Ontogeny Of Hemocyanin Function In The Dungeness Crab Cancer Magister: Hemolymph Modulation Of Hemocyanin Oxygen-Binding

Oxygen transport in crustaceans is affected by a myriad of factors. The gill ventilation rate, the gill and tissue perfusion rates and the oxygen-binding characteristics of hemocyanin can all be altered in response to changes in external and internal factors that the organism experiences. A change in environmental salinity can cause changes in hemolymph inorganic ion concentrations (for a review, see Mantel and Farmer, 1983). These changes in ion concentrations in turn affect the oxygen-binding characteristics of crustacean hemocyanins. Increases in Ca²⁺ and Mg²⁺ concentrations cause increases in both the affinity and cooperativity of hemocyanin from the crayfish Procambarus simulans (Larimer and Riggs, 1964). In the green crab Carcinus maenas, Mg²⁺ and Ca²⁺ increase the oxygen affinity of the hemocyanin and Mg²⁺ has an interaction with H⁺ that increases the effect of Mg²⁺ at higher pH (Truchot, 1975). In the blue crab Callinectes sapidus, Ca²⁺ increases hemocyanin oxygen-affinity but has no effect on cooperativity over the physiological range of Ca²⁺ concentration (Mason et al. 1983). Similarly, changes in environmental temperature cause changes in the function of hemocyanin in crustaceans. One of the many species in which this has been clearly demonstrated is Carcinus maenas (Truchot, 1975).

The products of metabolism and other naturally occurring constituents of crustacean hemolymph are internal factors that affect the oxygen-binding properties of hemocyanin. Crustaceans produce L-lactate as a result of anaerobic energy metabolism. L-Lactate raises the hemocyanin oxygen-affinity of many crustacean species, including Carcinus maenas, Cancer pagurus (Truchot, 1980), Callinectes sapidus (Booth et al. 1982) and Cancer magister (Graham et al. 1983). Not all crustacean hemocyanins are affected by the presence of lactate, however (reviewed in Mangum, 1983). Other naturally occurring organic effectors of hemocyanin oxygen-affinity in some species are urate (Morris et al. 1985, 1986), a purine product accumulating in crustaceans exposed to environmental hypoxia (Lallier et al. 1987; Lallier and Truchot, 1989), dopamine and related cardiac neuroamines (Morris and McMahon, 1989; Morris, 1990) and nitrogen metabolites, ammonium and trimethylamine (Sanders et al. 1992).

Developmental changes also affect crustacean oxygen transport. First, hemocyanin structure changes during ontogeny of Cancer magister (Terwilliger and Terwilliger, 1982) and Homarus americanus (Olson and McDowell Capuzzo, 1989). Terwilliger and Terwilliger (1982) and Terwilliger and Durstewitz (1996) have shown that hemocyanin from adult Cancer magister contains a polypeptide chain, subunit 6, that is not present in megalopa or early juvenile hemocyanins. These stage-specific hemocyanins of C. magister have different intrinsic oxygen affinities and yet have the same sensitivity to lactate and pH (Terwilliger et al. 1986).

Second, ionic regulation also shifts during development from megalopa to adult in C. magister, especially with respect to stage-specific differences in Ca²⁺ and Mg²⁺ regulation (Brown and Terwilliger, 1992). Cancer magister adults strongly hyporegulate [Mg²⁺]. Megalopas, first- and fifth-instar juveniles in 100 % sea water have hemolymph Mg²⁺ activities approximately twice as high as in the adult. The Ca²⁺ activities in hemolymph from the different life stages are not significantly different in 100 % sea water, but as ambient salinity decreases the adult and megalopa regulate [Ca²⁺], and the first- and fifth-instar juveniles hyperconform. Physiologically relevant changes in Ca²⁺ and particularly in Mg²⁺ concentrations cause differential effects on the oxygen affinity of the hemocyanin from first-instar juvenile and adult crabs (Terwilliger and Brown, 1993). Information on oxygen affinity and the effects of organic and inorganic effectors on purified hemocyanin, and studies on changes in osmoregulation during crustacean development, have often been analyzed to determine whether there are trends relating to environment or activity (see Young, 1972a,b; Mangum, 1983; McMahon, 1986; Charmantier and Charmantier-Daures, 1991). In C. magister, developmental changes in hemocyanin structure and function appear to be more closely related to ontogeny of ionic and osmotic regulation than to habitat selection.

