Oxygen uptake, diffusion limitation, and diffusing ...

Comparative Biochemistry and Physiology, Part A 143 (2006) 299 – 306 www.elsevier.com/locate/cbpa

Oxygen uptake, diffusion limitation, and diffusing capacity of the bipectinate gills of the abalone, Haliotis iris (Mollusca: Prosobranchia) Norman L.C. Ragg ⁎, H. Harry Taylor School of Biological Sciences, University of Canterbury, Private Bag 2400 Christchurch, New Zealand Received 23 June 2005; received in revised form 30 November 2005; accepted 1 December 2005 Available online 31 January 2006

Abstract Extant abalone retain an ancestral system of gas exchange consisting of paired bipectinate gills. This paper examines the hypothesis that fundamental inefficiencies of this arrangement led to the extensive radiation observed in prosobranch gas exchange organs. Oxygen uptake at 15 °C was examined in the right gill of resting adult blackfoot abalone, Haliotis iris Martyn 1784. Pre- and post-branchial haemolymph and water were sampled and oxygen content, partial pressure (Po2), pH, and haemocyanin content measured; in vivo haemolymph flow rate was determined by an acoustic pulsed-Doppler flowmeter. During a single pass across the gills, mean seawater Po2 fell from 138.7 Torr to 83.4 Torr, while haemolymph Po2 rose from 37.2 Torr to 77.0 Torr raising total O2 content from 0.226 to 0.346 mmol L− 1. Haemolymph flowed through the right gill at a mean rate of 9.6 mL min− 1 and carried 0.151 to 0.355 mmol L− 1 of haemocyanin (mean body mass 421 g). Only 34.7% of the oxygen carried in the arterial haemolymph was taken up by the tissues and less than half of this was contributed by haemocyanin. A diffusion limitation index (Ldiff) of 0.47–0.52, a well-matched ventilation–perfusion ratio (1.2–1.4) and a diffusing capacity (D) of 0.174 μmol O2 kg− 1 Torr− 1 indicate that the gills operate efficiently and are able to meet the oxygen requirements of the resting abalone. © 2005 Elsevier Inc. All rights reserved. Keywords: Diffusion limitation; Bipectinate gills; Abalone; Haliotis iris; Evolution

1. Introduction Evolutionary processes tend to exert pressure upon gas exchange organ design both directly and indirectly, as metabolic rates increase. In aquatic gas exchangers these evolutionary responses may increase the efficiency with which oxygen diffuses from water to blood, as well as accommodating the hydrostatic and hydrodynamic forces associated with increased blood flow and ventilation. Numerous physiological studies of fish provide an indication of the evolution of gas exchanger efficiency and structure in response to the demands of increasingly dynamic pelagic lifestyles (Jones and Randall, 1978; Hughes, 1982; Randall and Daxboeck, 1984). Gas exchange efficiency in decapod crustaceans has also received careful attention, both in relation to water breathing and the progressive adaptation to air (Scammell and Hughes, 1981; Mangum, 1983; McMahon and Wilkens, 1983; Taylor and ⁎ Corresponding author. Tel.: +64 3 3642861; fax: +64 3 3642590. E-mail address: [email protected] (N.L.C. Ragg). 1095-6433/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpa.2005.12.004

Taylor, 1992). Gastropod molluscs display perhaps the most diverse variety of gas exchange organ arrangements. Physiological efficiency of these structures and the effect of evolutionary processes are, however, rarely examined. Abalone, that collectively comprise the genus Haliotis and the family Haliotidae (Prosobranchia: Vetigastropoda), are generally regarded as primitive gastropods. This categorisation is based upon the morphological arrangement of extant species and upon the evolutionary stability of the family over geological time. For example, well preserved fossilized shells dated at 66 Ma have been assigned to the extant subgenus Paua (Lindberg, 1992), which has a single living member, Haliotis iris. The New Zealand blackfoot abalone, Haliotis iris Martyn 1784, displays a body plan characteristic of the family, including paired organs and a row of dorsal perforations (tremata) in the shell. Although the organs of Haliotids show a left-side bias due to the need to accommodate the large right shell adductor muscle (Yonge, 1947), their morphology demonstrates the bilateral symmetry and notched shell of the hypothetical ancestral gastropods (e.g., Voltzow, 1994, Voltzow et al., 2004).

300

N.L.C. Ragg, H.H. Taylor / Comparative Biochemistry and Physiology, Part A 143 (2006) 299–306

A conspicuous feature of prosobranch evolution has been the tendency to abandon paired gills in favour of many other arrangements. Initially, the trend was apparently towards gill reduction (Purchon, 1977; Fretter and Graham, 1994). More specialised vetigastropods (Trochoidea and Seguenzioidea) rely on a single bipectinate gill, while higher orders (Mesogastropoda, Neogastropoda) retain only ‘half a gill’, i.e., a monopectinate (pectinibranch) gill. However, many ecologically successful gastropod groups have either abandoned gills altogether and rely on cutaneous gas exchange, or have evolved secondary gills from other body surfaces (Kingston, 1968; Voltzow, 1994). In vetigastropods the gills are usually associated with secondary shell apertures. Ancestrally the aperture takes the form of a slot in extinct forms and some rare extant species, including Perotrochus sp. and Entemnotrochus sp., usually restricted to deep water (Voltzow et al., 2004). In abalone and the keyhole limpets, the slit has been partially occluded, forming one or more tremata. Distribution of these animals is usually restricted to rocky sub-littoral coastlines where they experience strong water movements and high levels of dissolved oxygen (Barnes, 1986). Water movement across the perforated shell drives water through the branchial chamber by a combination of ram and Bernoulli effects (Vogel, 1994) and provides augmentation to the endogenous ciliary ventilation (Murdock and Vogel, 1978; Voltzow, 1983; Taylor and Ragg, 2005a). Thus, a picture has been developed of abalone as animals possessing an archaic gill design that requires external assistance to ventilate adequately, despite inhabiting a high Po2 environment. That most abalone are inherently sluggish and have a large facultative anaerobic capacity (Gäde, 1988; Baldwin et al., 1992) seems to corroborate this view. The present study was therefore designed to specifically examine the efficiency of abalone gills as gas exchangers and to consider the broader implication that limitations imposed by the bipectinate design may have encouraged the evolutionary radiation leading to modern prosobranch gas exchangers. Blood (more correctly ‘haemolymph’) flow in the right gill, and oxygen partial pressures and concentrations in the afferent and efferent blood and ventilatory water flows, were measured in naturally ventilating and artificially ventilated abalone. These values were then used to analyze gas exchange efficiency in terms of the diffusing capacity of the water–haemolymph barrier, D, (Scheid, 1982) and the diffusion limitation index, Ldiff (Piiper, 1982). These indices thus provide a basis for comparison of the abalone branchial system with that of other aquatic animals. The importance of haemocyanin in oxygen transport in the blood is also discussed. 2. Methods 2.1. Experimental animals Twenty seven adult Haliotis iris (Prosobranchia: Vetigastropoda, 250–580 g) collected from South Bay, Kaikoura, New Zealand were acclimated for 2 months in a seawater facility maintained at 15 °C and weaned onto an artificial pellet diet using the techniques of Allen et al. (2001).

