The Regulation of Haemocyanin Oxygen Affinity During Emersion of the Crayfish Austropotamobius Pallipes: I. An In Vitro Investigation of the Interactive Effects of Calcium and L-Lactate On Oxygen Affinity

The effects of aerial exposure on respiration in decapod crustaceans have been investigated for a number of species and many of the respiratory responses to emersion are now well understood (e.g. Wheatly & Taylor, 1979; Taylor, Butler & Sherlock, 1973; McMahon & Wilkes, 1983).

The crayfish Austropotamobius pallipes has been observed to leave the water when the oxygen tension falls below 40 Torr and to breathe air (Taylor & Wheatly, 1980, 1981). Complete emergence of crustaceans into air often causes, initially, a haemolymph acidosis (Truchot, 1975a; Taylor & Wheatly, 1981; deFur & McMahon, 1984b), which is normally accompanied by a decrease in post-branchial haemolymph oxygenation (Taylor & Wheatly, 1980, 1981; deFur & McMahon, 1984a; deFur, McMahon & Booth, 1983). An increased anaerobiosis leads to an accumulation of both lactate and H⁺ in the haemolymph. Taylor & Wheatly (1981) have shown that during extended aerial exposure the initial acidosis that occurs in the haemolymph of A. pallipes can be almost completely compensated, over 24 h, by elevated levels of haemolymph HCO₃⁻. These authors also demonstrated that the accumulated lactate in the haemolymph of A. pallipes was either substantially excreted, sequestered in the tissues, or metabolized.

The role of pH in modulating the oxygen affinity of crustacean haemocyanins is now appreciated, as is the potentiating effect of Ca²⁺ (Larimer & Riggs, 1964; Truchot, 1975b; Weiland & Mangum, 1975; Arisaka & van Holde, 1979). The other major haemolymph variable during aerial exposure, L-lactate, is now known to increase haemocyanin oxygen affinity (Truchot, 1980). Although both of these factors are known to fluctuate in the haemolymph of A. pallipes, the effect of these ions on the oxygen affinity of the haemocyanin has until now received little attention. The study reported here considers the effects of Ca²⁺ and L-lactate separately and together on the in vitro oxygen affinity of haemocyanin from A. pallipes.

Specimens of the common British crayfish, Austropotamobius pallipes (Lereboullet) were collected and maintained as described by Taylor & Wheatly (1980, 1981). Haemolymph was withdrawn from intermoult animals via the pericardial space. The haemolymph was then frozen at −70°C and transported under solid CO₂ to Düsseldorf where the investigation was carried out. The frozen material was allowed to thaw on ice and was then pooled. Clotted material was broken up by pressing the haemolymph through a fine nylon mesh and subsequent centrifugation at 10000g for 10 min. The resulting haemolymph was divided into 500 μl aliquots which were frozen until required. The possibility that freezing may affect the oxygenation properties of A. pallipes haemocyanin has been discussed (Morris, Bridges & Grieshaber, 1986) and, although some changes in cooperativity may occur, n₅₀ is not markedly reduced and changes in P₅₀ were considered unlikely.

The concentrations of key ions in the haemolymph were determined using the methods described by Bridges, Morris & Grieshaber (1984), Morris, Bridges & Grieshaber (1985a) and Morris, Taylor, Bridges & Grieshaber (1985b), in which Ca²⁺ and Mg²⁺ were measured spectrophotometrically (test kits: 1028, Roche, Basel, Switzerland; Merkotest 3338, Darmstadt, FRG) and Cl⁻ was measured with a chloride titrator (CMT 10, Radiometer, Copenhagen, Denmark). The concentration of L-lactate in the haemolymph was measured according to the method of Gutmann & Wahlefeld (1974), modified by the addition of EDTA to the assay (Engel & Jones, 1978; Graham, Mangum, Terwilliger & Terwilliger, 1983). Each determination was made at least twice.

The concentration of haemocyanin was determined by spectrophotometric scanning of 10 μ1 of haemolymph in 1ml Ringer solution between 200 and 450 nm (Uvikon 810, Kontron, Munich, FRG). The haemocyanin concentration was then calculated from the absorbance maximum near 335 nm, assuming an extinction coefficient of 2·69(Nickerson & van Holde, 1971).