Crustacean hemolymph functions as an important interface between the organism and the environment and is critical to the animal’s internal transport, not only of oxygen-carrying hemocyanin but also of metabolites that may influence oxygen transport. By studying different stages of a single species, such as C. magister, that share an identical evolutionary history, we can explore both developmentally linked and habitat-related modifications in respiratory physiology without many of the complications of interspecific comparisons. In the present paper, we investigate whether the intrinsic differences in oxygen affinity between juvenile and adult crab hemocyanins are reflected in the oxygen-binding characteristics of whole hemolymph under in vivo conditions. We bring together the developmental changes in (1) ion regulation, (2) sensitivity of hemocyanin oxygen-affinity to divalent cations and (3) hemocyanin structure and function to examine how the combined effects of these changes are integrated to modulate the oxygen-transport properties of whole hemolymph from various life stages of C. magister. We find that the P₅₀ values of whole hemolymph, measured at what might be in vivo pH, and stage-specific divalent cation levels are surprisingly similar across developmental stages.

Cancer magister megalopas, juveniles and adults were caught and maintained as previously described (Terwilliger and Brown, 1993).

Hemolymph samples were obtained from the infrabranchial sinus of fifth-instar juvenile and adult crabs with a needle and syringe. Hemolymph samples were taken from the same place in first-instar juveniles, but the sampling was performed using a microcapillary pipette. Megalopa hemolymph was obtained by puncturing the heart with a microcapillary pipette. In order to avoid puncturing the yellowish-brown digestive tract, sampling of megalopas and first-instar juveniles was performed under a dissecting microscope. Hemolymph from five intermolt adults, 10 intermolt fifth-instar juveniles, 150 intermolt first-instar juveniles or 250 premolt megalopas was pooled for each whole hemolymph sample.

The fresh whole hemolymph samples were allowed to clot on ice for 30 min and then centrifuged at 12 000 g for 10 min at 4 °C to remove the aggregate. Samples to be used for tonometric oxygen equilibria were stored at 4 °C and used within 24 h. Samples of whole hemolymph for determination of oxygen equilibria using a diffusion method (Sick and Gersonde, 1969) were immediately centrifuged, stored at −73 °C and transported on dry ice to the University of Calgary, Alberta, Canada.

Samples of purified 25S hemocyanin from first-instar juveniles and adults were prepared by chromatography of fresh whole hemolymph on BioGel A5-m, as previously described by Terwilliger and Brown (1993).

The fresh whole hemolymph samples used in oxygen equilibrium determinations contained an unidentified component with a low broad absorbance peak at approximately 325 nm. This peak was apparent in the deoxygenated spectra of whole hemolymph but not in the deoxygenated purified hemocyanin (Fig. 1). Owing to the presence of this interfering component, the absorbance used to calculate the hemolymph oxygen-carrying capacity was based on the difference between the oxyhemocyanin and the deoxyhemocyanin absorbances at 340 nm (10 °C), adjusted for the altered baseline absorbance caused by the unidentified component. Examination of 47 paired sets of C. magister 25S oxyhemocyanin and deoxyhemocyanin absorbances at 340 nm revealed that deoxyhemocyanin absorbance for whole hemolymph containing the interfering component should be: OD_{deoxy,adjusted}=0.09(OD_oxy−OD_{deoxy,measured}) (r²=0.86). Given this relationship, the difference in absorbance of oxygenated and adjusted deoxygenated whole hemolymph was used in oxygen-carrying capacity calculations as the absorbance of fully oxygenated hemocyanin in the sample.

The extinction coefficient at 280 nm of a 1 % solution of C. magister hemocyanin in a 1 cm pathlength cuvette, ⁠, is 15 (Nickerson and Van Holde, 1971). Based on a ratio of 0.2 for the absorbance of purified C. magister oxyhemocyanin at 340 nm and 280 nm, the value of the extinction coefficient at 340 nm would be 3 (Graham, 1983). Since each hemocyanin subunit combines reversibly with one molecule of oxygen, the hemocyanin oxygen-carrying capacities of the whole hemolymph samples were calculated from the absorbance at 340 nm of a fully oxygenated sample, a value of at 340 nm, and an average subunit molecular mass of 75 kDa. The total oxygen-carrying capacity of the hemolymph includes the amount of oxygen physically dissolved in solution when ambient ⁠.

L-Lactate concentrations in the whole hemolymph samples were assayed enzymatically (Boehringer-Mannheim, no. 139 084). Urate concentrations in the samples were also assayed by enzymatic assay (Sigma Chemical Co., procedure no. 685).