2.2. Experimental treatments Abalone were held at rest at 15 °C in 1 L polycarbonate bowls of gently circulating sea water, replaced at 4 L h− 1. Oxygen partial pressures (Po2) and concentrations (Co2) and pH were determined in inhalant and exhalant water samples, and in afferent and efferent branchial haemolymph aspirated from indwelling cannulae. Simultaneously, measurements were made of haemolymph flow through the right gill, and of haemocyanin concentration. Sampling continued until 5 data sets were obtained from each animal ventilating naturally and in animals artificially ventilated via a mask covering the tremata (in a number of individuals cannula failure or loss of Doppler signal precluded obtaining a full dataset). 2.3. Animal preparation and surgery Abalone were starved for 24 h prior to surgery which was carried out in moist air at 4–6 °C. A section of shell bearing the exhalant tremata (all except occasionally the most anterior, forward-projecting holes, after Taylor and Ragg, 2005a), measuring approximately 40 × 10 mm, was removed with a hand-held grinding wheel (Dremel™), to expose the central region of the gills. Another window ∼10 × 10 mm, immediately posterior and to the right of the oldest patent shell hole, exposed the right efferent ctenidial vein. Further posterior, two 1.5 mm diameter holes were drilled either side of the heart and bareended copper leads (0.2 × 1 m) for measurement of heart rate were secured into each using cyanoacrylate gel. Each animal was then allowed to recover for 24 h in the holding system before proceeding further. The right efferent ctenidial vein was then punctured 20 mm anterior to the heart using a 23-gauge needle and a 10 mm length of cannula tubing (PE, 0.8 mm external diameter) was inserted retrograde to haemolymph flow. Pre-branchial haemolymph was sampled from the basibranchial sinus, which supplies both gills (Crofts, 1929). A small hole was made near the edge of the shell on the left side to support the cannula (PVC, 0.6 mm external diameter), which was inserted into a puncture near the anterior end of the left afferent ctenidial vein and fed through to the basibranchial sinus. Visual inspection confirmed the vein was not occluded by this procedure. For measurement of haemolymph flow through the right gill, probes were constructed by attaching a Doppler crystal subassembly (Iowa Doppler Products) to a short length of 0.86 mm internal diameter PE tubing. The assembly was slid over the outside of the efferent ctenidial cannula, coming to rest at its point of emergence from the vessel with the Doppler crystal resting on the thin mantle tissue overlying the vein at an angle of about 45° (Fig. 1). All leads and cannulae were secured to the shell with cyanoacrylate glue. The animal was then placed in the experimental chamber to recover for a further 24 h. On the measurement day, the cannulae were tested for patency, the cardiac leads were connected to an impedance coupler (Strathkelvin Instruments A100), the Doppler probe was connected to a directional pulsed-Doppler flowmeter (Bioengineering 545C-4) tuned to the fastest venous flow (largest

N.L.C. Ragg, H.H. Taylor / Comparative Biochemistry and Physiology, Part A 143 (2006) 299–306

301

Fig. 1. Schematic representation of retrograde venous cannula with customised pulsed-Doppler probe in position. The probe consists of a sub-assembly sealed in epoxy resin and mounted beneath a 0.86 mm∅ polyethylene sleeve. The 0.8 mm OD cannula is threaded through the sleeve prior to insertion into the vessel; the assembly is subsequently held in position on the tissue surface by the cannula, PVC wings prevent rolling and tension in the Doppler leads keeps the sleeve at the cannula insertion point.

signal), and signals were sampled at 100 Hz (PowerLab 4/20, Chart 4.1.2 software, ADInstruments). The ventilation mask consisted of a 10 cm section of 12 mm diameter PVC tubing cut obliquely to follow the shell contour. The mask was hot-glue sealed to the previously removed anterior shell section over the shell holes and the assembly reattached to the shell. A peristaltic pump drew water through the mask at 48 mL min− 1. 2.4. Haemolymph and water sampling procedure Each set of pre- and post-branchial haemolymph samples and inhalant and exhalant seawater samples was taken over a 10 min period in random order. For each haemolymph sample, 100 μL was first withdrawn (Hamilton Gastight™ syringe) to clear dead space, followed by a second 100 μL sample for determination of Po2 and pH (Microelectrodes Inc. MI-730 O2 electrode and Strathkelvin 781b meter, MI-710 pH electrode and Radiometer PHM84 meter, Cameron Instruments capillary microelectrode cell at 15 °C). The sample was then recovered, centrifuged at 9000 ×g for 3 min to remove cells, and frozen at − 18 °C for later haemocyanin analysis. A third 100 μL of haemolymph was used to determine Co2, as described below. The Po2 and pH of seawater were determined similarly. Mixed water in the holding chamber was assumed to represent inhalant seawater. In naturally ventilating abalone, exhalant water samples were taken from the shell window within the endogenous ciliary stream just above and between the two gills (c.f. Voltzow, 1983; Taylor and Ragg, 2005a). In preliminary trials a number of animals were sampled from this location before and after replacement of the shell section with tremata, without significant effect on the exhalant Po2 (F = 2.98, p = 0.10). In artificially ventilated abalone, the exhalant samples were taken from the mask. 2.5. Oxygen concentration of haemolymph Total oxygen content of the haemolymph (Co2, mmol.L− 1) was determined using a galvanic cell oxygen analyzer (‘OxyCon’ manufactured by the Department of Physiology, University of Tasmania), calibrated with air samples. Duplicate 50 μL haemolymph sub-samples were sequentially injected into a cuvette containing 1 mL of scrubber solution (6 g L− 1 potassium ferricyanide, 6 g L− 1 potassium cyanide, 1 mL L− 1 Triton-X, 1 mL L− 1 Sigma Antifoam-A, 15 g L− 1 Na3PO4·12H2O, 5 g L− 1 Na2HPO4) bubbled with a stream of nitrogen. The

