An Investigation of Haemocyanin Oxygen Affinity in the Semi-Terrestrial Crab Ocypode Saratan Forsk

Although the ghost crabs, family Ocypodidae, have been the subject of numerous ecological studies (Bliss, 1968; Wolcott, 1978; see Powers & Bliss, 1983 for review) few studies have been made of the respiratory physiology of this group (Storch & Welsch, 1975; Burnett, 1979). Some characteristics of haemolymph gas transport in land crabs have been described by Redmond (1968), Young (1972) and Burnett & Infantino (1984). A significant contribution to the study of air breathing in crabs was made by a series of investigations during the Alpha Helix cruise of 1979 (cf. Cameron, 1981a). These investigations were, however, restricted to the genera present on the Palau islands and did not include any member of the Ocypodidae nor did they investigate extensively the oxygen-binding properties of the haemocyanin.

Since the work of Truchot (1980), the existence of specific, organic modulators of haemocyanin oxygen affinity has become an area of increasing interest. Some comparative information is available about the distribution of these modulating effects throughout the Crustacea (Mangum, 1983a; Bridges, Morris & Grieshaber, 1984), but to date no terrestrial decapod has been considered.

The subtropical land crab, Ocypode saratan, although not fully terrestrial, is confined to a supralittoral habitat (von Linsenmair, 1967) and can be found up to 1 km inland. The object of the present study was to characterize some of the oxygentransporting properties of the haemolymph of this species and to investigate the specific modulation of haemocyanin O₂ affinity in a terrestrial decapod. The results of the investigation are discussed in terms of the possible physiological advantages to Ocypode saratan.

Adult Ocypode saratan were collected from local beaches by personnel of the University of Jeddah, Saudia Arabia and individually packed for transport in cooled, moist styrofoam cases. The animals were then air freighted to the Department of Zoology in Glasgow and installed in a warm aquarium room at 25 ± 2 °C. The crabs were maintained for 2 weeks in tanks containing autoclaved sand dampened with sea water. Fresh water and sea water were both provided and the animals were fed with fresh salad/fruit twice each week and occasionally with chopped meat.

Prior to sampling, intermoult male crabs of carapace width 35–49 mm (weight range 21–36 g) were maintained in separate polystyrene ice buckets, with moist sand, for a period of 1 day to ensure that L-lactate in the haemolymph was at resting levels. Venous haemolymph samples were then withdrawn, in less than 10 s, from the quiescent crabs via the arthrodial membrane at the base of the third walking leg. The samples were then pooled and frozen at — 70 °C prior to transport to Düsseldorf.

The concentrations of the key inorganic ions were determined in the undialysed haemolymph samples. The concentrations of Mg²⁺ and Ca²⁺ were determined spectrophotometrically (Merkotest 3338, Darmstadt, F.R.G. and Roche, No. 1028, Basel Switzerland, respectively) and that of Cl^- using a chloride titrator (CMT 10, Radiometer, Copenhagen, Denmark). The concentration of L-lactate in the haemolymph samples and Ringer solutions used was estimated using the method of Gutmann & Wahlefeld (1974) modified according to Graham, Mangum, Terwilliger & Terwilliger (1983).

Haemolymph protein and haemocyanin concentrations were calculated from the absorbance peaks of spectrophotometric scans between 250–400 nm using 1 : 100 dilutions of haemolymph in Ringer solution. The absorbance maxima at 280 nm and 335 nm were used to determine concentrations of total protein and haemocyanin respectively using the extinction coefficients of 14·3 ⁠, and 2·69 (Nickerson & van Holde, 1971).

Native Ocypode haemolymph, in aliquots of 1ml, was dialysed at 4 °C for 24 h against two changes (21) of Ocypode Ringer (pH = 8·2) with the following composition (in mmol I^-1) : NaCl, 386; KC1, 7; MgSO₄, 10; MgCl₂, 5; and NaHCO₃, 22. Dialysis was carried out twice on two separate haemolymph aliquots (A and B). The different haemolymph lactate concentrations were achieved in the following manner. Samples (100 μl) from the appropriate test haemolymph were centrifuged at 136 000 g (Airfuge ultracentrifuge, Beckman, California, U.S.A.) for 30min to pellet the haemocyanin. A portion of the plasma (5 μl) was then removed and replaced with 5 μl of neutralized (pH = 8·2) solutions of L-(+)-lactate (Boehringer, Mannheim, F.R.G.) of varying concentrations. The haemocyanin was then remixed with the plasma. The precise concentration of lactate in the haemolymph was subsequently determined by enzymatic assay. Control haemolymph solutions were made by using similar (5 μl) amounts of Ringer solution without lactate.