Fresh whole hemolymph samples for oxygen binding were buffered in stage-specific saline solutions (Table 1). The ion concentrations for each stage-specific saline corresponded to the concentrations measured in the hemolymph of megalopas, first-instar juveniles and fifth-instar juveniles in 100 % sea water, as reported by Brown and Terwilliger (1992). The saline for the adult hemolymph was made to match the formulation used by Graham et al. (1983). Titrating with Trizma base (Sigma Chemical Co.) rather than HCl provides a more consistent level of chloride for ion studies and/or analytical ultracentrifugation.

Oxygen equilibria of whole hemolymph samples were determined by two methods. First, oxygen equilibria were determined tonometrically (Benesch et al. 1965) at 10 °C using 0.4–0.5 ml of diluted whole hemolymph. Fresh whole hemolymph was diluted, 1:2 or 1:3, in appropriate stage-specific saline in order to obtain the desired pH and an absorbance at 340 nm near 0.3 in a 1 mm pathlength cuvette. Second, oxygen binding was also measured on previously frozen samples of undiluted whole hemolymph at 10 °C at the University of Calgary, using the diffusion chamber method described by Sick and Gersonde (1969). The pH of these samples was adjusted by using gas-mixing pumps to vary ⁠. Oxygen equilibria for purified 25S hemocyanin of first-instar juvenile and adult were also determined tonometrically at 10 °C in stage-specific salines: first-instar juvenile 25S hemocyanin in first-instar juvenile saline; adult 25S hemocyanin in both first-instar saline and adult saline. Purified 25S hemocyanin samples were dialyzed against the appropriate salines at 4 °C for 24 h prior to oxygen equilibrium determinations.

The Bohr coefficients (slope of logP₅₀versus pH) and the oxygen affinities were compared by analysis of covariance (ANCOVA). Mean values of cooperativity (n₅₀) were compared by Student’s t-test. P<0.05 was considered significant. Statistical analyses were performed using SYSTAT version 4.1 (SYSTAT, Inc.).

The concentration of hemocyanin in the hemolymph changes with developmental stage (Table 2). The intermolt first-instar juvenile had the lowest hemocyanin concentration, and the adult had the highest. The intermediate level seen in the megalopa is probably due to its premolt condition, the time in the molt cycle when hemocyanin levels reach a maximum (Terwilliger and Otoshi, 1994). These differences in concentration have a very pronounced effect on the total oxygen-carrying capacities of the four stages.

No urate was detected in any of the whole hemolymph samples from any of the four stages examined. L-Lactate levels varied among individual samples and among stages. The lowest detectable level of L-lactate was 0.05 mmol l⁻¹. L-Lactate concentrations in the megalopa hemolymph samples (N=7; 250 megalopa per pooled sample) for tonometry ranged from the limit of detection to 0.57 mmol l⁻¹. Only one first-instar juvenile hemolymph sample (N=6; 150 first-instar juveniles per pooled sample) for tonometry had a detectable L-lactate level, 0.47 mmol l⁻¹. There was no detectable L-lactate in any of the fifth-instar juvenile samples (N=8; 10 fifth-instar juveniles per pooled sample) for tonometry. L-Lactate concentration in adult hemolymph samples (N=12; five adults per pooled sample) for tonometry ranged from 0.18 to 3.44 mmol l⁻¹.

Among the samples for diffusion chamber oxygen equilibrium determination, the hemolymph L-lactate concentrations were 1.10 mmol l⁻¹ in the megalopa hemolymph (N=1; 250 megalopas), 0.69 mmol l⁻¹ in the first-instar juvenile hemolymph (N=1; 150 first-instar juveniles), 0.30 mmol l⁻¹ in the fifth-instar juvenile hemolymph (N=1; 10 fifth-instar juveniles) and 0.49 mmol l⁻¹ in the adult hemolymph sample (N=1; five adults).

The oxygen affinities (measured tonometrically) of megalopa, first-instar juvenile and fifth-instar juvenile fresh whole hemolymph samples with endogenous L-lactate and ion concentrations are indistinguishable (Fig. 2). The oxygen affinity of adult fresh whole hemolymph, with endogenous L-lactate and ion concentrations, is higher than those of the the megalopa and juvenile whole hemolymph samples (Fig. 2). There are no significant differences in the slopes of the regressions of logP₅₀versus pH of the whole hemolymph from the four stages (see Fig. 2 legend).