concentration of oxygen bound to haemocyanin was determined by subtracting the dissolved oxygen component from the total content assuming that the oxygen capacitance of haemolymph plasma was the same as seawater (1.618 μmol L− 1·Torr− 1 at 15 °C). 2.6. Haemocyanin concentration of haemolymph Cell-free haemolymph samples were thawed, diluted 10fold with pH 8.8 buffer (10 mmol.L− 1 EDTA, 50 mmol L− 1 glycine) and oxygenated by shaking in air. Haemocyanin concentrations ([Hcy] mmol L− 1) were calculated from their absorbance at 346 nm using the extinction co-efficient reported by Behrens et al. (2002) for H. iris haemolymph, based on copper concentrations (EmM Cu, 1 cm = 11.42, i.e., EmM Hcy, 1 cm = 22.84 assuming 2 Cu atoms per haemocyanin functional unit bind one O2 molecule). 2.7. Blood flow and heart rate The continuous outputs of the impedance coupler and pulsed-Doppler meter were recorded during each haemolymph sampling period. Mean heart rate was estimated from the impedance signal peaks. Instantaneous blood flow velocity (mm.s− 1) was calculated from the Doppler output voltage and the characteristics of the instrument and probe. Mean volumetric flow (mL min − 1 ) was determined from the integrated Doppler signal calibrated in situ with preset flows delivered from a peristaltic pump. After each trial, animals were rapidly dispatched by decapitation, the ventricle was punctured, and a suspension of Zeolite (80 μm filtered barbecue deodorizer) in seawater was pumped into the right efferent ctenidial vein cannula at a series of preset flow rates. The Zeolite was initially observed to back-flow into the ctenidium until all efferent pores were blocked and then flowed past the Doppler crystal and drained from the ventricle. A linear relationship was observed between the Doppler signal and the flow rate (r2 N 0.7). 2.8. Calculations Total oxygen uptake from the branchial chamber water (including both gills), Mo2 (gills) (μmol.g− 1.h− 1) was calculated from: MO2 ðgillsÞ ¼ adFvd ðPi −Pe Þ=M

302

N.L.C. Ragg, H.H. Taylor / Comparative Biochemistry and Physiology, Part A 143 (2006) 299–306

where α the solubility of oxygen in seawater at 15 °C (1.6218 μmol L− 1 Torr− 1), Fv the ventilatory flow (0.048 L min− 1 in artificially ventilated abalone and assumed to be 0.084 L min− 1 kg− 1 in naturally ventilating abalone; Taylor and Ragg, 2005a), Pi − Pe is the difference between inhalant and exhalant Po2 (Torr; 1 Torr ≡ 133 Pa), and M the blotted live mass of the abalone, including shell (g). The diffusion limitation index, Ldiff, was determined at each sample time (after Piiper, 1982) as: Ldif f ¼ ðPm −Pa Þ=ðPm −Pv Þ where Pm, is the mean Po2 of inhalant and exhalant seawater, and Pa and Pv are the Po2 values of the efferent and afferent ctenidial haemolymph, respectively. The total diffusing capacity for oxygen of the gills and branchial chamber surfaces, D (Scheid, 1982; also referred to as transfer factor or diffusive conductance, Gdiff), was calculated from: D ¼ MO2 ðgillsÞ =ðPm −Ph Þ where Ph the branchial haemolymph Po2 (mean of afferent and efferent haemolymph) and is expressed in terms of total animal mass (as μmol kg− 1 min− 1 Torr− 1 to facilitate comparison with literature values). Other parameters used to describe the functioning of gas exchangers are defined and tabulated in the Discussion section. All statistical analyses were performed using Statistica™ 6.0 software (StatSoft Inc., USA). Significant differences were detected using paired 2-tail t-tests, one- and two-way anova. Where appropriate, within-individual variability was nested within treatment effect. Statistical significance was accepted at p b 0.05. Values are expressed as means ± standard error of the mean (n = number of animals). 3. Results A total of 27 animals yielded Po2 data, 12 of these were subjected to forced ventilation and also provided Co2 and [Hcy] data. 3.1. Partial pressure and pH gradients During unassisted ventilation the resting abalone lowered seawater Po2 from 138.7 ± 1.4 to 83.4 ± 3.1 Torr (n = 27) in its passage through the branchial chamber, corresponding to an oxygen extraction efficiency of 40%. When a mask was applied and the animal ventilated artificially at 48 mL min− 1 the mean exhalant seawater Po2 was 91.3 ± 3.3 Torr, which was not significantly different from that in naturally ventilating animals (t = − 1.67, p = 0.07, n = 9). In naturally ventilating abalone the Po2 of haemolymph passing through the right gill was raised significantly from 37.2 ± 3.6 to 77.0 ± 4.2 Torr (paired t = 11.3, p b 0.001, n = 27), approaching equilibrium with the exhalant water (Fig. 2, Table 1). The mean change in Po2 was 39.8 ± 3.5 Torr. During forced ventilation, the Po2 of afferent and the efferent ctenidial haemolymph (41.8 ± 4.8 and 80.8 ± 8.5 Torr, respectively, n = 9) did not differ significantly from the naturally

