ABSTRACT
In the days immediately after moulting, manipulations of external pH, [HCO3−], and [Ca2+] were used to determine the nature of the rapid net Ca2+ influx and attendant apparent net H+ efflux in the blue crab (Callinectes sapidus Rathbun). Both fluxes were strongly inhibited by reductions in external [Ca2+], [HCO3−], or pH. The net Ca2+ influx was reversed at an external concentration of 2·5 mmol 1−1, and both fluxes were reversed by reducing the external [HCO3−] to 0·2 mmol 1−1. The correlation between net Ca2+ flux and apparent net H+ flux was 0·61 (P < 0·01), but the variability and the time course of most experiments indicated that the link was indirect, rather than a direct coupling or cotransport. This conclusion was also borne out by acid-base disturbances that occurred in the low-[Ca2+] treatment. The results are consistent with the hypothesis that inward calcium transport is accompanied by both inward HCO3− transport and outward H+ transport, probably by separate exchanges with ions of like charge such as Na+ and Cl−. Crustecdysone (β-ecdysone) does not appear to be involved in control of these post-moult fluxes and calcification.
INTRODUCTION
METHODS AND MATERIALS
Male and female blue crabs (Callinectes sapidus Rathbun), ranging from 64 to 211 g, were collected with unbaited, shaded pots from Redfish Bay, Texas. Pre-moult crabs, which were recognized by various external signs (Perry, Ogle & Nicholson, 1979), were maintained in individual aquaria in the laboratory until they moulted. Food was offered ad libitum, but feeding usually ceased at 1–2 days prior to the moult. Temperatures in the laboratory ranged from 18 to 22°C, but all experiments were performed between 19·6 and 20·4°C (average 20).
Several hours prior to the beginning of each experiment, the freshly-moulted crabs were gently transferred to darkened acrylic chambers which formed part of a smallvolume, recirculating system. The system was provided with vigorous aeration and temperature control. Animals from which blood samples were required were fitted with small neoprene foam septa, which were glued to the mid-dorsal carapace with cyanoacrylate cement. The experimental protocol consisted of periods of up to 8 h during during which water and blood samples could be taken and various changes of media performed with minimal disturbance to the animals.
Normal sea water (SW, salinity 28–32‰) was used as the control treatment. Acidified sea water (low [HCO3−]) was prepared by gradual titration of sea water to pH 6·7–6·8 with HC1 under vigorous aeration. Asea water with the same pH but with high [HCO3−] was prepared by equilibrating normal sea water with 1% CO2 (=7·5 Torr) in air at least overnight. For manipulations of [Ca2+] an artificial sea water was prepared from distilled water, containing (gl−1): NaCl, 25·12; KC1, 0·70; MgCl2·6H2O, 4·67; NaHCO3, 0·176; MgSO4, 3·13; and CaCl2·4H2O, 1·72. Calcium was reduced appropriately in some solutions without corresponding replacement of chloride. More information on the composition of the various solutions used is given in Table 1.
The ‘apparent net H+ excretion’ is an operational term which includes excretion of H+ or NH4+ ions and/or uptake of OH− or HCO3− ions. It was measured at each time interval by titration of a 10·0-ml water sample to pH 4·000 with 0·020 mol 1−1 in a micrometer burette, and by adding the total ammonia excreted to the difference in titration values over the time interval (cf. Cameron & Kormanik, 1982; Cameron & Wood, 1985). Absolute accuracy to the third decimal point is not claimed, but with low-drift electrodes (less than 0·005 pH day−1), the difference between successive samples could be measured with a reproducibility of ± 1·2μ 1, which corresponded to an overall sensitivity of approximately 10 μ equiv net H+ change in the system. Calcium and magnesium concentrations in the sea water were measured by atomic absorption spectrophotometry, and calcium in blood samples by a colorimetric cresolphthalein Complexone method (Sigma 586). The colorimetric assay required less sample volume; a series of standards and samples was analysed with both methods, giving essentially identical results. Total ammonia concentration in water was measured using the phenolhypochlorite method (Solorzano, 1969), and in blood samples with an enzymatic method (Sigma 170-UV). Total CO2 in water and blood were measured with a conductometric apparatus (Capni-Con 3, Cameron Instrument Co.), and blood pH with a capillary type microelectrode (Radiometer). Results were analysed with standard statistical methods, and are presented as ± 1 S.E. unless otherwise stated.