Oxygen equilibrium curves were constructed spectrophotometrically using 10 μl samples in a diffusion chamber (Sick & Gersonde, 1969) with the modifications of Lykkeboe, Johansen & Maloiy (1975) used by Bridges, Bicudo & Lykkeboe (1979). Gas mixtures were supplied to the chamber by serially connected Wösthoff gas mixing pumps (303 a/f Wösthoff, Bochum, F.R.G.). The pH of the haemolymph was varied by changing the partial pressure of CO₂ in the gas mixture ⁠. The pH of the haemolymph was measured in separate sub-samples (60 μl) tonometered in a BMS 2 (Radiometer) which was supplied with the same gas mixtures as the diffusion chamber. The pH of the equilibrated haemolymph was measured near the P50 using a microcapillary electrode (Type G299, Radiometer) thermostatted at the experimental temperature in BMS 2. These data were then used to calculate the magnitude of the Bohr shift.

Measurements on the effect of lactate were made at 30 °C. The P₅₀ values were estimated from regression analyses of saturation values between 25 and 75 % according to the Hill equation. The cooperativity (n₅₀) was also determined from these points.

The possible existence of a specific effect of CO₂ on the affinity of Ocypode saratan haemocyanin was investigated by comparing the fixed acid and CO2 Bohr effect for native haemolymph at 30 °C. A similar method to that of Morris, Taylor, Bridges & Grieshaber (1985) was used to make these determinations. This technique involved first centrifuging the haemolymph and then replacing a small amount of the supernatant plasma (5 μl) with 0·1 mol I^-1 HC1 or 0·1 mol I^-1 NaHCCh made up in Ringer solution. Control determinations were made using replacements of Ringer alone. Oxygen equilibrium curves were then constructed using two CO₂ concentrations, 0·4% and 3%, in order to compare the effects of changing pH with and without changes in CO₂.

Estimates of the buffer value of undialysed haemolymph and of dialysed haemolymph were made from CO₂ titration measurements. The was varied using gas mixing pumps (see above) and the pH measured. In each case the haemolymph bicarbonate concentration was calculated using the Henderson-Hasselbalch equation together with and α CO₂ estimated from the nomograms of Truchot (1976) for 30°C (⁠ and α CO₂ = 0·334 molT¹ Torr^-1).

Calculations made using data from the spectrophotometric scans of the haemolymph of Ocypode saratan revealed a protein concentration of 53·3 mg ml^-1 and a haemocyanin concentration that was virtually the same at 53·4 mg ml^-1. Using this information and assuming a molecular weight for haemocyanin of between 72 000 and 75 000 Da (Rochu & Fine, 1984) the oxygen-carrying capacity of the haemocyanin was calculated to be 0·72–0·74 mmol I^-1 (1·59–1·62 vol%).

The concentration of Cl^- in the native haemolymph was 425 ±15 mmol I^-1 and the concentration of Mg^2+- was 14·6 ± 0·8 mmol I^-1 compared to 9·4 ± 2·3 mmol I^-1 for Ca²⁺ ion concentration. The calculated concentration of HCO₃^- in the haemolymph after equilibration with 7·24TorrCO₂ was, at 30°C, 23·1 ± l·8mmoll^-1.

A large number of equilibrium curves resulted from this investigation and two typical sets of data for haemolymph at 20 and 35 °C are shown in Fig. 1. From these data it is apparent that the oxygen affinity of the haemolymph was temperature dependent; for example, the P₅₀ at 35 °C was 12·2 Torr (pH = 7·91) but decreased to a value of 8·3 Torr at 20°C (pH — 7·90). The same curves also clearly demonstrate the presence of a significant Bohr effect.

The CO₂ Bohr factor (φ = Δlog P₅₀/ΔpH) was calculated for native Ocypode haemolymph at all experimental temperatures using the data presented in Fig. 2. The value of φ was found to be essentially temperature-independent (Fig. 2) and a mean CO₂ Bohr factor of -0·670 ±0·02 could be calculated for the temperature range 20–35 °C.

The temperature sensitivity of the haemocyanin was analysed further by comparing the ΔH values (Table 1) at two different pH values using data calculated from the regression lines presented in Fig. 2. The similarity of AH at the two pH values (7·4 and 7·8) confirms the low temperature sensitivity of φ. At a constant pH 7·8, the change in the heat of oxygenation of the respiratory pigment accompanying a rise in temperature from 20 to 35 °C was found to be — 18·4kJ mol^-1. An examination of the change in ΔH accompanying increases of 5 °C within this temperature range showed, however, a distinct decrease in temperature sensitivity near to the normal environmental temperature. Increasing the temperature of the haemolymph from 25 to 30 °C produced a ΔH value of less than — 4 kJ mol^-1.