Fig. 2. Pre- and post-branchial haemolymph and seawater Po2 measured in adult Haliotis iris during natural ventilation and during forced ventilation at 48 mL min− 1. Error bars represent +1 SEM.

ventilated state (t = 0.117, p = 0.455 and t = − 0.374, p = 0.359, respectively). The mean (inhalant) pH of seawater in the experimental chambers was 7.65 ± 0.04 and this decreased to pH 7.23 ± 0.04 in the exhalant water of non-masked abalone. There was no significant difference between the pH values of afferent and efferent ctenidial haemolymph (7.16 ± 0.02 and 7.17 ± 0.03; Table 1). 3.2. Haemolymph flow and heart rate Doppler-flow records were successfully obtained from 24 resting abalone. Haemolymph flow in the right efferent ctenidial vein was highly pulsatile. Midstream flow velocity rose steeply, typically peaking at N40 mm s− 1 before abruptly falling to zero between pulses. Volume flow calibration was achieved in 12 animals (mean mass 421 g). Right gill haemolymph flow during unassisted ventilation was 9.60 ± 0.38 mL min− 1 (n = 12). In artificially ventilated abalone, haemolymph flow through the right gill showed a small, but significant increase to 11.96 ± 0.30 mL min− 1 (F = 6.98, p = 0.01). Mean heart rate in naturally ventilating animals was 29.1 ± 0.5 min− 1 and, under forced ventilation, was slightly but significantly higher at 30.7 ± 0.2 min − 1 (F = 8.30, p = 0.005). Heart rate varied inversely with animal mass (linear regression, r2 = 0.65, y-intercept 53.6 ± 1.8 min− 1, slope − 0.048 ± 0.004 min− 1 g− 1, n = 88). 3.3. Oxygen and haemocyanin concentrations of haemolymph Mean oxygen concentration of the haemolymph increased from 0.226 ± 0.031 mmol L− 1 in pre-branchial haemolymph to 0.346 ± 0.023 mmol L− 1 (n = 12) in efferent ctenidial samples. Subtracting dissolved oxygen indicated that the concentration of oxygen bound to haemocyanin increased significantly from

N.L.C. Ragg, H.H. Taylor / Comparative Biochemistry and Physiology, Part A 143 (2006) 299–306 Table 1 Oxygen content and partial pressures in pre- and post-branchial haemolymph of the right gill of resting Haliotis iris at 15 °C

Po2 (Torr) Co2 (mmol O2 L− 1) Hcy-bound O2 (mmol L− 1) [Hcy] (mmol L− 1) mol O2: mol Hcy pH

Pre-branchial haemolymph (Basibranchial sinus)

Post-branchial haemolymph (R. efferent ctenidial vein)

Statistical significance

37.2 ± 3.6 0.226 ± 0.031 0.160 ± 0.030

77.0 ± 4.2 0.346 ± 0.023 0.218 ± 0.026

⁎⁎ ⁎⁎ ⁎⁎

0.237 ± 0.020 0.63 ± 0.07 7.16 ± 0.02

0.237 ± 0.019 0.91 ± 0.05 7.17 ± 0.03

N/S ⁎⁎ N/S

Values represent mean ± sem; data are paired, n = 27 for Po2 data, n = 12 for all other values. Statistical differences between pre- and post-branchial data are indicated; N/S = no significant difference (p N 0.05), ⁎ = significant difference (p b 0.05), ⁎⁎ = highly significant difference (p b 0.001). Test details are given in the body of the text.

0.160 ± 0.030 to 0.218 ± 0.026 mmol L− 1 (anova F = 68.49, p b 0.001; Table 1). Haemocyanin concentration estimated from absorbance at 346 nm was similar in pre- and post-branchial haemolymph (0.237 ± 0.020 mmol L− 1; Table 1) although varying widely among individuals (0.15 to 0.36 mmol L− 1). Thus, the molar ratio of O2:haemocyanin (i.e., saturation) rose significantly from 0.63 ± 0.07 to 0.91 ± 0.05 (Table 1; F = 71.62, p b 0.001). The average oxygen capacitance coefficient of haemolymph (βh) perfusing the right gill was determined in 11 individual abalone for which eight or more paired measurements of Co2 and Po2 were available spanning the arterial and venous range (mean Po2 37–77 Torr, Table 1). As this range lies above the steepest part of the oxygen equilibrium curve (see Discussion section), βh was satisfactorily approximated by the linear regression slope (mean 2.64 ± 0.17 μmol L− 1 Torr− 1; c.f. βseawater = 1.62 μmol L− 1 Torr− 1; Dejours, 1981). 3.4. Oxygen uptake and the role of haemocyanin The total rate of oxygen uptake by both gills from seawater passing throughthe branchial chamber, M˙o2(gills) and the rate of delivery of oxygen into the haemolymph from the right gill M˙o2(R gill) were determined in eight artificially ventilated abalone for which haemolymph flow measurements and at least three measurements for each of Pio2, Peo2, Pvo2, and Pao2 were available. Mean M˙o2(gills) was 0.539 ± 0.041 μmol g− 1 h− 1 and mean M˙o2(R gill)was 0.199 ± 0.049 μmol g− 1 h− 1 (n = 8), accounting for 39.0 ± 12.2% of uptake. 3.5. Indices of diffusion The mean diffusion limitation index, Ldiff, for the right gill was 0.47 ± 0.05 in naturally ventilating abalone water and 0.52 ± 0.13 during forced ventilation (the difference was not significant, paired t-test, n = 8). Within individuals, Ldiff was relatively constant over a period of several hours but among individuals there were quite large differences (range from 0.25 ± 0.08 to 0.75 ± 0.04, F = 5.26, p b 0.001).

