Modulation of Haemocyanin Oxygen-Affinity by L-Lactate and Urate in the Prawn Penaeus Japonicus

During the last 15 years, many studies have dealt with the functions of crustacean haemocyanins and with the regulation of oxygen transport in the haemolymph under different kinds of stresses (e.g. osmotic, hypoxic, exercise) (reviewed in Mangum, 1983a; McMahon, 1985). Most of this work has concentrated on crabs, crayfishes and lobsters, with little attention paid to natantian decapods (e.g. Weber & Hagerman, 1981; Taylor et al. 1985; Morris et al. 1985,1988c). Moreover, there are no physiologically relevant data available for penaeid prawns, a group considered as the most primitive among decapods (Bowman & Abele, 1982).

Besides the well-known effects of inorganic ions, two organic cofactors reportedly increase haemocyanin oxygen-affinity in decapod crustaceans: L-lactate (Truchot, 1980) and urate (Morris et al. 1985). Numerous studies have shown that the L-lactate effect is present in many decapod crustaceans (Mangum, 19836; Bridges & Morris, 1986). Since the recent finding that urate enhances the oxygen affinity of Austropotamobius pallipes haemocyanin (Morris et al. 1985), only a few other species have been investigated (Morris & Bridges, 1986; Lallier, 1988). To be considered as true modulators, these two molecules must be shown to appear in haemolymph under stress conditions and to modify in vivo haemocyanin oxygen affinity. L-Lactate, the main end-product of anaerobiosis in Crustacea (Gäde, 1983), fulfils these requirements during muscular exercise (Booth et al. 1982) but its role during environmental hypoxia is not clear (Bouchet & Truchot, 1985; Lallier et al. 1987). Urate is a byproduct of purine catabolism, which is poorly documented in Crustacea (Claybrook, 1983).

Penaeus japonicus, a euryhaline species originating from southeast Asia, may encounter hypoxic water in the brackish, shallow lagoons where the adults live. This species is now cultured in brackish water ponds on the Atlantic coast of France. In this new environment, it may also face periodic hypoxic conditions of moderate amplitude, down to 9 kPa at 20°C at night (F. Lallier & J. P. Truchot, personal observations), or even more severe hypoxia, and possibly anoxic waters, in cases of eutrophication. Respiratory adaptations of the prawns to such decreases in environmental oxygen availability may involve an enhanced oxygen transport in the blood which could be mediated through an increased haemocyanin oxygen-affinity.

To examine how P. japonicus deals with such conditions, we (i) investigated the in vitro effects of L-lactate and urate on the oxygen-binding properties of P. japonicus haemocyanin and (ii) measured the variations of L-lactate and urate concentrations, together with the changes of haemocyanin oxygen-affinity in vivo, after exposure of the prawns to environmental hypoxia.

Animals were obtained from a culture farm located near the Gironde estuary (southwest France) at the end of summer (mean mass 12–15 g) and kept in the laboratory in a large tank with continuously renewed natural sea water maintained at 20–22°C, salinity 27 ‰, for several weeks. They were fed on artificial pellets (Aqualim), specifically made for prawns and identical to those used during their growth in culture.

The effects of L-lactate and urate on haemocyanin oxygen-affinity were studied on dialysed blood samples. A pool of haemolymph was sampled from the pericardial sinus of several individuals, using plastic syringes and G23 needles inserted between the thorax and the first abdominal segment. Since Penaeus japonicus haemolymph clots immediately, samples were treated by sonication and the liquid phase was collected after centrifugation before further analysis. The pooled sample (approx. 20ml) was then dialysed at 4°C, using an ultrafiltration method with immersible units (Millipore CX-10, cutoff point 10 000 Da). One unit was immersed in the haemolymph and connected to a peristaltic pump allowing withdrawal of water and low molecular weight solutes, which were automatically replaced with an equivalent volume of Ringer’s solution (in mmol 1⁻¹: NaCl, 430; CaCl₂,15; KC1,15; MgCl₂, 5; MgSO₄, 5; NaHCO₃, 5; pH7·6 at 25°C). After 12h, when about 50 ml of Ringer had been filtered, the process was repeated with a new filter unit. Haemolymph was continuously stirred and its equilibration with Ringer was periodically checked by comparison of the ultrafiltrate with Ringer for Na⁺, K⁺ and Ca²⁺ concentrations and osmotic pressure. The dialysed pool (approx. 20 ml) was separated into identical 1 ml samples and frozen at −20°C. After gentle thawing at 4 °C, each sample was prepared by adding various volumes of either a neutral 0·5 mol 1⁻¹ sodium L-lactate solution in Ringer or a saturated uric acid solution (approx. 2–3 mmol l⁻¹) in Ringer, diluted with Ringer to obtain the same final volume (1.2 ml). Using this procedure, the composition of all samples was the same, except for sodium L-lactate or urate concentration. L-Lactate and urate concentrations were measured for each sample after completion of the preparation procedure. The possible effects of freezing on oxygen-binding properties of haemocyanin have been assessed and no significant changes in affinity, cooperativity or the Bohr effect were found (Lallier, 1988).

