksoni in 100% SW was similar to that of other marine elasmo- branchs. The mean PCO2 (0.30±0.02 kPa) was 30% Fig. 2A–C A The mean CO2 dissociation curves derived from lower than that determined for the species by Grigg pooled oxygenated and deoxygenated whole blood of H. portus- jacksoni transferred to either 100%, 75% or 50% SW for 72 h and (1974), but like the CCO2 and pH, was well within the ) B the relationship between whole blood [HCO3 ] and pH under range of other marine elasmobranchs (Lenfant and Jo- equivalent conditions. Regression of the data from each salinity ) hansen 1966; Piiper and Schumann 1967; Butler et al. group provided the equations: 100% SW, [HCO3 ]=67.86– 2 ) 2= 1979; Truchot et al. 1980; Bushnell et al. 1982; Duthie 7.75.pH (r =0.96); 75% SW, [HCO3 ]=42.60–4.61.pH (r 0.95); ) 2= and Tort 1985; Wood et al. 1994; Tufts and Perry 1998). 50% SW, [HCO3 ]=30.10–4.05.pH (r 0.94). C The CO2 disso- ciation curves redrawn as a double log plot for H. portusjacksoni Compared to resting leopard sharks (Lai et al. 1990), the after exposure to 100%, 75% and 50% SW. The exponent of the a–v differences in Port Jackson sharks are small and slope DCCO2/DPCO2 (B) around the normal blood Pco2 provides barely significant. The metabolic rate of the resting Port a measure of CO2 capacitance (after Piiper 1986). Blood pooled Jackson sharks is extraordinarily low and produced very from equal volume samples taken from six fish in each treatment small a–v differences in [O2] (Cooper and Morris 2003) and thus accounts for the small a–v difference in CCO2. reach this new equilibrium as rapidly as possible and to Additionally, samples from the caudal vein may not be minimise further water influx. entirely representative of mixed venous blood, which In adjusting to 75% SW, the Jout of H. portusjacksoni could not be sampled in this study, and may minimise was increased, but after 24 h the GFR stabilised as did the observed a–v differences. However, a–v samples were total body Na, plasma Na and Na efflux but at adjusted pooled for further analysis of salinity effects. The blood values. Marine elasmobranchs are hypo-natric com- acid–base and CO2 status of H. portusjacksoni trans- pared to SW, thus the tendency for passive Na+ influx. ferred to either 75% or 50% SW showed small but sig- In 100% SW the inward concentration gradient for Na+ nificant changes. The changes in the acid–base status of ) was 203 mmol l 1, and while this decreased through Na H. portusjacksoni were similar to those observed in some ) clearance to 138 mmol l 1 in sharks in 75% SW and to teleost fishes transferred to low salinity waters (Perry 218 c ) Fig. 3A–C The pH/HCO3 diagrams for changes in the acid–base status of H. portusjacksoni transferred to either 100% SW (A), 75% SW (B), or 50% SW (C) for up to 1 week (168 h). Values shown as means±SEM. The broken lines indicate the non-bicarbonate buffering and are derived from the data in Fig. 2. The hash mark ) indicates [HCO3 ] and the asterisk shows pH different from that of ) sharks in 100% SW; star indicates [HCO3 ] and the tick shows pH different from that in sharks exposed to 75% SW for the same length of time. The arrow indicates significantly elevated PCO2 as determined by post hoc analysis (Tukeys) following one-way ANOVA of the combined arterial–venous PCO2 values. The PCO2 isopleths were derived from the Henderson–Hasselbalch relation- )1 )1 ship using a pKapp=6.029 and aCO2=0.3435 lmol l kPa (n=6 at each time and each treatment; total n=96 sharks) and Heming 1981; Nonnotte and Truchot 1990; Clai- borne et al. 1994) and can be best described by pH vs. ) [HCO3 ] diagrams (Fig. 3). The mean blood pH