303

The total diffusing capacity of the branchial chamber for oxygen, standardized to the live mass of the animal, D, was 0.174 ± 0.021 μmol O2 kg− 1 min− 1 Torr− 1 (n = 8). Replicate estimates of D were relatively constant within individuals but varied significantly between individuals (range 0.123 ± 0.011 to 0.296 ± 0.027 μmol O2 kg− 1 min− 1 Torr− 1). 4. Discussion 4.1. The role of the haemolymph in oxygen transport and storage The mean post-branchial oxygen content (Co2) and haemocyanin concentrations of the haemolymph of H. iris (0.35 and 0.24 mmol L− 1, respectively) were similar to values reported for other sedentary abalone such as H. laevigata and H. rubra (Co2 0.46 and 0.45 mmol.L− 1, respectively; Ainslie, 1980), but somewhat lower than in more active gastropods such as H. roei (Co2 0.64 mmol L− 1; Ainslie, 1980) or Busycon (Co21.6 mmol L− 1; Mangum and Polites, 1980). Given the very large haemolymph volume of H. iris, (∼57 mL 100 g− 1 tissue, 44 mL 100 g− 1 including the shell; Taylor and Ragg, 2005b), it is likely that one of the factors militating against higher haemocyanin concentration is the cost of its synthesis. Benefits of low haemocyanin concentrations also include the avoidance of problems of high oncotic pressure (Mangum, 1983) and viscosity (Wells and Smith, 1987) in low pressure open vascular systems. The modest haemocyanin concentration and large circulating volume of H. iris support earlier conclusions (Wells et al., 1998; Behrens et al., 2002) that the function of H. iris haemocyanin is directed more towards storage of oxygen to support aerobic tissues during periods of reduced oxygen availability (e.g., during clamping to the substratum) than towards oxygen transport. A storage function is also implied by the relatively high values of both pre- and post-branchial oxygen pressures recorded here (Table 1, Fig. 2) in relation to the strong oxygen affinity and reverse Bohr shift of H. iris haemocyanin (P50 ∼4 Torr at 15 °C and pH 6.9 increasing to 10–20 Torr at pH 7.7; Wells et al., 1998; Behrens et al., 2002). This interpretation is also supported by oxygen content data for the right gill haemolymph (Table 1), which indicate that only 35% of the oxygen transported from the right efferent ctenidial vein was consumed by the tissues, less than half of which (48%) was donated by haemocyanin. The venous reserve of 0.226 mmol L− 1 (Table 1) represents about 90 μmol kg− 1 body mass of oxygen which would last for about 10 min at the resting rate of oxygen consumption but considerably longer at rates measured in hypoxia (Taylor and Ragg, 2005a). By contrast, Ainslie (1980) inferred that in three Australian haliotids (H. rubra, H. laevigata, and H. roei) 86–97% of oxygen delivered to the tissues was donated by haemocyanin. The more effective exploitation of haemocyanin in these species apparently resulted from lower oxygen binding affinities than in H. iris. Among other molluscs, the cephalopod Nautilus pompilius and the amphineuran Cryptochiton stelleri similarly unload relatively small proportions of their haemolymph

304

N.L.C. Ragg, H.H. Taylor / Comparative Biochemistry and Physiology, Part A 143 (2006) 299–306

oxygen in the tissues (∼40% and 48%, respectively), retaining a large venous reserve (Mangum and Polites, 1980), whereas octopus, squid, and the gastropod Busycon canaliculatum unload N80% of their haemolymph oxygen (Mangum and Polites, 1980; Wells and Smith, 1987; Wells, 1992). In artificially ventilated H. iris, 40% of the oxygen taken up from the seawater entered the haemolymph of the right gill. Uptake by the slightly larger left gill might therefore account for most of the remaining uptake. However, ciliary pumping is considered to be energetically costly (Vogel, 1994) and high rates of oxygen consumption have been demonstrated in molluscan gill epithelia (Busycon, Mangum and Polites, 1980; Crassostrea, Willson and Burnett, 2000). In the gastropods Hemifusus and Busycon (Depledge and Phillips, 1986) and the bivalves Mytilus and Modiolus (Booth and Mangum, 1978; Famme, 1981) direct uptake by tissues accounted for relatively large proportions of total oxygen uptake. Thus further studies are needed to partition oxygen uptake between transport across the two gills and consumption by the epithelia of the branchial chamber in H. iris. 4.2. Ventilation–perfusion matching Efficient operation of aquatic gas exchangers involves minimal cost of pumping fluids and this is partly achieved by optimal matching of ventilation and perfusion flows. The ventilation/perfusion ratio (Qw / Qh) of the right gill of H. iris was estimated to be 1.4 during endogenous ciliary ventilation and 1.2 for artificially ventilated abalone (Table 2). These values are low compared with many other water breathers, e.g., up to 4 in decapod crustaceans and up to 20 in fish gills (Jones and Randall, 1978; Randall and Daxboeck, 1984; Taylor and Taylor, 1992) but are consistent with the generally higher oxygen carrying capacity of the blood of the latter groups. A more useful index of ventilation/perfusion matching is the ratio of ventilatory water conductance to haemolymph conductance (Qw · βw / Qh · βh), which takes into account both oxygen capacitance and flow (Scheid, 1982). The present estimate of

0.96 for the right gill of naturally ventilating resting abalone (Table 2) is close to the optimal value of 1.0 and indicates that, on this basis at least, employment of ciliary currents for branchial ventilation should not be dismissed as an inefficient mechanism. 4.3. Branchial diffusion of oxygen In gill systems, diffusion across the water–blood barrier is often the rate-limiting step in the transfer of oxygen between the environment and the mitochondria (Piiper, 1982). However, oxygen partial pressures in the branchial water and blood of H. iris imply quite effective equilibration between these fluids with the Po2 of the right efferent ctenidial blood approaching that of the exhalant water (Fig. 2). More formally, the extent of diffusion limitation in a gas exchanger may be expressed by the diffusion limitation index, Ldiff, introduced by Piiper (1982). The index returns a value between 0 and 1; a value approaching 1 indicates that oxygen transfer is entirely determined by a low diffusing capacity at the blood/water interface whereas a value close to zero indicates a perfusionlimited system in which the barrier offers negligible resistance to diffusion and blood flow through the organ determines oxygen uptake. Mean Ldiff values in the range 0.47–0.52 obtained for the right gill of H. iris therefore suggest that diffusion and perfusion acted equally to limit oxygen uptake and that blood flow was well-matched to the diffusing capacity of the gill under the conditions of these measurements. These values are comparable with Ldiff values obtained in fish (e.g., 0.6 in Scyliorhinus stellarius; Piiper, 1982) and are lower than values reported for aquatic decapod crustaceans (0.5–0.9 Taylor and Taylor, 1992) in which the relatively thick branchial cuticle and epithelium are believed to provide a significant barrier to diffusion. Few data are available for diffusion limitation in other molluscs. From Po2 values published for the conch, Busycon (Mangum and Polites, 1980), an Ldiff of 0.37 was estimated for the pectinibranch gill plus mantle.