Aliquots of A. pallipes haemolymph (0·5 ml) were dialysed against a series of Ringer solutions. In each case the dialysis was carried out at 4°C for 24 h with the Ringer solution (21) being replaced after approximately 12 h. The stock Ringer was formulated on the basis of measured and literature values and had the following composition in mmol l⁻¹: NaCl, 181; KC1, 4·7; CaCl₂, 17; MgCl₂, TO and NaHCO₃, 3·0 (pH = 8·0). Ringer solutions were also prepared with CaCl₂ concentrations of 49 mmol l⁻¹ and 9 mmol l⁻¹ to encompass the in vivo variation in [Ca²⁺] that occurs during aerial exposure of the crayfish (E. W. Taylor, unpublished results). The concentration of NaCl was changed correspondingly to ensure that the concentration of Cl⁻ remained constant. In addition to this each of the three Ringer solutions was prepared with either low lactate (1·0 mmol l⁻¹ L-( + )-lactate, Sigma Chemie GmbH, Taufkirchen, F.R.G.) or high lactate (approx. 8·0 mmol l⁻¹) to give a total of six different Ringer solutions against which native A. pallipes haemolymph was dialysed. The exact concentrations of Ca²⁺ and L-lactate in each dialysed haemolymph preparation were measured subsequent to the construction of oxygen equilibrium curves.

The possible existence of a specific effect of CO₂ that might increase the oxygen affinity of A. pallipes haemocyanin was investigated using the methods of Morris et al. (1985b) and of Morris & Bridges (1985). Dialysed haemolymph samples (100 μl) were centrifuged in an ‘air-fuge’ (Beckman, California, USA) for 45 min at 140000g in a precooled rotor. After centrifugation, 40 μ1 of the supernatant was removed and a measured amount of this replaced by either 0·1 moll⁻¹ HC1 or 0·1 moll⁻¹ NaOH (5–30 μ1) in A. pallipes Ringer. The resulting mixture was carefully remixed with the haemocyanin pellet. Control measurements were made using solutions in which a 30 μ1 replacement was made with Ringer alone. Oxygen equilibrium curves were then constructed at constant tensions of 0·7 and 10·5 Torr for haemolymph in which the pH had been fixed by the addition of acid or base. By comparing this ‘fixed acid’ Bohr shift to the Bohr shift induced by CO₂, any specific effect of CO₂ on oxygen affinity could then be distinguished from the CO₂ Bohr shift.

The curves were constructed using a spectrophotometric method on 8 /il samples in a diffusion chamber (Sick & Gersonde, 1969) as described by Bridges, Bicudo & Lykkeboe (1979). Briefly, oxygen tension was varied using gas mixing pumps (Wösthoff, Bochum, FRG), and the pH of the haemolymph was varied by adjusting the CO₂ content of the gas mix between 0·1 and 2·0%. The same gas mixtures were also supplied to the tonometers of a BMS II (Radiometer) to enable the haemolymph pH to be measured near the P₅₀ using the microelectrode (G299, Radiometer) of the BMS II. All determinations were made at 15°C. The values of P50 and cooperativity (n₅₀) were calculated by regression analysis of the saturation values between 25% and 75% using the Hill equation. Where appropriate, values throughout this paper are given as means ±1 S.D. unless otherwise stated.

The haemocyanin concentration of the native A. pallipes haemocyanin (Hey) used in the investigation was 53·5 mg Hey ml⁻¹ and the dialysed haemolymph preparations had a mean concentration of 50·1 ±6·8 mg Hey ml⁻¹. Consequently, the concentration of haemocyanin was assumed to remain effectively constant throughout the experiment.

In the pooled haemolymph, L-lactate averaged 1·01 ± 0·02 mmol l⁻¹. The concentrations of L-lactate measured in the dialysed haemolymph preparations are given in Figs 1, 2, 3. The measured ion values of untreated A. pallipes haemolymph were 9·7 ± 2·1 mmol l⁻¹ for Ca²⁺ and 198 ± 1·0 mmol l⁻¹ for Cl⁻, the concentration of Mg²⁺ was low at 1·9 ± 0·2 mmol l⁻¹. In the dialysed haemolymph preparations, these ions, with the exception of Ca²⁺, were maintained constant. The concentration of Ca²⁺ is given, in each case, together with the appropriate figure.

The result of dialysing the haemolymph was quite clearly to reduce the apparent affinity of A. pallipes haemocyanin for oxygen (Fig. 1A,B). The data show a clear specific effect of L-lactate (Fig. 1A), which increased the oxygen affinity of dialysed haemolymph. In the example given (Fig. 1A), increasing L-lactate from 0·9 mmol l⁻¹ to 7·2 mmol l⁻¹ reduced the P₅₀ from 6·4 Torr to 4 Torr (pH = 7·9).

Similar results were obtained when pH and L-lactate concentration were maintained at whole haemolymph values and the concentration of calcium ions was increased (Fig. 1B). Increasing the concentration of Ca²⁺ from the whole haemolymph value of 9·7 mmol l⁻¹ to 17 mmol l^{− 1} produced an increase in oxygen affinity of 2·8 Torr. Increasing the concentration of Ca²⁺ further to 45 mmol l⁻¹ elicited a further increase.