in control sharks (100% SW) ) varied by less than 0.07 pH units and the [HCO3 ]by 1.4 mmol l)1, none of which represented a significant change (Fig. 3A). After 6–12 h in 75% SW the PCO2 of H. portusjacksoni was significantly elevated from 0.30 kPa to 0.39 kPa and 0.38 kPa, respectively (Fig. 3B), but declined again by 24 h exposure. The source of this small hypercapnia is difficult to determine but was coincident with the rapid decline in osmotic pressure and [Na+], as well as a very large and transient increase in plasma volume. The plasma dilution caused significant loss of Hct from 18–20% to 14% after 6 h and a lowering of blood [Hb] (Cooper and Morris 1998a, 2003), which would encourage increased PCO2. There is the possibility that, unlike the diffusion-limited system in teleosts, elasmobranchs have a perfusion-lim- ited CO2 excretion (Perry and Gilmour 2002) due to the occurrence of plasma and branchial vasculature car- bonic anhydrase (Henry and Swenson 2000; Gilmour et al. 2001, 2002). However, there was no compensatory increase in cardiac output in H. portusjacksoni (Cooper and Morris 2003) and thus any excess dissolved CO2 was not expressly excreted. Experimental anaemia depresses CO2 excretion in both elasmobranch and teleost fishes (Wood et al. 1982, 1994; Perry and Gilmour 1993). Furthermore, the rate of CO2 excretion was positively correlated with Hct (Perry et al.1982; Tufts and Perry 1998). The rate of dehydra- ) tion of HCO3 to CO2 in the blood of the dogfish, The whole blood non-bicarbonate buffering capacity Scyliorhinus canicula, becomes markedly reduced at of H. portusjacksoni in 100% SW (b=)7.7) was similar haematocrit values <15% (Wood et al. 1994). Extrap- to that of marine elasmobranchs in which the b values olation from the data of Wood et al. (1994) for Scylio- range between )6and)10 (Albers and Pleschka 1967; rhinus canicula, showed the reduction in the Hct of H. Burger 1967; Piiper and Baumgarten-Schumann 1968a, portusjacksoni would have been equivalent to a 20% 1968b; Truchot et al. 1980; Weber et al. 1983; Heisler ) reduction in the rate of HCO3 dehydration, and et al. 1988; Lai et al. 1990; Wood et al. 1994; Gilmour therefore of CO2 excretion, promoting hypercapnia. et al. 2002). The absence of any significant Haldane ef- ) Interestingly, the HCO3 dehydration reaction, and thus fect in H. portusjacksoni was typical for elasmobranchs carbonic anhydrase (CA) function of Squalus acanthias, (see: Butler and Metcalfe 1988 for review). The non- was unaffected by low [urea] and [TMAO] (Wood et al. bicarbonate buffering capacities of many teleosts moved 1994) and it is unlikely that the CA of H. portusjacksoni to low salinity water are maintained or even increased was altered by the lowered TMAO and urea (Cooper (Maxime et al. 1990; Nonnotte and Truchot 1990). and Morris 1998a). However, in H. portusjacksoni transferred to either 75% 219 or 50% SW, the capacity declined by >40%. Changes in indistinguishable from the original condition or that of Hct will also markedly alter blood buffer values (Tufts sharks in 100% SW (Fig. 3B). Transferring the sharks and Perry 1998) and directly determine the CO2 capac- from 75% to 50% SW introduced another increase in ) itance (Piiper 1986). Haematocrit and non-bicarbonate [HCO3 ] within 6 h compared to both sharks in 100% buffering capacity of H. portusjacksoni transferred to and 75% SW (Fig. 3C). At some time between 3 days diluted SW could be described by the linear relationship: and 7 days this was exacerbated into a metabolic alka- 2 b=)9.110+(0.993ÆHct) R =0.99. The CO2 dissociation losis exceeding 0.1 pH units (Fig. 3C). It is possible curve can be