Table 2 Parameters characterizing gas exchange in the right gill of H. iris (after Piiper and Scheid, 1984) Parameter

Definition1

Artificially ventilated 2

Naturally ventilating 2

Statistical significance

Diffusion limitation index (Ldiff) Haemolymph flow (mL min− 1) Ventilation extraction efficiency, % Ventilation / perfusion ratio Ventilatory / perfusive conductance ratio Diffusing capacity (D, μmol O2 kg− 1 min− 1 Torr− 1)

(Pm − Pa) / (Pm − Pv) Qh 100.(Pi − Pe) / Pi Qw · Qh Qw · βw / Qh · βh M˙o2 (gills) / (Pm − Ph)

0.52 ± 0.13 11.96 ± 0.30 36.1 ± 2.2 1.21 ± 0.09 1.41 ± 0.18 0.174 ± 0.021

0.47 ± 0.05 9.60 ± 0.38 39.8 ± 2.2 1.4 ± 0.2 0.96 ± 0.15

N/S ⁎ N/S N/S ⁎

Values are means ± sem (n = 8 for all data except ventilation extraction efficiency, where n = 27 for naturally ventilated and n = 9 for artificially ventilated individuals). Statistical differences between naturally and artificially ventilated data are indicated; N/S = no significant difference (p N 0.05), ⁎ = significant difference (p b 0.05), ⁎⁎ = highly significant difference (p b 0.001). Ventilation extraction efficiency: unpaired t = 0.92, p = 0.36. Ventilation / perfusion ratio paired t = 1.40, p = 0.20. Ventilatory / perfusive conductance ratio paired t = − 3.64, p = 0.008. Remaining test details are given in the body of the text. 1 Pi and Pe are the Po2 of inhalant and exhalant water, respectively, and Pm is their mean; Pv and Pa the pre-branchial (‘venous’) and post-branchial (‘arterial’) haemolymph Po2 values, and Ph is their mean. Qw and Qh represent the flow rates of ventilatory water and haemolymph, respectively. βh is the oxygen capacitance of the haemolymph (2.64 ± 0.17 μmol L− 1 Torr− 1) in the arterio-venous Po2 range. M˙o2 (gills) is the net O2 uptake from the ventilatory water stream (μmol g− 1 h− 1). Mean flesh mass of abalone was 421 g. Temperature 15 °C. 2 The calculations assume that the right gill receives half of the ventilatory flow (114 mL min− 1 kg− 1 in artificially ventilated abalone; 84 mL min− 1 kg− 1 in endogenously ventilating abalone; Taylor and Ragg, 2005a).

N.L.C. Ragg, H.H. Taylor / Comparative Biochemistry and Physiology, Part A 143 (2006) 299–306

Besides the absence of a cuticle, diffusion across abalone gills is likely to be facilitated by several other aspects of their design. 1) Ciliary ventilation would reduce the establishment of unstirred layers. 2) Branchial transmural pressures in abalone are very low (0.7–2 Torr, estimated from measurements of prebranchial haemolymph pressures in H. corrugata, Bourne and Redmond, 1977; H. kamtschatkana, Krajniak and Bourne, 1988; H. rubra, Russell and Evans, 1989; H. iris, Ragg, 2003) compared, for example, to fish gill lamellae (20–40 Torr; Randall and Daxboeck, 1984), permitting a much thinner lamellar epithelium. 3) Ciliary ventilation establishes a countercurrent ventilation–perfusion relationship at the level of the gill lamellae (Yonge, 1947; Fretter and Graham, 1994; Wanichanon et al., 2004). Piiper and Scheid (1984) suggest that Ldiff must be less than 0.37 before a countercurrent exchanger has an inherent advantage over a co-current system. Interestingly, Ldiff measured in fish rarely falls this low, but many individual abalone had values in this range. While Ldiff is an operational index that describes whether an animal is effectively under- or over-perfusing its respiratory organs in relation to their diffusing capacity, the magnitude of the total diffusing capacity (D) is an equally important physiological/structural adaptation of gills for gas exchange. D expresses how effectively an animal is able to exploit the medium/blood Po2 gradient. The present estimate of 0.174 μmol O2 kg− 1 min− 1 Torr− 1 for H. iris relates to the total diffusing capacity of both gills normalized to animal live mass. The value is comparable with determinations of D in moderately active fish such as the haemoglobinless icefish, Chaenocephalus aceratus (0.12 μmol O2 kg− 1 min− 1 Torr− 1), catfish and flounder (∼0.3 μmol O2 kg− 1 min− 1 Torr− 1) but rather lower than in the more active dogfish (0.4–0.67 μmol O2 kg − 1 min − 1 Torr − 1 ) (Johansen, 1982; Piiper, 1982). Comparisons with other invertebrate gills are more difficult as most published values of D have been derived from morphometric data. ‘Morphometric D’ estimates generally ignore the potential effects of unstirred layers, mucus or non-functional regions of the exchanger. Thus the very high D values calculated for Nautilus (2.4–6.2 μmol O2 kg− 1 min− 1 Torr− 1; Hughes, 1982; Eno, 1994) octopus (2.2 μmol O2 kg− 1 min− 1 Torr− 1), squid (23.3 μmol O2 kg− 1 min− 1 Torr− 1; Eno, 1994) and Carcinus maenas (2.2 μmol O2 kg− 1 min− 1 Torr− 1; Scammell and Hughes, 1981) are likely to overestimate ‘physiological D’ values by a large factor (Piiper, 1982; Maina, 1988). As noted above, the empirical estimates of right gill Ldiff and total D incorporate a number of unknown factors that might have influenced diffusion across the branchial barrier. Both parameters varied quite widely among individuals, perhaps attributable to differential secretion of mucus or to lamellar recruitment analogous to fish gills (Randall and Daxboeck, 1984). Abalone gill lamellae exhibit marked spontaneous contractility and tactile sensitivity (unpublished observations) which could have contributed to heterogeneous perfusion of the lamellae. Within individuals there was a strong reciprocal relationship between Ldiff and D (Fig. 3). The explanation for this relationship is undoubtedly complex but does imply that decreases in total diffusing capacity were