For each sample, dissociation curves were constructed using a diffusion chamber (Sick & Gersonde, 1969) and a step-by-step procedure (Lykkeboe et al. 1975; Bridges et al. 1979). Gas mixtures were obtained using Wösthoff pumps, and pH changes were induced by varying the CO₂ tension in the gas mixtures. pH was measured with a capillary pH electrode (G299, Radiometer) on a separate sub-sample equilibrated with the same gas mixture in the tonometers of a blood gas analyser (BMS2, Radiometer). Parabolic curves were fitted to data pairs (pH, logP₅₀) using a quadratic polynomial least-squares regression, providing estimations of P₅₀ and Bohr factor at any pH value in the range studied (7.1-7.7).

Six groups of 7–10 prawns were acclimated for 24 h in normoxic sea water in a 20-1 experimental tank with a sandy bottom and supplied with sea water (31 min⁻¹) from a 50-1 thermostatted tank in which sea water was renewed continuously (0·31 min⁻¹). This acclimation period was followed by either normoxic (control) or hypoxic exposure (⁠ and 4·4kPa) for 3 or 48 h. Water oxygen partial pressure, was controlled by bubbling mixed air and nitrogen from mass lowmeters (FC260, Tylan), adjusted to obtain the desired value. was Wecked periodically by measurements with a thermostatted electrode. Postbranchial blood samples (0·3–0·7 ml) were withdrawn from each individual as rapidly as possible after removing the animal from the water, and treated as described above to remove the clot. Since it was impossible to get enough haemolymph for individual measurements, a representative pool of whole haemolymph was constituted with equal volumes from each sample, and used for further determinations of L-lactate and urate concentrations and production of equilibration curves.

L-Lactate concentrations were assessed with an enzymatic method (Boehringer, no. 139084) on samples deproteinized in 8% perchloric acid. This method reportedly provides stable readings for copper-containing samples (Gäde, 1984). Urate concentrations were determined using a spectrophotometric method (Sigma Diagnostic no. 685) at 520nm. At this wavelength, haemocyanin absorbance interfered and test absorption measurements (At) had to be corrected correspondingly. For this purpose, a blank measurement (Ab) was made on a sample containing the same volume of haemolymph as the test sample diluted in a phosphate buffer (pH 7·8, 250 mmol 1⁻¹) identical to the Sigma kit buffer. A standard sample measurement (As) served as the reference [Sigma Standard no. 685·1 diluted 1/1 (v/v) in distilled water, [Ustd]=0·149mmoll⁻¹]. Urate concentration was calculated using the following formula: [U] = [(At-Ab)/As] × [Ustd]. This method gave linear results over a wide range of urate concentrations (0–0·3 mmol 1⁻¹), with an average coefficient of variation of 2–3 %, amounting to a maximum 12·5 % at the lowest concentrations.