adequately described by a power function given the projected lower salinity limited of 27% (Piiper 1986): SW that the alkalosis in the 50% SW sharks would also  have been removed eventually. B Cco2 Pco2 The strong ion difference (SID) of the plasma of H. ¼ portusjacksoni was calculated according to Maxime et al. Cco20ðÞ Pco20ðÞ (1990) (but with Ca2+ and Mg2+ included) from plasma where (0) indicates values for the normal condition and ion measurements reported previously for similarly where the value of the exponent B depends on the treated H. portusjacksoni (Cooper and Morris 1998a). ) magnitude of b and thus varies significantly with expo- The changes in [HCO3 ] were within the range of error sure of the sharks to lowered salinity (Fig. 2C). The of the ion measurements but the transient metabolic value of B declines markedly from B=2.18 with expo- alkalosis in 75% SW-exposed sharks was concomitant sure to dilute SW and was as low as B=0.63 in 50% SW with a 6.5-mEq l)1 increase in SID after 72 h which was (Fig. 2C). It thus appears that increased plasma volume subsequently removed. Further exposure to 50% SW and haemodilution precipitously (see also Cooper and was matched by an increase in SID of 4.5 mEq l)1 Morris 1998a) decreased the CO2 capacitance and buffer which persisted together with the alkalosis. These capacity of the blood. The maximal increase in the blood changes in blood SID are sufficient to explain the PCO2 of H. portusjacksoni transferred to diluted SW metabolic alkalosis. was only 0.2 kPa which would be equivalent to a The linear relationship between the in vivo pHpl and ) )1 transient increase in [HCO3 ] of only 1 mmol l pHer of H. portusjacksoni described by the equation; (Fig. 3). A rapid and acute decrease in CO2 capacitance pHer=1.03+(0.78ÆpHpl), had a similar slope to that would promote an increase in PCO2 until such time as calculated for the dogfish, Squalus acanthias,overan the excess could be excreted. Gilmour et al. (2001, 2002) equivalent pH range from the data of Wells and Weber proposed the relatively high buffer capacity (b) of elas- (1983) but somewhat greater than for teleost fishes mobranch plasma normally provides a greater supply of which have DpHpl/DpHer of 0.6 (Tetens and Lykkeboe H+ compared to, for example, teleost fish with lower 1981; Jensen and Weber 1982; Heming et al. 1986). The plasma b value and that this imparts an important role pHer of 75% SW increased from pH 7.08±0.02 to to membrane bound branchial CA. Reducing b in H. pH 7.19±0.02, and similarly in 50% SW-exposed portusjacksoni by as much as 48% may thus lower H+ sharks from pH 7.03±0.03 to pH 7.15±0.02 after ) availability and slow HCO3 hydration at the gill 1 week. Ordinarily this would require a decrease in thereby impeding CO2 excretion. Direct determination PCO2, which was not the case in H. portusjacksoni of plasma protein and buffering separate from cellular (Fig. 3). However, the low salinity exposure treatments buffering would reveal any special role in the response to promoted a transient lowering of mean cell haemoglobin salinity transfer. Nonetheless, the small hypercapnia of concentration (MCHC) in 75% SW and a more persis- sharks in the 75% SW appears to be an indirect effect of tent reduction of up to 30% of MCHC in 50% SW- haemodilution on carriage of CO2 in the blood and is exposed sharks (Cooper and Morris 1998a). Together eventually accommodated. The sharks transferred to with the loss of Hct, this probably accounted for the 50% SW had previously been acclimated to 75% SW. very larger reductions in CO2 capacity and buffering of The 50% SW treatment thus represents sharks that had the blood (Fig. 2). Perturbations, in intra-erythrocytic already compensated for a small hypercapnia and a pH are thus primarily due to dilution rather than any small subsequent metabolic alkalosis (below). The acute changes per se in CO2 or Na and Cl management. exposure to 50% SW did not induce any further loss of There is little evidence of any limitation of either H+ Hct but did cause a pronounced increase in ventilation or CO2 excretion as a direct result of reduced salinity, at volume (Cooper and Morris 2003) which is consistent least as low as 50% SW. Much longer term and gradual with the absence of further acute dilution induced acclimations to even lower salinity in attempt to remove hypercapnia. the inward Na+ gradient may reveal whether or not The transfer of H. portusjacksoni from 100% to 75% Na+ limitations specifically impede acid–base balance in SW resulted in a small but significant accumulation of H. portusjacksoni. It seems more likely, however, that ) [HCO3 ] after 12 h which after 24 h and 72 h had these sharks will be limited by unsustainable salt loss manifest as a similarly small but significant metabolic before they adjust to the projected 27% SW required. In alkalosis of 0.11 pH units (Fig. 3B). This metabolic such circumstances, CO2 and acid–base perturbations alkalosis was completely removed between 3 days and would not be prime determinants limiting the penetra- ) 7 days and the values of both pH and [HCO3 ] were tion of Port Jackson sharks into very dilute SW. The loss 220 of Hct and MCHC have important implications for Cooper AR (1999) The effect of changes in seawater salinity on the O -uptake and transport which may be compromised by blood physiology of the Port Jackson shark, Heterodontus 2 portusjacksoni. PhD Thesis, University of Sydney, Australia, the dilution effects. pp 315 Cooper AR, Morris S (1998a) Osmotic, ionic and haematological Acknowledgements We are indebted to A. Broadhurst for the col- response of the Port Jackson shark, Heterodontus portusjack- lection of the sharks and to the Darling Harbour aquarium for soni, and the common stingaree, Trygonoptera testacea, upon facilities. This work was carried while A.R.C. was in receipt of exposure to diluted seawater. Mar Biol 132:29–42 an Australian Research Council Post-Graduate Award. The work Cooper AR, Morris S (1998b) The blood respiratory, haemato- was carried out under animal ethics approval LO4/9-94/3/1079 logical, acid–base and ionic status of the Port Jackson shark, and supported by funds from Morlab and Natural Events Heterodontus portusjacksoni, during recovery from anaesthesia (http://www.natural-events.com). and surgery: a comparison with sampling by direct caudal puncture. Comp Biochem Physiol 119:895–903 Cooper AR, Morris S (2003) Haemoglobin function and respira- tory status of the Port Jackson shark, Heterodontus portus- References jacksoni, in response to lowered salinity. J Comp Physiol B (In press) Albers C (1970) Acid–base balance. In: Hoar WS, Randall DJ (eds) Curtis BJ, Wood CM (1992) Kidney and urinary bladder responses Fish physiology, Vol IV. Academic Press, New York, pp of freshwater rainbow trout to isosmotic NaCl and NaHCO3 173–208 infusion. J Exp Biol 173:181–203 Albers C, Pleschka K (1967) Effect of temperature on CO2 trans- De Vlaming VL, Sage M (1973) Osmoregulation in the euryhaline port in elasmobranch blood. Respir Physiol 2:261–273 elasmobranch, Dasyatis sabina. Comp Biochem Physiol 45:31– Beyenbach KW, Kirschner LB (1976) The unreliability of mam- 44 malian glomerular markers in teleostean renal studies. J Exp Duthie GG, Tort L (1985) Effects of dorsal aortic cannulation Biol 64:369–378 on the