Using P₅₀ values from the O₂ equilibrium curves made at various temperatures an estimate was made for the temperature dependence of pH at a constant ⁠. This calculated in vitro value obtained at a of 7·4 ± 0·1 Torr (a CO₂ tension selected from the literature as being close to the in vivo tension) was extremely low (<0·001 pH units °C^-1). Interestingly, the calculated n₅₀ values exhibited no obvious dependence on temperature (Fig. 2) and changed minimally, if at all, with a change in haemolymph pH.

No detectable specific effect of carbon dioxide on O₂ binding by the haemocyanin of Ocypode saratan was found within the C O₂ concentration range 0·4–3·0 % (Fig. 3). The P₅₀ values calculated from equilibrium curves made at these two concentrations and at different fixed acid concentrations were judged to describe one line for the relationship between pH and log P₅₀ (the fixed acid Bohr shift). The calculated regression line for the combined data from the fixed acid determinations was; log P₅₀ = 6·31–0·67pH (r=0·97). A similar Bohr value was found for untreated haemolymph in which the Bohr shift was induced by CO₂ (see Figs 2, 3). The identity of the CO₂ and fixed acid Bohr shifts is clearly evident in Fig. 3, as is the absence of any effect of the small dilutions of the native plasma by the addition of acid or base.

The measured L-lactate concentration of native Ocypode saratan haemolymph was 1·12 ± 0·01 mmol l^-1, a value sufficiently low to permit the addition of neutralized L-lactate to give concentrations in the physiological range (Wood & Randall, 1981; Smatresk & Cameron, 1981; Taylor & Davies, 1982). The error of the lactate estimates was at all times less than 1 %. Increasing the lactate concentration of the undialysed haemolymph was found to increase the haemocyanin oxygen affinity (Fig. 4). Increasing the haemolymph lactate concentration to 9·51 mmol I^-1 resulted in a pH-dependent decrease in the log P₅₀ value. For example Δ log P₅₀ is 0·123 at pH 7·8 (Δ P50 — 3·7Torr) but increases to 0·195 at pH7·4 (Δ P₅₀— 10·5Torr) Increasing the concentration of haemolymph lactate not only increases affinity but also significantly (covariance analysis gives 0·01 <P< 0·05) reduces the size of the Bohr factor (Fig. 4). The n₅₀ values calculated from the same data showed no evidence of lactate-dependent cooperativity.

For the purposes of determining if other effectors of O₂ affinity were present in the haemolymph of Ocypode saratan, similar experiments were carried out on haemolymph dialysed against Ocypode Ringer. Lactate exponentially increased O₂ affinity in both dialysed and undialysed haemolymph. For comparison, plots of log P50 against log L-lactate concentration for dialysed and undialysed haemolymph have been presented in Fig. 5. The data were in all cases calculated using values from the regression lines for log P50 vs pH and include data shown in Fig. 4. The response of haemocyanin affinity to increasing lactate was dissimilar in dialysed and undialysed haemolymph. Firstly, the O2 affinity of the undialysed haemolymph never exceeded that of haemolymph that had been dialysed and secondly, at lactate concentrations lower than approximately 7 mmol I^-1, the affinity of undialysed haemolymph became progressively lower. This effect resulted in a large difference in O₂ affinity between dialysed and undialysed haemolymph at low lactate concentrations. Near to the measured lactate concentration for the original native haemolymph, this difference was calculated to be 8·8Torr and 3·5 Torr at pH7·4 and pH7·8, respectively. Similar results were obtained with the two separately dialysed aliquots (Fig. 5: A, B). The relationship between log P₅₀ and log L-lactate concentration was calculated for native haemolymph and dialysed haemolymph by regression analysis (Table 2). The calculated coefficients in Table 2 indicate that dialysed haemolymph exhibited a reduced sensitivity to L-lactate when compared to that of undialysed haemolymph. A maximum effect of lactate was achieved at much lower concentrations (approx. 4 mmol I^-1) in dialysed blood than in undialysed blood (approx. 7 mmol I^-1). At concentrations greater than 7 mmol I^-1 lactate, the effect was maximal in all preparations and the oxygen affinity was essentially the same at any given pH value.