L-Lactate increased the oxygen affinity of Penaeus japonicus haemocyanin (covariance analysis, P<0·001) (Fig. 1). At pH7·6, which may be near the physiological value at the experimental temperature used, the drop in P₅₀ amounts to 0·9 kPa when L-lactate concentration was increased from 0·05 to 4·72 mmol 1⁻¹. Penaeus japonicus haemocyanin showed a large Bohr shift, reaching values of Φ=ΔlogP₅₀/ΔpH= −1·5 at pH7·.6. The lactate effect seemed to become quickly saturated as there was no further increase of affinity at L-lactate concentrations above about 5 mmol 1⁻¹. The cooperativity of oxygen binding to haemocyanin was not significantly affected by either L-lactate or pH (Fig. 1), with n₅₀ values ranging from 3·5 to 4·5 and averaging 4·1±0. ·2 (S.E., N=16). The lactate effect, measured as the ratio ΔlogP₅₀/Δlogflactate], was –0·067 at pH 7·2 compared with −0·077 at pH7·.6. However, this difference was not statistically significant (P>0·5), supporting the idea that the lactate effect may be independent of pH for P. japonica haemocyanin.

Urate also enhanced haemocyanin oxygen-affinity (covariance analysis, P<0. ·001) (Fig. 2). From a qualitative point of view, the effect of urate was similar to that of L-lactate, although more moderate. For example, at pH7. ·6, P₅₀ decreased from 2·9 to 2·3 kPa when urate concentration was increased from 0. ·001 to 0. ·96 mmol 1⁻¹. Raising urate concentration did not induce any change in cooperativity (Fig. 2). Saturation of the urate effect was not clearly evident. Whether it occurred at higher urate concentrations could not be tested owing to the low solubility of urate in haemolymph (McNabb & McNabb, 1980). The ratio ΔlogP₅₀/Δlogfurate], a numerical estimation of the urate effect, was −0·026 at pH 7·2 and −0·029 at pH 7·.6, values which were not statistically different (P>0·5), again suggesting that the urate effect was independent of pH.

For each of the six experimental series, we determined L-lactate and urate concentrations. Data in Table 1 came from triplicate measurements on pooled haemolymph and did not allow evaluation of individual variations. Nevertheless, major trends could be observed. A 3-h hypoxic exposure elicited only a small rise in L-lactate level, which remained below 0·5 mmol 1⁻¹ (Table 1), whereas urate concentration increased to a greater extent, from about 0·01 mmol 1⁻¹ in normoxia to over 0. ·10 mmoll⁻¹ in hypoxia, whatever the (Table 1). The difference between normoxic and hypoxic levels tended to be reduced after 48 h of exposure for both L-lactate and urate (Table 1). This seemed mainly to be due to a marked increase in the normoxic groups, especially for urate.

A 3-h exposure to environmental hypoxia increased the haemocyanin oxygen affinity compared with that of normoxic blood, regardless of pH (covariance analysis, P<0·001) (Fig. 3A). At pH7·.6, P₅₀ was about 3·7kPa in the pooled blood of Penaeus exposed at and about 2·9kPa at a of 6·3 kPa (Fig. 3A). A similar value prevails after 3h at a of 4·4 kPa. After a 48-h hypoxic exposure, only a slight, insignificant increase (covariance analysis, P>0·.25) of haemocyanin oxygen-affinity was seen, whatever the (Fig. 3B). Neither the Bohr effect nor the cooperativity were altered between normoxia and hypoxia.

Although the prawns we had at our disposal were relatively large (12–15 g, about 10–12cm long), only small volumes (approx. 0–3–0–7ml) of haemolymph could be withdrawn from the pericardial cavity of any individual. Sampling in the abdominal sinus yielded even poorer results, presumably because of the considerable abdominal muscular mass. This prohibited paired design and individual measurement of all parameters. Moreover, Penaeus japonicus haemolymph forms a dense clot within a few seconds after withdrawal, requiring homogenization before liquid blood can be recovered, and precluding measurements such as in vivo pH and ⁠, parameters which would have allowed further understanding of oxygen transport and hypoxia adaptation. Despite these limitations, our study produced data which will be discussed in the context of present knowledge concerning haemolymph oxygen-binding properties in decapod crustaceans and the in vivo responses of the oxygen transport system to environmental hypoxia.