respiration and haematology of Mediterranean living Boutilier RG, Heming TA, Iwama GK (1984) Physiological Scyliorhinus canicula L. Comp Biochem Physiol 81:879–883 parameters for use in fish respiratory physiology. In: Hoar WS, Eddy FB, Lomholt JP, Weber RE, Johansen K (1977) Blood Randall DJ (eds) Fish physiology, vol X. Academic Press, New respiratory properties of rainbow trout (Salmo gairdneri) kept York, pp 403–430 in water of high CO2 tension. J Exp Biol 67:37–47 Burger JW (1965) Roles of the rectal gland and kidneys in salt and Evans DH (1993) Osmotic and ionic regulation. In: Evans DH (ed) water secretion in the spiny dogfish. Physiol Zool 38:191–196 The physiology of fishes. CRC, Florida, pp 315–341 Burger JW (1967) Problems in the electrolyte economy of the spiny Forster RP, Goldstein L (1976) Intracellular osmoregulatory role dogfish Squalus acanthias. In: Gilbert PW, Mathewson RF, of amino acids and urea in marine elasmobranchs. Am J Rall DP (eds) Sharks, skates and rays. Johns Hopkins, Balti- Physiol 230:925–931 more, pp 177–185 Gilmour KM, Perry SF (1996) The effects of experimental anaemia Burton RF (1987) On calculating concentrations of ‘‘HCO3’’ from on CO2 excretion in rainbow trout, Oncorhynchus mykiss. Fish pH and Pco2. Comp Biochem Physiol 87:417–422 Physiol Biochem 15:259–272 Bushnell PG, Lutz PL, Steffensen JF, Oikari A, Gruber SH (1982) Gilmour KM, Perry SF, Bernier NJ, Henry RP, Wood CM (2001) Increases in arterial blood oxygen during exercise in the lemon Extracellular carbonic anhydrase in the dogfish, Squalus shark (Negaprion brevirostris). J Comp Physiol 147:41–47 acanthias: a role in CO2 excretion. Physiol Biochem Zool Butler PJ, Metcalfe JD (1988) Cardiovascular and respiratory 74:477–492 systems. In: Shuttleworth TJ (ed) Physiology of elasmobranch Gilmour KM, Shah B, Szebedinszky C (2002) An investigation of fishes. Springer, Berlin Heidelberg New York, pp 1–47 carbonic anhydrase in the gills and blood plasma of brown Butler PJ, Taylor EW, Davison W (1979) The effect of long-term, bullhead (Ameiurus nebulosus), longnose skate (Raja rhina), and moderate hypoxia on acid–base balance, plasma catecholam- spotted ratfish (Hydrolagus colliei) J Comp Physiol B 172:77–86 ines and possible anaerobic end products in the unrestrained Goldstein L, Forster RP (1971) Osmoregulation and urea metab- dogfish Scyliorhinus canicula. J Comp Physiol 132:297–303 olism in the little skate Raja erinacea. Am J Physiol 220:742–746 Chan DKO, Wong TM (1977a) Physiological adjustments to Greenaway P (1981) Sodium regulation in the freshwater/land crab dilution of the external medium in the lip-shark, Hemiscyllium Holthuisana transversa. J Comp Physiol 142:451–456 plagiosum (Bennett) I. Size of body compartments and osmolyte Greenaway P, Taylor HH, Morris S (1990) Adaptations to a ter- composition. J Exp Zool 200:71–84 restrial existence by the robber crab Birgus latro. VI. The role of Chan DKO, Wong TM (1977b) Physiological adjustments to the excretory system in fluid balance. J Exp Biol 152:505–519 dilution of the external medium in the lip-shark, Hemiscyllium Grigg GC (1974) Respiratory function of blood in fishes. In: plagiosum (Bennett). III. Oxygen consumption and metabolic Florkin M, Scherr BT (eds). Chemical zoology, vol VIII. Aca- rate. J Exp Zool 200:97–102 demic Press, New York, pp 331–368 Chantler C, Garnett ES, Parsons V, Veall N (1969) Glomerular Haywood GP (1974) The exchangeable ionic space, and salinity filtration-rate measurement in man by the single injection effects upon ion, water, and urea turnover rates in the dogfish, method using 51Cr-EDTA. Clin Sci 37:169–180 Poroderma africanum. Mar Biol 