Interactive effects of these two modulators were investigated by increasing the concentration of both Ca²⁺ and L-lactate together (Fig. 2), which increased oxygen affinity from a P₅₀ value of 6·4 Torr to a P₅₀ value of 2·9 Torr. A simple additive effect of the two ions (i.e. adding the lactate effect in Fig. 1A to the Ca²⁺ effect in Fig. 1B) would be expected to increase oxygen affinity beyond this to a P₅₀ value near 1·5 Torr (Fig. 2). There was, therefore, clear evidence of agonism in the effects of L-lactate and calcium ions.

The interdependent nature of the potentiating effects of Ca²⁺ and L-lactate on the oxygen affinity of A. pallipes haemocyanin was investigated further and a larger range of experimental conditions were considered (Fig. 3). Examination of these data revealed that: (a) the increase in haemocyanin oxygen affinity brought about by increasing L-lactate concentrations became progressively reduced as the concentration of Ca²⁺ became more elevated; and (b) the effect of Ca²⁺ on oxygen affinity was similarly reduced by increasing the concentration of L-lactate. It was not possible to conclude that either L-lactate or Ca²⁺, alone or together, were able significantly to affect the cooperativity (n₅g) or the Bohr value (−0·455 ± 0·033). The interdependence of these modulators can be more accurately expressed by the equations in Table 1. The specific effect of L-lactate has been shown, when present, to be correlated as log [lactate⁻] with logP₅ (see Truchot, 1980; Bridges et al. 1984) and identical methods have been employed here. Although Truchot (1975b) correlated with logP₅₀, a more significant relationship was found in this study when log[Ca²⁺] was used (see also Mason, Mangum & Godette, 1983). All equations in Table 1 were calculated by least-squares regression analysis of values calculated from the regression equations of the data in Fig. 3. The progressive decrease in the magnitude of the lactate and calcium effect coefficients (Table 1) clearly demonstrates the decreased efficacy of lactate when [Ca²⁺] was increased, and also the reduced effect of Ca²⁺ when [L-lactate] was increased. Covariance analysis of the P₅₀vs pH data obtained in the presence of high and low L-lactate concentrations and at three different concentrations of calcium ions further demonstrated this interdependence. Increasing the calcium concentration from 9 mmol l⁻¹ to 17 and then to 47 mmol l⁻¹ could be correlated with a decrease in the variance ratio (F = 228, 34 and 13·6, respectively) obtained from comparisons of data from the high and low [L-lactate] preparations. The third F value (highest Ca²⁺ concentration) was significant at the 1% probability level, whereas those for data obtained at the middle and lower concentrations showed a more significant lactate effect (probabilities of 0·5 and 0·1%).

The strong potentiating effects of Ca²⁺ on the oxygen affinity of A. pallipes haemocyanin reported here concur with the early observations of Hogben (1926) working on Homarus and of Stedman & Stedman (1926) working on Cancer haemocyanin. Investigations of the binding of Ca²⁺ to haemocyanin (e.g. Morimoto & Kegeles, 1971; Brouwer, Bonaventura & Bonaventura, 1978; Arisaka & van Holde, 1979) have usually employed conditions that are outside the physiological range considered here for A. pallipes. Nevertheless, these more mechanistic studies demonstrate several features of Ca²⁺-haemocyanin interaction. For example, Kuiper et al. (1979) demonstrated that Ca²⁺-binding by the haemocyanin caused a pH-dependent H⁺ liberation and that the binding of Ca²⁺ was dependent on oxygenation state. Although a dependency of cooperativity (n₅₀) on [Ca²⁺] has been reported (Miller & van Holde, 1981; Larimer & Riggs, 1964; Chantler, Harris & Bannister, 1973), the present study, like that of Mason et al. (1983), found no significant increase in n₅₀ when [Ca²⁺] was increased. These authors (Mason et al. 1983) also compared the relatively high value for ΔlogP₅₀/Δlog[Ca²⁺] of −0·82 in Callinectes sapidus with the lower value of −0·28 in Carcinus maenas (Truchot, 1975b). Using their criteria, the effect of Ca²⁺ on the haemocyanin O₂ affinity in A. pallipes must also be considered to be large when [L-lactate] is at resting values (ΔlogP₅₀/Δlog[Ca²⁺] = −0·735).

Data demonstrating the binding of L-lactate to haemocyanin are less plentiful as the discovery of this effect is relatively recent (Truchot, 1980) and is at present limited to the work of Mangum (1983a) and of Johnson, Bonaventura & Bonaventura (1984). The effect of L-lactate on the oxygen affinity of haemocyanin from A. pallipes is, however, similar to previously quantified lactate effects (Truchot, 1980; Bridges et al. 1984; Morris & Bridges, 1985; Morris et al. 1985a). When compared with the estimates of ΔlogP₅₀/Δlogflactate] tabulated by Bridges et al. (1984), which ranged from 0 to −0·560 (pH 7·8), the value of − 0·194 (pH 7·9) for A. pallipes haemocyanin can be seen to represent a marked but unexceptional effect. There is no direct evidence that lactate binding is related to the number of H⁺-binding sites on the haemocyanin molecule as is apparently the case with the Ca²⁺ effect (Larimer & Riggs, 1964; Kuiper et al. 1979) and the postulated alternatives for the mechanism of the lactate effect (Johnson et al. 1984) would seem to indicate that this is not the case.

