Crab Gill Intra-Epithelial Carbonic Anhydrase Plays a Major Role in Haemolymph CO₂ and Chloride Ion Regulation

Considerable evidence points to an important role for gill intra-epithelial carbonic anhydrase in CO₂ excretion by fish (cf. recent reviews by Cameron, 1979 and by Haswell, Randall & Perry, 1980). There is also evidence that lung intra-epithelial carbonic anhydrase may be important in mammalian CO₂ excretion (Klocke, 1977; Effros, Chang & Silverman, 1978; Crandall & O’Brasky, 1978). By way of contrast, Aldridge & Cameron (1979) report that in Callinectes sapidus, a crab which is known to possess gill intra-epithelial carbonic anhydrase (Aldridge, 1977), this enzyme is unimportant to CO₂ excretion. Therefore, we decided to investigate the role of gill intra-epithelial carbonic anhydrase in another crab, Pachygrapsus crassipes (Randall).

In water-breathing species, at least part of the CO₂ excreted (presumably HCO₃⁻) is thought to be implicated in the movement of chloride ions between the animal and its environment (Cameron, 1979; Haswell et al., 1980). Since mammalian erythrocyte carbonic anhydrase plays a major role in mobilizing intracellular CO₂ to HCO₃⁻, which in turn bidirectionally exchanges with plasma chloride (Roughton, 1964), we were interested in establishing the relation between haemolymph chloride 10n regulation and gill carbonic anhydrase function in a crab which regulated haemolymph chloride concentration in both hypo-and hypersaline environments. Therefore, the effects of carbonic anhydrase inhibition by sodium acetazolamide (generic name ‘Diamox’) were studied over a range of external salinities where we had established that the animal could regulate its haemolymph chlorinity. The effects of carbonic anhydrase inhibition on chloride regulation, haemolymph CO₂ content and haemolymph pH were investigated in crabs at different salinities.

Pachygrapsus crassipes were collected along the rock jetties of Mission Bay, San Diego, California, and held in ten-gallon glass aquariums thermostated to 23 ± 1 °C. Only male crabs weighing between 10 and 35 g were used. Crabs were fed frozen smelt three times each week, but were starved for 24 h before experimentation.

Crabs acclimated to 34‰, sea water (530 mM chloride) for at least two days were killed quickly and various tissues were dissected from them. Prior to killing, haemolymph samples were taken as described below. Solid tissues were washed with cold assay buffer. All samples were then placed in 20 vol ice-cold assay buffer and disrupted using either ground glass homogenizers or sonication. In no case were the assay results different using these two types of disruption. Homogenates were then centrifuged at 100000g for one hour at 4 °C. The supernatants were decanted for assay and the pellets were resuspended by vortexing in 20 vol ice-cold assay buffer which was 1 % v/v in Triton X-100 (Sigma; a non-ionic detergent). The suspension was centrifuged as before and the supernatant (presumably containing any enzyme which was tightly lipid-associated) decanted for assay. The pellet was resuspended in ice-cold assay buffer and kept for assays. Triplicate 100 μl samples of the supernatants and the pellet suspension were assayed as described above.

To achieve a specific circulating concentration of Diamox in crab haemolymph by injection, the haemolymph volume must be accurately known. Since the haemolymph volume of P. crassipes is unknown, the possibility of Diamox penetrating crab tissues from the sea water was investigated using tritium-labelled Diamox.

Tritium-labelled Diamox was prepared according to the method of Roblin & Clapp (1950) using ³H-acetic anhydride (New England Nuclear). Our product melted from 260 to 261 °C (reported: 258–259 °C; Roblin & Clapp, 1950). The physiological activity of this product was verified by showing that it was equipotent with commercial Diamox (Lederle) as an inhibitor of carbonic anhydrase (from P. crassipes gills) in our assay procedure. Assuming that it was 100% Diamox, the specific activity of our product was 19×10⁹ cpm/mole. Radio-activities were determined by liquid scintillation counting in Aquasol-2 (New England Nuclear). All samples were counted until a statistical sampling error of 3% was achieved.

The function of gill carbonic anhydrase was investigated by inhibiting the gill orbonic anhydrase in vivo with Diamox and measuring the changes in haemolymph pH, CO₂ content and chloride ion concentration. Enzyme function was studied different salinities.