RESULTS
Effects of low pH and [HCO3−]
The effects of the acidified sea water and high CO2 sea water treatments on six crabs are shown in Fig. 1. In acidified sea water (pH = 6·7, [HCO3−] = 0·22mmol 1−1), both net Ca2+ influx and apparent net H+ efflux were strongly inhibited (P< 0·01, analysis of variance, ANOVA). The high-CO2 sea water treatment, which lowered pH by the same amount and increased [HCO3−], did not, however, significantly (by ANOVA) inhibit either net Ca2+ influx or apparent net H+ efflux (Fig. 1).
Effects of altered sea water [Ca2+]
The effects on seven crabs of reducing external [Ca2+] to 2·45 mmol 1−1 are shown in Fig. 2. Originally the artificial sea water was intended as an additional control, but due to excess hydration of the reagent and attendant weighing errors, the [Ca2+] was reduced from 9·49 to 7·23 mmol 1−1 (Table 1), which significantly (P<0·01, ANOVA) affected both net Ca2+ influx and apparent net H+ efflux (Fig. 2). Treatment with 2·45 mmol 1−1 Ca2+ resulted in a reversed Ca2+ net flux (i.e. a net efflux), and a 65% reduction in the apparent net H+ efflux.
The low-calcium treatment also had marked effects on the blood. Blood [Ca2+] fell from 11·60 ±0·20 to 5·30 ± 0·34 mmol 1−1 between the initial control and the low calcium treatment, while the blood [Mg24−] remained unchanged at 16·8 ± 0·7 mmol 1−1. There was also an acid-base disturbance (Fig. 3). During the low-calcium treatment, a non-respiratory (‘metabolic’) alkalosis developed, which was probably due to differential effects of the low [Ca2+] on Na+ and Cl− permeability, resulting in changes in the strong ion difference (SID) (Stewart, 1978; Henry & Cameron, 1983).
Effects of elevated [HCO3−] and [Ca2+]
Increased external [Ca2+] did not have any measurable effect on either the net Ca2+ influx or the apparent net H+ efflux (N = 5 crabs). In successive 1·5-h treatments of control SW, high-[Ca2+] and control SW, net Ca2+ influx rates were 8·56 ± 0·87, 7·42±0·67 and 6·25 ±1·73mequivh−1; apparent net H+ efflux rates were 9·30 ± 1·21, 8·82 ± 1·22 and 9·98 ± 0·97 mequiv h−1. There were some significant changes in the blood, with [Ca2+] rising from 9·31 ±0·35 to 10·94 ± 0·48 mmol 1−1, and the blood pH falling slightly, but not significantly. These changes were opposite to those observed in the low-[Ca2+] treatment, but much less pronounced.
By increasing the [HCO3−] from about 2 to nearly 5 mmol 1−1, a small but significant (P<0·05, ANOVA) increase in the apparent net H+ efflux was observed (from 8·52 ± 1·65 to 11·05 ± 1·51 mequivh−1), but there was no change in the net Ca2+ influx (N = 6 crabs). The blood pH in the high-[HCO3−] treatment rose from 7·798 ± 0·037 to 7·937 ± 0·035 (P < 0·01, paired t-test) and the blood [HCO3−] from 9·81 ± 0·61 to 11·22 ± 0·47 mmol 1−1 (P<0·05), with no changes in [Ca2+] or [Mg2+].
Analysis of the combined data
Day-to-day fluctuations in natural sea water composition, plus the variability included in the various experimental solutions provided a wide range of external [Ca2+] and [HCO3−] in the combined experimental series. Pooling all of the data resulted in the relationship between net Ca2+ influx and apparent net H+ efflux shown in Fig. 4. The slope of the line was 0·650 ± 0·089 (±S.D., N=87), indicating a stoichiometry of approximately 3H+:2Ca2+. The relationships between net Ca2+ influx and external [Ca2+], and between apparent net H+ efflux and external [HCO3−] are shown in Fig. 5A and Fig. 5B, respectively.
Both simple and multiple correlation analyses were carried out on the pooled data, and the simple correlation coefficients are given in Table 2. Because some of the variables were changed in a concerted manner, and thus internally correlated, certain partial correlation coefficients were examined. The apparent net H+ efflux, for example, was closely correlated with the net Ca2+ influx when all other variables were held constant (r = 0·61, P< 0·001), but not to an extent greater than shown by the simple correlation coefficient. The apparent net H+ efflux was also about equally correlated with water pH and water [HCO3−], contrary to the apparently greater sensitivity to [HCO3−] shown in Fig. 1. The higher [HCO3−] and lower pH were offsetting effects in that treatment. Finally, the correlation of net Ca2+ influx with external [Ca2+] is only marginally improved by partial analysis.