Comparison of ΔlogP₅₀/Δlog[Ca²⁺] values with those for ΔlogP₅₀/Δlog[lactate] indicates that the molecular relationship between haemocyanin and Ca²⁺ is different to that between haemocyanin and L-lactate. In A. pallipes Ca²⁺ would appear to be more effective on a molar basis than lactate in enhancing haemocyanin oxygen affinity.

It was concluded that the effect of either Ca²⁺ or lactate on the O₂ affinity of dialysed A. pallipes haemocyanin was progressively reduced by the presence of ions of the other species. The possibility of Ca²⁺ and lactate interacting as free ions was empirically investigated by Ghosh & Nair (1970), who found that the reaction Ca²⁺ + Lactate⁻ ⇌ CaLactate⁺ occurred in aqueous solution but that CaLactate₂ was not formed. A similar complexing of lactate with Ca²⁺ was found in turtle blood by Jackson & Heisler (1982). In the anoxic turtle blood, the concentration of Ca²⁺ rose to 68 mmol l⁻¹ and that of L-lactate to 145 mmol l⁻¹ which should not be considered comparable to the 8–10 mmol l⁻¹ that occurs in aerially exposed A. pallipes (Taylor & Wheatly, 1981). The formation of CaLactate⁺ in crustacean haemolymph was shown to be unlikely by the work of Graham et al. (1983). These authors showed that the haemocyanin of Cancer magister exhibits lactate sensitivity but were unable to demonstrate a change in free Ca²⁺ concentration when lactate concentrations were increased. Interactive effects of Ca²⁺ and lactate may not be a general feature of crustacean haemocyanin.

The effect of these two ions on the oxygen affinity of A. pallipes haemocyanin can be summarized graphically. By substituting a given value for logP₅₀ at pH7·9 into the equations (Table 1) it was possible to solve the relationship for either [Ca²⁺] or [lactate⁻]. The resulting points were used to construct Fig. 4. The interdependence of the effects of Ca²⁺ and L-lactate is manifest in the curved isopleths describing given P₅₀ values for A. pallipes haemocyanin. Interestingly, Mangum (1983a) comments on the similar effect of Ca²⁺ on Callinectes sapidus haemocyanin O₂ affinity, concluding that a specific effect of lactate added to the large effect of Ca²⁺ would increase affinity so much as to render it non-adaptive. It would appear from the data now available from A. pallipes that this additive effect is reduced due to a suppression of one by the other.

During emersion the arterial haemolymph in A. pallipes was reported to rise from 3·0 Torr in submerged normoxic animals to 10·6Torr after 12h in air (Taylor & Wheatly, 1981). No role could be attributed to CO₂ in the haemolymph of A. pallipes. Literature reports of a specific effect of CO₂ on oxygen affinity are limited to Truchot (1973), Weber & Hagermann (1981) and Morris et al. (1985b). Other studies such as those of Morris & Bridges (1985) and Burnett & Infantino (1984) were also unable to demonstrate specific CO₂ effects and the absence of such effects would now seem to be the normal condition in crustaceans.

This study has demonstrated, however, the presence of at least one unidentified factor that increases the oxygen affinity of the haemocyanin. Although the low specificity of this factor has been recently demonstrated (Morris et al. 1985a) little is known about its role in A. pallipes. In a recent study (Morris et al. 1986) the effects of factors other than L-lactate and Ca²⁺ have been considered and may offer an explanation of the absence of a lactate effect in some species (see Mangum, 1983b).

The present observations suggest that the reduction in the oxygen affinity of A. pallipes haemolymph during aerial exposure, which might be expected as a result of a Bohr shift, may not occur. Instead it is possible that the combined interactions of Ca²⁺, L-lactate and HCO₃⁻ in the haemolymph compensate for the effects of acidosis on the oxygen affinity of the haemocyanin and may even increase the affinity above that found in resting crayfish under normoxia.

We should like to thank Professor M. K. Grieshaber for making available the facilities of the Institut für Zoologie IV, Universität Düsseldorf without which this study would not have taken place. We are especially grateful to Dr C. R. Bridges for invaluable discussions throughout the period of the study and during the preparation of the manuscript. Thanks are due also to Ms A. Lundkowski for technical assistance. Financial support was provided by The Royal Society, London (SM, European Exchange Fellowship) and by the Science and Engineering Research Council (RT-J, Postgraduate Studentship).