Oxygen binding to Penaeus japonicus haemocyanin is characterized by a high level of cooperativity (n₅₀≈4) and a large Bohr shift. The Bohr effect was similar for dialysed and whole blood (see Figs 1 and 3, for example), amounting to Φ= −0.99 at pH7·4 and Φ= −2.01 at pH7·8. These data agree with previous studies of penaeid haemocyanin (Brouwer et al. 1978; Mangum, 1982; Mangum & Burnett, 1986) which also found the combination of a high Bohr effect and a high cooperativity in P. setiferas, P. monodon and P. duorarum. Such a high pH-sensitivity of oxygen affinity seems to be typical of natantian decapods, since similarly high values have been reported for Palaemon adspersus (Φ= −2.0 and −0·9 at pH7.85 and 7·40, respectively, Weber & Hagerman, 1981) and Palaemon elegans (Φ=−1. ·17 for pH between 7·4 and 8.0, Taylor et al. 1985).

Penaeus japonicus haemocyanin exhibits a moderate lactate effect compared with other species. In a recent review, Bridges & Morris (1986) reported values of ΔlogP₅₀/Δlog[lactate] ranging from −0·04 in Ocypode saratan to −0·66 in Palaemon serratas. Our value of −0·077 for Penaeus japonicus is in the lower range among crustaceans. Furthermore, as is the case in most decapods investigated, this coefficient appears to be relatively independent of pH in the physiological range. It is also interesting to note that, in Penaeus japonicus, a moderate lactate effect occurs in combination with a strikingly large Bohr effect. The roughly proportional relationship between these two effects suggested by Bridges & Morris (1986) could not therefore be confirmed in Penaeus. This, however, makes sense physiologically, since only small amounts of lactate appear in the haemolymph during hypoxia in this species.

Oxygen affinity was increased when urate was added to dialysed Penaeus haemolymph. The presence of a urate effect enhancing oxygen binding by crustacean haemocyanins, first discovered in the crayfish Austropotamobius pallipes (Morris et al. 1985), has been reported in some decapod species -Callinectes sapidus, Carcinus maenas, Palaemon serratus, Homarus vulgaris -but seems to be absent in others, for example Coenobita clypeatus (Morris & Bridges, 1986). Because of the recent finding of the effect of urate, there are few complete studies reporting values of ΔlogP₅₀/Δlog[urate]. Morris et al. (1985) found a value of −0·39 for the crayfish Austropotamobius pallipes, and Lallier (1988) measured a value of −0.22 in Carcinus maenas at pH7·8 and 15 °C. Compared with these data, the urate effect found in Penaeus japonicus seems to be quite small (−0. ·029 at pH7. ·6 and 25°C). Nevertheless, it is interesting to note that, despite the relative primitiveness of penaeids among decapod crustaceans (Bowman & Abele, 1982), both lactate and urate effects are present in Penaeus japonicus. This suggests that lactate and urate sensitivities may be early characteristics of decapod crustacean haemocyanins: lack of one or both of these effects in some species (e.g. Procambarus clarkii, Mangum, 1983b; Glyphocrangon vicaria, Arp & Childress, 1985; Holthuisiana transversa, Morris et al. 1988a; and more generally air-breathing species, Morris et al. 1988b) may therefore indicate that they have subsequently disappeared.

The physiological significance of the modulation of the haemocyanin oxygen affinity by organic cofactors remains unclear. That L-lactate is the main anaerobic end-product in decapod crustaceans suggests that the lactate effect might participate in the adjustment of haemolymph oxygen transport in situations such as environmental hypoxia or muscular exercise. However, for lactate and urate to be considered as physiological modulators, one must demonstrate first that they accumulate in haemolymph, and then that this results in an increase of the oxygen affinity in vivo.