305

Fig. 3. Relationship between the diffusion limitation index (Ldiff) and diffusive conductance (D) in the right gill of Haliotis iris. Replicate data from individual animals bear the same symbol.

not associated with corresponding decreases in blood flow, at least in the right gill. 4.4. Conclusions The ventilation–perfusion ratio, diffusion limitation index, and diffusing capacity all reveal the right gill of H. iris to be an efficient gas exchange organ whose operational characteristics resemble those of other competent water-breathers. About 40% of the oxygen was extracted from sea water and under some conditions it can reach 70% (Taylor and Ragg, 2005a). These observations imply that the paired bipectinate gill system is well-adapted to meet the oxygen demand of the resting abalone. The capacity of the system to accommodate an increase in oxygen demand now remains to be determined. For example, in H. kamtschatkana, M˙o2 was increased considerably during locomotion (Donovan and Carefoot, 1997).The Ldiff of about 0.5 and ventilation / perfusion conductance ratio near unity indicate that increase in either ventilation, or perfusion, or both could facilitate increases in M˙o2. In fact H. iris displays little flexibility in the rate of endogenous ciliary ventilation rate although branchial water currents induced by external flow undoubtedly could augment ventilation (Voltzow, 1983; Taylor and Ragg, 2005a). Recruitment of additional gill lamellae is also possible. Alternatively, adjustments to the relative perfusions of left and right gill could produce flexibility in gas exchange. This latter suggestion is the focus of on-going work by the authors. The findings presented therefore provide no support for the suggestion that inherent inefficiency of bipectinate gills as gas exchangers has driven evolutionary processes towards the abandonment of this design by higher prosobranchs. Acknowledgements The authors extend their gratitude for technical assistance to Gavin Robinson, Jan McKenzie, Victor Menzel and Franz Ditz. This project was generously supported by the Marsden Fund of New Zealand (contract UOC 804).

306

N.L.C. Ragg, H.H. Taylor / Comparative Biochemistry and Physiology, Part A 143 (2006) 299–306

References Ainslie, R., 1980. The quantitative role of haemocyanin in the respiration of abalone. J. Exp. Zool. 211, 87–99. Allen, V.J., Marsden, I.D., Ragg, N.L.C., 2001. The use of stimulants as an aid to wean fishery caught blackfoot abalone (Haliotis iris) to artificial food. J. Shellfish Res. 20 (2), 647–651. Baldwin, J., Wells, R.M.G., Low, M., Ryder, J.M., 1992. Tauropine and Dlactate as metabolic stress indicators during transport and storage of live paua, (New Zealand abalone) (Haliotis iris). J. Food Sci. 57 (2), 280–282. Barnes, R.D., 1986. Invertebrate Zoology, 5th edition. Saunders College Publishing. Behrens, J.W., Elias, J.P., Taylor, H.H., Weber, R.E., 2002. The archaeogastropod mollusc Haliotis iris: tissue and blood metabolites and allosteric regulation of haemocyanin function. J. Exp. Biol. 205, 253–263. Booth, C.E., Mangum, C.P., 1978. Oxygen uptake and transport in the lamellibranch mollusc Modiolus demissus. Physiol. Zool. 51, 17–32. Bourne, G.B., Redmond, J.R., 1977. Hemodynamics in the pink abalone, Haliotis corrugata (Mollusca, Gastropoda): I. Pressure relations and pressure gradients in intact animals. J. Exp. Zool. 200, 9–16. Crofts, D.R., 1929. Haliotis. In: Johnstone, J., Daniel, R.F. (Eds.), Liverpool Marine Biology Committee Memoirs on Typical Marine Plants and Animals, vol. 29. University Press of Liverpool. Dejours, P., 1981. Principles of Comparative Respiratory Physiology, 2nd edition. Elsevier/North-Holland Biomedical Press, Amsterdam. Depledge, M.H., Phillips, D.J.H., 1986. Circulation, respiration and fluid dynamics in the gastropod mollusc, Hemifusus tuba (Gmelin). J. Exp. Mar. Biol. Ecol. 95, 1–13. Donovan, D.A., Carefoot, T.H., 1997. Locomotion in the abalone Haliotis kamtschatkana: pedal morphology and cost of transport. J. Exp. Biol. 200, 1145–1153. Eno, N.C., 1994. The morphometrics of cephalopod gills. J. Mar. Biol. Assoc. U.K. 74, 687–706. Famme, P., 1981. Haemolymph circulation as a respiratory parameter in the mussel, Mytilus edulis L. Comp. Biochem. Physiol. A 69, 243–247. Fretter, V., Graham, A., 1994. British Prosobranch Molluscs: Their Functional Anatomy and Ecology. The Ray Society, London. Gäde, G., 1988. Energy metabolism during anoxia and recovery in shell adductor and foot muscle of the gastropod mollusc Haliotis lamellosa: formation of the novel anaerobic end product tauropine. Biol. Bull. 175, 122–131. Hughes, G.M., 1982. An introduction to the study of gills. In: Houlihan, D.F., Rankin, J.C., Shuttleworth, T.J. (Eds.), Gills. Society for Experimental Biology Seminar Series 16. Cambridge University Press, Cambridge. Johansen, K., 1982. Respiratory gas exchange of vertebrate gills. In: Houlihan, D.F., Rankin, J.C., Shuttleworth, T.J. (Eds.), Gills. Society for Experimental Biology Seminar Series 16. Cambridge University Press, Cambridge. Jones, D.R., Randall, D.J., 1978. The respiratory and circulatory systems during exercise. Fish Physiol. VII, 425–501. Kingston, R.S., 1968. Anatomical and oxygen electrode studies of respiratory surfaces and respiration in Acmaea. The Veliger 11, 73–78 (Supplement). Krajniak, K.G., Bourne, G.B., 1988. Aortic and venous blood pressures in the intact Pinto abalone, Haliotis kamtschatkana. Comp. Biochem. Physiol. A 89, 405–407. Lindberg, D.R., 1992. Evolution, distribution and systematics of Haliotidae. In: Shepherd, S.A., Tegner, M.J., Guzman del Proo, S.A. (Eds.), Abalone of the World: Biology, Fisheries and Culture. Blackwell, Oxford, pp. 3–17. Maina, J.N., 1988. The Gas Exchangers: Structure, Function and Evolution of the Respiratory Processes. Springer-Verlag, Berlin. Mangum, C.P., 1983. Oxygen transport in the blood. In: Mantel, L.H. (Ed.), The Biology of Crustacea Volume 5: Internal Anatomy and Physiological Regulation. Academic Press, New York. Mangum, C.P., Polites, G., 1980. Oxygen uptake and transport in the prosobranch mollusc Busycon canaliculatum: I. Gas exchange and the response to hypoxia. Biol. Bull. 158, 77–90.