The P₅₀ of each sample in Fig. 2 was adjusted to what it would be if each sample contained an identical concentration of L-lactate, equal to the lowest detectable level of 0.05 mmol l⁻¹. The value used to adjust the P₅₀ values of all four stages was ΔlogP₅₀/Δlog[L-lactate]=−0.29. This value is based on the coefficients for the relationship of logP₅₀versus log[L-lactate] for adult C. magister hemocyanin (−0.287) and for juvenile hemocyanin (−0.291) (Terwilliger et al. 1986). When the P₅₀ values of the whole hemolymph samples were adjusted for the effect of L-lactate, there were no significant differences in oxygen affinity or Bohr coefficient among any of the four stages (Fig. 3).

When the affinities of the frozen whole hemolymph samples (diffusion chamber method) were adjusted to 0.05 mmol l⁻¹ L-lactate, they overlapped entirely with the P₅₀ values of the tonometer samples in the physiological pH range, pH 7.2–8.2 (Graham et al. 1983; Brown and Terwilliger, 1992) (Fig. 4). Whether the slight trend in increasing P₅₀ for tonometric analyses is indicative of a specific effect of CO₂ is unknown.

Some crustacean hemocyanins exhibit a specific CO₂ effect while others do not (see Mangum, 1997). The Bohr coefficients and the statistically adjusted mean values for logP₅₀ were not significantly different between the tonometer data and the diffusion chamber data.

The cooperativities of the fresh whole hemolymph samples among the four stages from the tonometric method were not significantly different (n₅₀=3.08±0.48, mean ± S.D.; Fig. 5). The mean cooperativities of the diffusion chamber oxygen equilibria were significantly lower than those of the tonometric samples (n₅₀=2.24±0.24, mean ± S.D.; P<0.05; Fig. 5). This decrease in cooperativity in the frozen samples is consistent with previous reports on the effects of freezing hemocyanin (Mangum, 1983; Morris, 1988).

The oxygen affinities of purified 25S hemocyanins from first-instar juvenile and adult crabs were compared under different salinity regimes. The relationship between logP₅₀ and pH for purified 25S hemocyanin from first-instar juveniles and adults in stage-specific salines is shown in Fig. 6. When both hemocyanins were dialyzed against first-instar juvenile saline, the P₅₀ values at pH 7.8 differed by 54 %: 4.68 kPa for the first-instar juvenile and 2.16 kPa for the adult. When the two types of hemocyanin were dialyzed against their respective stage-specific salines, the affinities differed by only 28 %: 3.39 kPa for the adult and still 4.68 kPa for the first-instar juvenile. Thus, a major part of the intrinsic difference in stage-specific hemocyanins could be reduced by stage-specific differences in hemolymph ion concentrations. Cooperativity values for both hemocyanins in first-instar juvenile saline were significantly lower than cooperativity values for adult hemocyanin in adult-type saline (Fig. 7).

The oxygen affinities of hemolymph from juvenile and adult Cancer magister are indistinguishable at in vivo pH and cation levels. These results are initially surprising in the light of the significant differences in intrinsic oxygen affinities of 25S hemocyanins from the various life stages of C. magister (Terwilliger and Terwilliger, 1982; Terwilliger and Brown, 1993). The stage-specific hemocyanins also have different sensitivities to the allosteric effectors Ca²⁺ and Mg²⁺. While both ions increase hemocyanin oxygen-affinity, the P₅₀ of juvenile hemocyanin is more sensitive to [Ca²⁺], and especially to [Mg²⁺], than is adult hemocyanin (Terwilliger and Brown, 1993). In addition, C. magister undergoes ontogenetic changes in ionic and osmotic regulation (Brown and Terwilliger, 1992) such that both megalopas and first-instar juveniles have more than twice the adult hemolymph Mg²⁺ level. We suggest that the intrinsically low-affinity hemocyanin of the first-instar juvenile serves to counterbalance the increase in oxygen affinity caused by both high hemolymph Mg²⁺ levels and increased sensitivity to [Mg²⁺]. As the ionic regulatory capabilities of the young crab develop and it becomes better able to maintain low Mg²⁺ levels in the hemolymph, the synthesis of higher-affinity adult hemocyanin is initiated, and it gradually replaces juvenile hemocyanin. Expression of adult-type hemocyanin, as indicated by the appearance of hemocyanin subunit 6 mRNA, begins at approximately the sixth juvenile instar (Durstewitz and Terwilliger, 1997). It is the combination of developmental differences in (1) hemocyanin intrinsic affinity and (2) hemocyanin sensitivity to allosteric effectors, in parallel with (3) developmental changes in capacities for ion regulation, that interact and allow the crab to maintain a constant oxygen affinity of its blood as ontogeny progresses. Divalent cations are clearly significant effectors of crustacean hemocyanin oxygen-binding properties. The importance of the differences in hemolymph Mg²⁺ levels and in sensitivity to [Mg²⁺] as an allosteric modifier are emphasized when we compare the results of oxygen-binding experiments on first-instar juvenile 25S hemocyanin and adult 25S hemocyanin in stage-specific salines. When functional studies were carried out on the two types of hemocyanin in identical salines, the affinity of the first-instar juvenile hemocyanin was 54 % lower than that of the adult (Fig. 5), consistent with earlier reports on purified hemocyanin. Half of this intrinsic difference in oxygen affinity at pH 7.8 between the two stages was eliminated when first-instar juvenile and adult hemocyanins were each tested in their own stage-specific saline. The most notable differences between the salines are the Ca²⁺ and Mg²⁺ concentrations.