Accumulation of L-lactate in decapod crustaceans during environmental hypoxia has been documented largely by biochemical (Zebe, 1982; Gade, 1983) or physiological studies (Bridges & Brand, 1980; Spotts, 1983; Mauro & Malecha, 1984). None of these studies, however, describes a steady-state relationship between haemolymph lactate concentration and ambient oxygen levels, which would ensue, at least to a certain extent, if the lactate effect was stabilizing oxygen transport to the tissues. In a recent work on resting crabs, Carcinus maenas, under precisely controlled ⁠, L-lactate has been shown to accumulate in haemolymph only under severely hypoxic conditions ( below 2·3 kPa at 15°C), suggesting that lactate is produced in substantial amounts or, alternatively, is released from the tissues only below a threshold (Lallier et al. 1987). At least during environmental hypoxia, the lactate effect may provide only an emergency response. Present data show that there is no large increase in L-lactate concentration in the haemolymph of quiescent Penaeus japonicus exposed to a as low as 4. ·4 kPa at 25°C. These conditions can be considered as a relatively severe hypoxia, since the critical partial pressure (Pc) is about 4–5 kPa for Penaeus japonicus at this temperature (Truchot & Jouve-Duhamel, 1985). In the shrimp Crangon crangon, haemolymph L-lactate concentration increases only below a of 2 kPa at 20°C (Hagerman & Szaniawska, 1986), also supporting the notion of a threshold. To test whether more severe hypoxia could induce a marked accumulation of L-lactate in Penaeus japonicus haemolymph, an experiment was conducted at a of 2·7kPa (25°C). This resulted in considerable mortality among prawns within a few hours, suggesting that resting Penaeus japonicus do not rely significantly upon anaerobic metabolism under sublethal hypoxic conditions.

The pattern of urate release in haemolymph of crustaceans during hypoxia seems to be quite different from that of lactate, though little studied. In a previous work on Carcinus maenas (Lallier et al. 1987), urate levels in haemolymph were found to be closely related to water oxygenation, increasing proportionately with decrease after hypoxic exposure, and decreasing after hyperoxia. An increase of urate concentration has also been reported in hypoxic crayfish (Czietrich et al. 1987). These studies also emphasized the large variations of urate concentrations among individuals, though they were apparently in the same physiological conditions. Such individual variability could not be assessed in our study because of the small volume of haemolymph sampled from each prawn. Nevertheless, measurements on pooled samples clearly show an important and stupid increase in haemolymph urate concentration upon hypoxic exposure, suggesting that the urate response in P. japonicus may be similar to that observed in Carcinus. There are, however, few data concerning purine catabolic pathways in crustaceans, especially for urate production in the cells and subsequent release into haemolymph under hypoxic conditions.

Increased haemocyanin oxygen-affinity has been described previously in the haemolymph of hypoxic crustaceans (McMahon et al. 1978; Burnett, 1979), but this was attributed solely to the concomitant elevation of blood pH. Whether oxygen affinity increases in vivo independently of pH, as would be expected if there is an action of lactate and urate, and possibly of other unidentified organic cofactors, remains poorly documented. In the crayfish Orconectes rusticus exposed for 3×5 weeks to moderate ambient hypoxia, Wilkes & McMahon (1982) found an increased oxygen affinity, of which only 50% was accounted for by pH variations, the remaining 50 % being unrelated to changes in ionic composition of haemolymph (i.e. no effect of calcium). In Homarus vulgaris, Bouchet & Truchot (1985) also found a pH-independent increase of oxygen affinity in haemolymph sampled from hypoxic lobsters, which was only partly due to elevated L-lactate concentrations. In the present study, there was also a moderate, pH-independent increase of oxygen affinity after a 3-h hypoxic exposure in Penaeus, which had disappeared after 48 h. This increase may be almost fully explained by the concomitant in vivo variations of L-lactate and urate concentrations in the haemolymph. This would exclude the possibility of another unidentified cofactor involved during hypoxia in Penaeus. However, it must be recalled that this interpretation relies upon the assumption that the effects of urate and lactate are additive, a factor not assessed in this study. It should be noted that, in the crayfish Austropotamobius pallipes, the effect of L-lactate was found to be markedly reduced in the presence of high urate concentrations (Morris et al. 1986), suggesting that these effects are agonistic.

Most water breathers, such as fishes and crustaceans, increase their ventilatory flow rate when faced with declining ambient oxygen tensions (Dejours, 1981). Hyperventilation generally results in a hypocapnic alkalosis, as has been observed in many crustacean species (see Truchot, 1987). Thus, although blood pH could not be measured in this study, it may increase during environmental hypoxia in Penaeus japonicus. Because of the large Bohr shift of Penaeus japonicus haemocyanin, this elevation of pH should lead to a large increase in haemocyanin oxygen-affinity. Therefore, the in vivo increase of affinity during hypoxia should be the result of the synergistic action of pH, L-lactate and urate increases.

This work formed part of FL’s doctoral thesis and was supported by a fellowship from IFREMER, France.