Possible endocrine involvement
Since the ‘switching on’ of calcium uptake and associated processes after the moult is quite dramatic, some simple experiments were performed to test the involvement of the eyestalk neuroendocrine system. In the first series, cross-transfusions of 1 ml of defibrinated blood were performed between an inter-moult and a post-moult crab. Despite double defibrination, the animals appeared to die from coagulation of the cross-injected blood in the pericardial cavity. Next, eyestalk extracts from 1 day postmoult crabs were prepared by homogenizing the soft tissues from both eyestalks in 0·5 ml saline and injecting them into intermoult crabs. There was no effect of this treatment on either net Ca2+ influx or apparent net H+ efflux. Two further crabs were injected with 50 μg of β-ecdysone on day 3 post-moult, with no measurable effect on any parameter. Finally, measurements of net Ca2+ influx and apparent net H+ efflux were made on two crabs at 1 day post moult, their eyestalks ligated, and on the next day the measurements repeated. Other than the normal slight decline in fluxes with time after moult, there was no detectable effect of the eyestalk ligation.
DISCUSSION
In our previous work we demonstrated that the source of CO2 for newly deposited CaCO3 in the shall had to be at least partly external HCO3−, since the rate of calcification exceeded the metabolic rate of CO2 production, and there was a net uptake of CO2 for several days after the moult (Cameron & Wood, 1985). The data in Fig. 3 actually show a normal partial pressure gradient for CO2 from inside to out, so the metabolic CO2 contribution to carbonate formation is probably very small. That is, if the new carapace were a significant metabolic CO2 sink, the partial pressure of CO2 in the blood would be depressed.
The results in Figs 1 and 2 are consistent with the model of Fig. 6, in that they show that reductions of external [Ca2+] or [HCO3−] reduce the influx of those ions. The data of Fig. 1 also indicate that a reduction of external pH or increase of external [H+], reduces the observed apparent net H+ efflux, as would be predicted from the model in Fig. 6.
The coupling between the movements of Ca2+, HCO3− and H+ is not very tight, however, as indicated by various experiments. In Fig. 2, for example, the low-[Ca2+] treatment actually reversed the Ca2+ flux, but the apparent net H+ efflux continued, albeit at a reduced rate. In the high external [HCO3− ] experiments, an increase in the apparent net H+ efflux was shown without any increase in the net Ca2+ influx; and finally, temporary inequalities of the various fluxes produced acid base disturbances in the blood, as shown in Fig. 3 and discussed in connection with the high external [HCO3−] treatment (above). The link between the various fluxes appears to be indirect, possibly controlled by changes in either the blood acid-base status, or that of various intracellular compartments. Any further conjecture concerning the actual transport mechanisms seems premature on the basis of present evidence.
The data of Fig. 5 seem to show that the uptake mechanisms (for Ca2+ and HCO3−) are saturated at normal sea water concentrations, but the experimental design was not intended to give details information about the shape of the kinetic curve.
It seems quite likely that the post-moult fluxes are under neuroendocrine control. Ecdysone was an early suspect, based on reports that it affects calcium deposition and mobilization in the crayfish (McWhinnie, Cahoon & Johanneck, 1967; McWhinnie, Kirchenberg, Urbanski & Schwartz, 1972). According to some early reports, ecdysone reaches a peak shortly after ecdysis (Adelung, 1967; Faux et al. 1969), but later studies have shown a peak before ecdysis (Hopkins, 1983; Soumoff & Skinner, 1983). Ecdysone has also been accorded a general role in epidermal events during the moult cycle (Highnam & Hill, 1977). Since neither eyestalk ligation, injection of eyestalk extracts, nor direct injection of ecdysone produced any measurable effects, I conclude that neither ecdysone nor the other distinguishable peptide hormones of the eyestalk complex is responsible for post-moult calcification in the blue crab. There is also considerable neurosecretory activity in the central nervous system, however, and it seems probable that a calcium neuro-hormone will eventually be discovered.
ACKNOWLEDGEMENTS
This work was supported by NSF Grants PCM80-20982 and PCM83-15833 to the author.