Both urate and lactate increase the oxygen affinity of hemocyanin from a variety of crustaceans (Truchot, 1980; Graham et al. 1983; Mangum, 1983; Morris et al. 1985, 1986). We can rule out the influence of urate and L-lactate in these experiments, however. First, no urate was detected in any of the hemolymph samples. Second, the effects of L-lactate on the oxygen affinity of C. magister hemocyanin are pronounced (Graham et al. 1983), and circulating L-lactate levels can vary widely, depending on the activity of an individual crab (Booth et al. 1982; McDonough, 1990). We have accounted for L-lactate differences among samples, however, in order to compare the function of the hemolymph. It should be noted that there may be stage-specific differences in the capacity of C. magister to produce L-lactate. Sastry and Ellington (1978) reported that the larval stages of Cancer irroratus differ in the activity of lactate dehydrogenase, which may imply stage-specific differences in anaerobic capacity and L-lactate production.

The similarities in whole-hemolymph P₅₀ are due in large part to the interplay between the lower-affinity hemocyanin and high Mg²⁺ levels of the juvenile versus the higher-affinity hemocyanin and lower Mg²⁺ levels of the adult crab. Additional stage-dependent factors in the hemolymph also contribute to equalizing the oxygen affinities of juvenile and adult hemolymph. We did not measure hemolymph concentrations of dopamine, other cardiac neuroamines or ammonium metabolites, and there have been suggestions of unidentified compounds in other crustacean hemolymphs that affect oxygen affinity (see Bridges et al. 1997; Lallier and Truchot, 1997).

The cooperativity of adult C. magister 25S hemocyanin was higher in adult saline than in juvenile saline. This may be due to independent or additive effects of higher concentrations of both Ca²⁺ (a specific divalent cation effect) and sodium chloride (an ionic strength effect) in adult saline. Ionic effects on cooperativity have been shown for other crustacean hemocyanins (for a review, see Mangum, 1997).

It has been known for some time that the oxygen-carrying capacity of crustacean hemolymph varies with the stage of the molt cycle (Drach, 1939), partly because of the increased water uptake just before ecdysis and partly because of a change in biosynthesis of hemocyanin (Mykles, 1980; Mangum et al. 1985; Spindler et al. 1992). The present study indicates that the oxygen-carrying capacity of C. magister hemolymph varies with developmental stage as well, increasing as the animal gets older. The low oxygen-carrying capacity of the hemolymph of juvenile crabs may impose constraints on oxygen transport and, therefore, on the activity of the crabs under certain conditions; this remains to be determined. The oxygen-carrying capacity calculated for the adult crab (1.75 ml O₂ dl⁻¹) is comparable to values reported by other researchers for adult C. magister (McMahon et al. 1978) and Callinectes sapidus (Booth et al. 1982).

In summary, the functional importance of the low-affinity hemocyanin in juvenile C. magister is to compensate for immature renal development and weak ion regulation during early development. This study demonstrates that the inherent functional differences in the oxygen transport proteins of this crab are modulated by stage-specific constituents of the native hemolymph. There are additional unknown factors in the hemolymph of megalopas, first-instar juvenile, fifth-instar juvenile and adult crabs that shift the intrinsic affinities and cooperativities of the stage-specific hemocyanins so that these respiratory properties are indistinguishable in the whole hemolymph of these stages. Coordinated changes in ion regulation and hemocyanin during development maintain homeostasis of oxygen affinity.

We thank Dr Steven Morris, University of Sydney, and Dr Brian McMahon, University of Calgary, for stimulating discussions, fruitful experiments and generous hospitality at the University of Calgary. This study was supported by NSF grants DCB 8908362, IBN 9217530 and IAB 9603521 to N.B.T.