Buffer values (β) calculated using pH and data from the oxygen equilibrium curves for undialysed blood are given in Table 3. It would appear that an increase in haemolymph lactate concentration results in a progressively decreasing buffer value. Furthermore, this relationship was not linear but exponential, like the relationship between log P₅₀ and lactate concentration. A regression equation calculated for these data by the method of least squares is given below :

Although no in vivo measurements were made during this study it would appear, from the literature, that the Ocypodidae might differ from other terrestrial decapods in at least one respect, the pH of the haemolymph. Data for Ocypode quadrata, which exhibits a similar mode of life to Ocypode saratan (von Linsenmair, 1967 ; Wolcott, 1978), indicate anuiw pH value in excess of pH 7·9 (Burnett, 1979). Investigation of other semi-terrestrial crustaceans (Burggren & McMahon, 1981; McMahon & Burggren, 1981; Smatresk & Cameron, 1981; Cameron, 1981b; Wood & Randall, 1981 ; Taylor & Davies, 1982) all report pH values lower than 7·8. These workers agree, however, that haemolymph levels normally lie within the range 6·12 Torr. The in vitro pH 7·89 measured at a of 7·4 Torr (30 °C) in Ocypode saratan haemolymph is in close agreement with the data of Burnett (1979) and may be important in confirming that the Ocypodidae maintain a high haemolymph pH. This may be facilitated in Ocypode saratan through a high haemolymph bicarbonate level which can be up to 50 % higher than those reported in other land crabs (McMahon & Burggren, 1981; Cameron,.1981b; Wood & Randall, 1981; Taylor & Davies, 1982).

The maintenance of a high blood pH is of significance when estimating oxygen affinity in vivo. At pH 7·8 and near environmental temperatures (30°C) the P50 is approximately 12 Torr. This is not an exceptionally high affinity in comparison with the P₅₀ values from other terrestrial crabs, e.g. 3·9Torr (pH 8·18) in Cardisoma guanhumi (Young, 1972) and 4·5 Torr (pH 7·53) in the same species (Redmond, 1962), 20 Torr (pH7·3) in Gecarcinus lateralis (Taylor & Davies, 1981), 13 Torr (pH 7·46) and 25 Torr (pH 7·46) in Cardisoma carnifex and Birgus latro, respectively (Burggren & McMahon, 1981). The mean CO₂ Bohr factor of — 0·67 in Ocypode saratan is lower than the — 0·76 reported for Ocypode quadrata (Burnett, 1979) and the — 0·95 found for Ocypode occidentalis (Burnett & Infantino, 1984).

It is evident from Fig. 1 and Table 1 that oxygen affinity is temperature dependent in Ocypode saratan. The Δ H values in Table 1 are, however, low in comparison with values from other crustaceans (Δ H = — 12 to —143 kJ mol^-1; Mangum, 1980). Young (1972) reports values, calculated at constant pH, of — 45 kJ mol^-1 for Car-disoma, —43 kJ mol^-1 for Gecarcinus and —46 kJ mol^-1 for Goniopsis within the temperature range 25–30 °C. Maximum values for Δ H (approx. — 26 kJ mol^-1) are observed in the present study at extreme temperatures. A similar temperaturedependent effect has been shown in Palaemon elegans (Morris et al. 1985), and low temperature sensitivity of oxygen binding has been reported in a number of crustaceans, including Callianasa califomiensis (Miller & van Holde, 1974), Pagurus bernhardus (Jokumsen & Weber, 1982), Euphasia superba (Bridges et al. 1983) and Palaemon elegans (Morris et al. 1985). Values given for AH in the present study are calculated for a constant pH. In vivo pH values will change with temperature: McMahon & Burggren (1981) found a Δ pH/ Δ T of — 0·023 and — 0·017 in Birgus latro and Cardisoma carnifex, respectively. In vitro measurements in Ocypode saratan, however, indicate a far lower temperature sensitivity for pH. The Bohr effect may remain temperature-independent (Fig. 2), but if affinity is decreased at higher temperature then the same pH change will produce a more pronounced change in Pso (cf. logP₅₀) than at lower temperatures.

Investigation of a specific CO₂ effect on oxygen binding by haemocyanin in Ocypode saratan did not reveal any changes in oxygen affinity, which is in agreement with other recent work (Burnett & Infantino, 1984; A. C. Taylor, S. Morris & C. R. Bridges, in preparation). Differences between fixed acid and CO₂ Bohr effects have seldom been reported (Truchot, 1973; Weber & Hagerman, 1981; Morris et al. 1985). Young (1972) has suggested, however, that his findings and those of Redmond (1955) possibly indicate the presence of a specific CO₂ effect in land crabs. No effect was detected in Ocypode (see also Burnett & Infantino, 1984), but this group (Ocypodidae) was not considered by Young and may well be phylogenetically different.