McMahon, B.R., Wilkens, J.L., 1983. Ventilation, perfusion, and oxygen uptake. In: Mantel, L.H. (Ed.), The Biology of Crustacea Volume 5, Internal Anatomy and Physiological Regulation. Academic Press. Murdock, G.R., Vogel, S., 1978. Hydrodynamic induction of water flow through a keyhole limpet (Gastropoda, Fissurellidae). Comp. Biochem. Physiol. A 61, 227–231. Piiper, J., 1982. A model for evaluating diffusion limitation in gas-exchange organs of vertebrates. In: Taylor, C.R., Johansen, K., Bolis, L. (Eds.), A Companion to Animal Physiology. Cambridge University Press, pp. 49–64. Piiper, J., Scheid, P., 1984. Model analysis of gas transfer in fish gills. Fish Physiol. XA, 229–262. Purchon, R.D., 1977. The Biology of the Mollusca, second edition. Pergamon Press, Oxford. Ragg, N.L.C., 2003. Respiratory circulation in the abalone, Haliotis iris. Ph.D. dissertation, University of Canterbury, New Zealand. Randall, D.J., Daxboeck, C., 1984. Oxygen and carbon dioxide transfer across fish gills. Fish Physiol. XA, 263–314. Russell, C.W., Evans, B.K., 1989. Cardiovascular anatomy and physiology of the Black-lip abalone, Haliotis ruber. J. Exp. Zool. 252, 105–117. Scammell, C.A., Hughes, G.M., 1981. Comparative study of the functional anatomy of the gills and ventilatory currents in some British decapod crustaceans. Biol. Bull. 156, 35–47. Scheid, P., 1982. A model for comparing gas-exchange systems in vertebrates. In: Taylor, C.R., Johansen, K., Bolis, L. (Eds.), A Companion to Animal Physiology. Cambridge University Press, pp. 3–16. Taylor, H.H., Ragg, N.L.C., 2005a. The role of body surfaces and ventilation in gas exchange of the abalone, Haliotis iris. J. Comp. Physiol. B 175, 463–478. Taylor, H.H., Ragg, N.L.C., 2005b. Extracellular fluid volume, urine filtration rate and haemolymph mixing time in the abalone, Haliotis iris Martyn: a comparison of 51Cr-EDTA and 14C-inulin as extracellular markers. Mar. Freshw. Res. 56 (8), 1117–1126. Taylor, H.H., Taylor, E.W., 1992. Gills and lungs: the exchange of gases and ions. Microscopic Anatomy of Invertebrates 10, Decapod Crustacea. WileyLiss Inc, pp. 203–293. Vogel, S., 1994. Life in Moving Fluids: The Physical Biology of Flow, 2nd edition. Princeton University Press, Princeton, New Jersey. Voltzow, J., 1983. Flow though and around the abalone Haliotis kamtschatkana. Veliger 26, 18–21. Voltzow, J., 1994. Gastropoda: prosobranchia. In: Harrison, F.W., Kohn, A.J. (Eds.), Microscopic Anatomy of Invertebrates Volume 5: Mollusca I. John Wiley and Sons Inc, New York. Chapter 4. Voltzow, J., Morris, P.J., Linsley, R.M., 2004. Anatomy of and patterns of water currents through the mantle cavities of Pleurotomariid gastropods. J. Morphol. 262, 659–666. Wanichanon, C., Laimek, P., Linthong, V., Sretarugsa, P., Kruatrachue, M., Upatham, E.S., Poomtong, T., Sobhon, P., 2004. Histology of hypobranchial gland and gill of Haliotis asinina Linnaeus. J. Shellfish Res. 23, 1107–1112. Wells, M.J., 1992. The cephalopod heart: the evolution of a high-performance invertebrate pump. Experientia 48, 800–808. Wells, M.J., Smith, P.J.S., 1987. The performance of the octopus circulatory system: a triumph of engineering over design. Experientia 43, 487–499. Wells, R.M.G., Baldwin, J., Speed, S.R., Weber, R.E., 1998. Haemocyanin function in the New Zealand abalones Haliotis iris and Haliotis australis: relationships between oxygen-binding properties, muscle metabolism and habitat. Mar. Freshw. Res. 49, 143–149. Willson, L.L., Burnett, L.E., 2000. Whole animal and gill tissue oxygen uptake in the Eastern oyster, Crassostrea virginica: effects of hypoxia, hypercapnia, air exposure and infection with the protozoan parasite Perkinsus marinus. J. Exp. Mar. Biol. Ecol. 246, 223–240. Yonge, C.M., 1947. The pallial organs in the aspidobranch gastropoda and their evolution throughout the mollusca. Philos. Trans. R. Soc. London 232B, 443–519.