26:69–75 Choe KP, Evans D (2003) Compensation for hypercapnia by a Heisler N (1982) Transepithelial ion transfer processes as euryhaline elasmobranch: effect of salinity and role of gills and mechanisms for fish acid–base regulation in hypercapnia and kidneys in fresh water. J Exp Zool A 297:52–63 lactacidosis. Can J Zool 60:1108–1122 Claiborne JB, Heisler N (1984) Acid–base regulation in the carp Heisler N (1988) Acid–base regulation. In: Shuttleworth TJ (ed) (Cyprinus carpio) during and after exposure to environmental Physiology of elasmobranch fishes. Springer, Berlin Heidelberg hypercapnia. J Exp Biol 108:25–38 New York, pp 215–252 Claiborne JB, Walton JS, Compton-McCullough D (1994) Acid– Heisler N, Toews DP, Holeton GF (1988) Regulation of ventilation base regulation, branchial transfers and renal output in a and acid–base status in the elasmobranch Scyliorhinus stellaris marine teleost fish (the long-horned sculpin Myoxocephalus during hyperoxia induced hypercapnia. Respir Physiol 71:227– octodecimspinosus) during exposure to low salinities. J Exp Biol 246 193:79–95 Heming TA, Randall DJ, Boutilier RG, Iwama GK, Primmett D Claiborne JB, Edwards SL, Morrison-Shetlar AI (2002) Acid–base (1986) Ionic equilibria in red blood cells of rainbow ) ) + regulation in fishes: cellular and molecular mechanisms. J Exp trout (Salmo gairdneri): Cl , HCO3 and H . Respir Physiol Zool 193:79–95 65:223–234 221

Henderson IW, Brown JA, Oliver JA, Haywood GP (1978) Hor- Payan P, Goldstein L, Forster RP (1973) Gills and kidneys in mones and single nephron function in fishes. In: Gaillard PJ, ureosmotic regulation in euryhaline skates. Am J Physiol Boer HH (eds) Comparative endocrinology. Elsevier, North 224:367–372 Holland Biomedical, Amsterdam, pp 217–222 Perry SF, Gilmour KM (1993) An evaluation of factors limiting Henry RP, Swenson ER (2000) The distribution and physiological carbon dioxide excretion by trout red blood cells in vitro. J Exp significance of carbonic anhydrase in vertebrate gas exchange Biol 180:39–54 organs. Respir Physiol 121:1–12 Perry SF, Gilmour KM (2002) Sensing and transfer of respiratory Hickman CP, Trump BF (1969) The kidney. In: Hoar WS, Randall gases at the fish gill. J Exp Zool 293:249–263 DJ (eds) Fish physiology, vol I. Academic Press, New York, Perry SF, Heming TA (1981) Blood ionic and acid–base status in pp 91–239 rainbow trout (Salmo gairdneri) following rapid transfer from Holmes WN, Donaldson EM (1969) Body compartments and freshwater to seawater: effect of pseudobranch denervation. distribution of electrolytes. In: Hoar WS, Randall DJ (eds) Fish Can J Zool 59:1126–1132 physiology, vol I. Academic Press, New York, pp 1–89 Perry SF, Davie PS, Daxboeck C, Randall DJ (1982) A comparison Hopkins TE, Cech JJ (2003) The influence of environmental of CO2 excretion in spontaneously ventilating blood-perfused variables on the distribution and abundance of three elasmo- trout preparation and saline-perfused gill preparations: contri- branchs in Tomales Bay, California. Environ Biol Fishes bution of the branchial epithelium and red blood cell. J Exp 66:279–291 Biol 101:47–60 Iwama GK, Heisler N (1991) Effect of environmental water salinity Piermarini PM, Evans DH (1998) Osmoregulation of the Atlantic on acid–base regulation during environmental hypercapnia in stingray (Dasyatis sabina) from the freshwater Lake Jesup of the rainbow trout (Oncorhynchus mykiss). J Exp Biol 158:1–18 the St. Johns River, Florida. Physiol Zool 85:553–560 Janech MG, Piermarini PM (1997) Urine flow rate and urine Piiper J (1986) Gas exchange and acid–base status. In: Heisler N composition of freshwater Atlantic stingrays, Dasyatis