The recent finding (Truchot, 1980) that L-lactate increases the oxygen affinity of some haemocyanins has prompted more detailed investigations of this effect (Mangum, 1983a; Bridges et al. 1984). The calculated coefficients for the lactate effect (Δ logP₅₀/ Δ log [lactate^-]) of -0·240 and -0·162 in undialysed haemolymph at pH 7·4 and pH 7·8, respectively (Table 2), represent a marked effect of lactate when compared to the known range of values of 0·00 to — 0·630 (Bridges et al. 1984). Some care should be taken when interpreting these values as the coefficient can be influenced by several parameters (Bridges et al. 1984). The coefficient does, however, demonstrate the pH sensitivity of the effect seen in some (see Bridges et al. 1984) but not all species (Truchot, 1980; Graham et al. 1983). The apparent reduction in the magnitude of the lactate effect in dialysed blood at low lactate concentration was unexpected. Although both sets of data (Fig. 5) show typical saturation kinetics the effect is near maximum at a much lower lactate concentration in dialysed haemolymph (4mmoll^-1) than in undialysed haemolymph (7 mmol I^-1). The cause of this difference is unclear.

The presence of an unidentified modulator of oxygen affinity has been shown in a number of crustaceans (Mangum, 1983a,b; Bridges et al. 1984), but these all increased the oxygen affinity. Lactate binding to haemocyanin does not show a 1 :1 stoichiometry and lactate is thought to bind either between subunits or to a particular subunit (Johnson, Bonaventura & Bonaventura, 1984). It is possible that competitive inhibition could occur at either of these sites.

The exponential decrease in the buffer value with increasing lactate concentration (Table 3) also suggests an allosteric change in the structure of the haemocyanin such that the number of available proton buffering sites are decreased. This is corroborated by the decrease in the CO₂ Bohr effect with increasing lactate concentration. Much speculation surrounds the identity and mechanism of haemocyanin oxygen affinity modulators (Bridges et al. 1984; Johnson et al. 1984) : the results for Ocypode saratan raise new questions which await further investigation at the molecular level.

Exercise or environmental hypoxia may exert substantial demands on the oxygen transport system of crustaceans and in particular air-breathing species like Ocypode saratan. Cardiac output together with ventilation may control the efficiency of gas exchange and therefore modulate Pao2 (Wood & Randall, 1981), but venous oxygen tensions will depend upon the position of the equilibrium curve and on oxygen demand and supply.

A right shift of the curve, due to respiratory or metabolic acidosis, will aid unloading at the tissues to the detriment of arterial saturation, if values are initially low due to gas exchange limitations or if the haemocyanin has an intrinsically low oxygen affinity. A left shift of the curve, due to respiratory alkalosis or to the presence of lactate, will enhance loading at the gill, if arterial saturation was previously impaired, and also defend the venous reserve, depending upon oxygen demand (Booth, McMahon & Pinder, 1982; Graham et al. 1983).

Inequalities in the efflux rates of protons and lactate from the tissues and their removal from the haemolymph after exercise have been reported in two species of land crab, Cardisoma carnifex (Wood & Randall, 1981) and Gecarcinus lateralis (Smatresk, Preslar & Cameron, 1979). Lactate was found to have a longer residence time in the haemolymph after exercise and may therefore play a significant role during recovery. This may represent an ideal adaptation for Ocypode saratan, which exhibits short periods of rapid locomotion and more sustained burrowing activity (von Linsen-mair, 1967) the lactate effect would prevent the immediate depletion of haemolymph oxygen stores and could help maintain oxygen delivery under conditions of high oxygen demand.

The magnitude of both the Bohr and lactate effects together with the buffer value of the haemolymph, which is lactate-dependent in Ocypode saratan (Table 3), will all play a decisive role in determining the response of the oxygen transport system to both internal and external demands.

We are indebted to the personnel of the University of Jeddah, Saudi Arabia for supplying the animals for these experiments and we should like to thank Dr R. J. A. Atkinson and Dr A. C. Taylor for supervising the maintenance of the animals prior to sampling. Thanks are due also to Annette Lundkowski for technical assistance. Financial support was provided by the Heinrich-Hertz Stiftung, Nordrhein-Westf alen to SM and by the Deutsche Forschungsgemeinschaft (Gr 456/10-1) to CRB.