sabina, (ed) Acid–base regulation in animals. Elsevier, Amsterdam, from the St, Johns River, FL. Comp Physiol Biochem Published pp 49–81 abstracts:147 Piiper J, Baumgarten-Schumann D (1968a) Effectiveness of O2 and Janssen RG, Randall DJ (1975) The effects of changes in pH and CO2 exchange in the gills of the dogfish (Scyliorhinus stellaris). Pco2 in blood and water on breathing in rainbow trout, Salmo Respir Physiol 5:338–349 gairdneri. Respir Physiol 25:235–245 Piiper J, Baumgarten-Schumann D (1968b) Transport of O2 and Jensen FB, Weber RE (1982) Respiratory properties of tench blood CO2 by water and blood in gas exchange of the dogfish (Scy- and hemoglobin. Adaptation to hypoxic-hypercapnic water. liorhinus stellaris). Respir Physiol 5:326–337 Mol Physiol 2:235–250 Piiper J, Schumann D (1967) Efficiency of O2 exchange in the gills Lai NC, Graham JB, Burnett L (1990) Blood respiratory properties of the dogfish, Scyliorhinus stellaris. Respir Physiol 2:135–148 and the effect of swimming on blood–gas transport in the Pleschka K, Wittenbrock I (1971) The solubility of carbon dioxide leopard shark Triakis semifasciata. J Exp Biol 151:161–173 in elasmobranch plasma between 10EC and 22EC. Pflugers Last PR, Stevens JD (1994) Sharks and rays of Australia. CSIRO, Arch Ges Physiol 329:186–190 Australia Rall DP, Sheldon W (1961) Transport of organic dyes by the Lenfant C, Johansen K (1966) Respiratory function in the elas- choroid plexus of the spiny dogfish, Squalus acanthias. Biochem mobranch Squalus suckleyi G. Respir Physiol 1:13–29 Pharmacol 11:169 Lin H, Randall D (1995) Proton pumps in fish gills. In: Shuttle- Schmidt-Nielsen B, Truniger B, Rabinowitz L (1972) Sodium- worth TJ, Wood CM (eds) Cellular and molecular approaches linked urea transport by the renal tubule of the spiny dogfish, to fish ionic regulation. Academic Press, San Diego, pp 229–255 Squalus acanthias. Comp Biochem Physiol 42:13–25 Maas AHJ, Rispens P, Siggaard-Anderson O, Zijlstra WG (1984) Shaw J (1963) Kinetic aspects of ion regulation in aquatic animals. On the reliability of the Henderson–Hasselbalch equation in In: Carthy JD, Duddington CL (eds) Viewpoints in biology, routine clinical acid–base chemistry. Ann Clin Biochem 21:26– vol II. Butterworths, London 39 Shuttleworth TJ (1988) Salt and water balance—extrarenal mech- Madsen SS, Larsen BK, Jensen FB (1996) Effects of freshwater to anisms. In: Shuttleworth TJ (ed) Physiology of elasmobranch seawater transfer on osmoregulation, acid–base balance and fishes. Springer, Berlin Heidelberg New York, pp 171–199 respiration in river migrating whitefish (Coregonus lavaretus). Smith HW (1931a) The absorption and excretion of water and salts J Comp Physiol B 166:101–109 by the elasmobranch fishes. II. Marine elasmobranchs. Am Maetz J, Payan P, DeRenzis G (1976) Controversial aspects of J Physiol 98:296–310 ionic uptake in fresh-water animals. In: Spencer-Davies P (ed) Smith HW (1931b) The absorption and excretion of water and salts Perspectives in experimental biology, vol I. Plenum, New York, by the elasmobranch fishes. I. Freshwater elasmobranchs. Am pp 77–92 J Physiol 98:279–295 Maxime V, Peyraud-Waitzenegger M, Claireaux G, Peyraud C Sulikowski JA, Maginniss LA (2001) Effects of environmental (1990) Effects of rapid transfer from sea water to fresh water on dilution on body fluid regulation in the yellow stingray, Urol- respiratory variables, blood acid–base status and O2 affinity of ophus jamaicensis. Comp Biochem Physiol 128:223–232 hemoglobin in Atlantic salmon (Salmo salar L.). J Comp Tetens V, Lykkeboe G (1981) Blood respiratory properties of Physiol 160:31–39 rainbow trout, Salmo gairdneri: responses to hypoxia acclima- McKim JM, Kolanczyk RC, Lien GJ, Hoffman AD (1999) tion and anoxic incubation of blood in vitro. J Comp Physiol Dynamics of renal excretion of phenol and major metabolites in 145:117–125 the rainbow trout (Oncorhynchus mykiss). Aquatic Toxicol Thorson TB (1958) Measurement of the fluid compartments of four 45:265–277 species of marine Chondrichthyes. Physiol Zool 31:16–23 McLaughlin RH, OGower AK (1970) Underwater tagging of the Thorson TB (1961) The partitioning of body water in Osteichthyes: Port Jackson shark, Heterodontus portusjacksoni (Meyer). Bull phylogenetic and ecological implications in aquatic vertebrates. Inst Oce´anogr Monaco 69:1–11 Biol Bull 120:238–254 McLaughlin RH, OGower AK (1971) Life history and underwater Thorson TB, Cowan CM, Watson DE (1973) Body fluid solutes of studies of a heterodont shark. Ecol Monogr 41:271–289 juveniles and adults of the euryhaline bull shark Carcharinus Nonnotte G, Truchot JP (1990) Time course of extracellular acid– leucas from freshwater and saline environments. Physiol Zool base adjustment under hypo- or hyperosmotic conditions in the 46:29–42 euryhaline fish Platichthys flesus. J Fish Biol 36:181–190 Thorson TB, Brooks DA, Mayes MA (1983) The evolution Pang PKT, Griffith RW, Atz JW (1977) Osmoregulation in elas- of freshwater adaptation in stingrays. National Geographic mobranchs. Am Zool 17:365–377 Society Research Reports (1974 Projects) 15:663–694 222

Toews DP, Holeton GF, Heisler N (1983) Regulation of the acid– lium in freshwater teleost fish: an immunological approach to base status during environmental hypercapnia in the marine ion-transport protein localization. J Exp Biol 203:2279–2296 teleost fish Conger conger. J Exp Biol 107:9–20 Wood CM, McMahon BR, McDonald DG (1979a) Respiratory, Truchot JP (1987) Comparative aspects of extracellular acid–base ventilatory, and cardiovascular responses to experimental balance. Springer, Berlin Heidelberg New York anaemia in the starry flounder, Platichthys stellatus. J Exp Biol Truchot JP, Toulmond A, Dejours P (1980) Blood acid–base bal- 82:139–162 ance as a function of water oxygenation: a study at two different Wood CM, McMahon BR, McDonald DG (1979b) Respiratory ambient CO2 levels in the dogfish, Scyliorhinus canicula. Respir gas exchange in the resting starry flounder, Platichthys stellatus. Physiol 41:13–28 A comparison with other teleosts. J Exp Biol 78:167–179 Tufts BL, Perry SF (1998) Carbon dioxide transport and excretion. Wood CM, McDonald DG, McMahon BR (1982) The influence of In: Perry SF, Tufts B (eds) Fish respiration. Academic Press, experimental anaemia on blood acid–base regulation in vivo London, pp 229–282 and in vitro in the starry flounder (Platichthys stellatus) and the Weber RE, Wells RMG, Rossetti JE (1983) Allosteric interactions rainbow trout (Salmo gairdneri). J Exp Biol 96:221–237 governing oxygen equilibria in the haemoglobin system of the Wood CM, Wheatly MG, Ho¨ be H (1984) The mechanisms of acid– spiny dogfish, Squalus acanthias. J Exp Biol 103:109–120 base and ionoregulation in the freshwater rainbow trout during Wells RMG, Weber RE (1983) Oxygenational properties and environmental hyperoxia and subsequent normoxia. III. Bran- phosphorylated metabolic intermediates in blood erythrocytes chial changes. Respir Physiol 55:175–192 of the dogfish, Squalus acanthias. J Exp Biol 103:95–108 Wood CM, Perry SF, Walsh PJ, Thomas S (1994) HCO3) dehy- Wilson JM, Laurent P, Tufts BL, Benos DJ, Donowitz M, Vogl dration by the blood of an elasmobranch in the absence of a AW, Randall DJ (2000) NaCl uptake by the branchial epithe- Haldane effect. Respir Physiol 98:319–337