Crabs held in sea water of different salinities and 23 ± 1 °C for a minimum of two days (Prosser, Green & Chow, 1955) were transferred to 6 1 glass aquaria which were completely covered with opaque plastic. A maximum of three crabs were placed in each aquarium containing 2 1 of vigorously aerated sea water of the appropriate salinity. In some tanks, Diamox was added to a final concentration of 2 × 10⁻⁴ M. Crabs were held in these containers for 12 h, removed individually and their haemolymph sampled from the infrabranchial sinus at the base of the third or fourth walking leg. Glass syringes containing the haemolymph samples were kept on ice and the haemolymph pH determined using a Radiometer Acid-Base Analyzer (PHM71) and pH electrode (G297/G2) thermostated to 23±0·1 °C. Total haemolymph CO₂ was measured using the method of Cameron (1971). The chloride ion concentration of haemolymph was determined in triplicate, using the titrametric method of Schales & Schales (1952) with 100 μ samples. In all cases only one haemolymph sample (~0·6 ml) was taken from each crab.

To determine if there was any specific effect of Diamox on CO₂ production, oxygen uptake (which estimates CO₂ production) was measured in crabs at three different salinities before and after 12 h exposure to 2 × 10⁻⁴ M Diamox. Oxygen uptake was determined using a method described by Burnett, Chambers & Coster (1980a).

The following tissues contained no detectable carbonic anhydrase activity: haemolymph, muscle, heart, hepatopancreas and male gonads. The possibility that Triton X-100 in the second supernatant and resuspended pellets was inhibiting any carbonic anhydrase present was ruled out. When a sample of gill supernatant, which exhibited carbonic anhydrase activity (described below), was made 1% v/v in Triton X-100 and then assayed, it was evident that Triton X-100 at that concentration did not inhibit P. crassipes carbonic anhydrase. The possibility that carbonic anhydrase in non-gill tissues was present at very low concentrations or was relatively inactive due to some endogenous inhibitor released by disruption, was made unlikely by our finding that concentrated preparations, made by pooling materials from five to ten crabs in a volume of buffer equal to what we would normally use for tissue from one crab, still exhibited no detectable carbonic anhydrase activity. Suspensions of large gill fragments exhibited no activity, suggesting that the enzyme is not located extra-cellularly.

Carbonic anhydrase activity in the Triton X-100 free supernatants of individual crab’s gills (all gill filaments were used from both sides of each crab) ranged from 4·74 mg⁻¹ to 12·82 mg⁻¹. The average of 18 animals was 7·75 mg⁻¹. There appears to be no carbonic anhydrase with which membrane is tightly associated, since neither the Triton X-100 supernatant nor the resuspended pellet from the gill homogenates exhibited detectable activity. All these crabs had been acclimated to 34‰, sea water.

When Diamox was added to a final concentration of 2 × 10⁻⁴ M in the assay reaction mixture, we detected no carbonic anhydrase activity in supernatants of gill tissue (Table 1). At a concentration of 5 × 10⁻⁵ M, Diamox partially inhibited the carbonic anhydrase activity.

Since Diamox is a basic compound we thought it necessary to investigate the direct effect of Diamox at 2 × 10⁻⁴ M on haemolymph pH in vitro. When Diamox was added to haemolymph and the mixture air-equilibrated for 20 min, no change in pH was observed (Table 2).

To determine whether Diamox would penetrate P. crassipes in vivo from the external medium, and to determine how long it would take for a given haemolymph concentration to be reached in the presence of a specified water concentration, six crabs were placed in 34‰ sea water containing 2 × 10⁻⁴ M radioactive Diamox (3800 cpm/ml). Three crabs were removed and their haemolymph sampled at one hour and another three crabs at three hours after introduction of radioactive Diamox. Under these conditions, radioabel in the external medium and crab haemolymph equilibrated within one hour (Table 3).

Pachygrapsus crassipes regulates its haemolymph chloride ion concentration over a wide range of salinities (Fig. 1). Twelve hours exposure to 2×10⁻⁴M Diamox severely impairs the crab’s ability to regulate haemolymph chloride under hypo-osmotic (P⪡ 0·001; analysis of covariance) but not under hyperosmotic conditions (P~ 0·055; analysis of covariance).

Diamox decreases O₂ uptake (P<0·05 according to a paired t test; Fig. 2), and by inference decreases CO₂ production assuming RQ remains constant. On the other hand, Diamox produces an elevation in haemolymph CO₂ content at all salinities studied (Fig. 2); the increase is greatest at low salinities, the same salinities at which haemolymph chloride regulation was most impaired by Diamox (Fig. 1). Furthermore, haemolymph CO₂ content at the highest salinity is significantly different from haemolymph CO₂ content at the lowest salinities (P< 0·005; Newman-Keuls test), While haemolymph pH changes little in control crabs over the range of salinities studied, it is raised by Diamox exposure at low, but not high, salinities (Fig. 2).

Our results indicate that the gills of P. crassipes, like that of C. sapidus (Aldridge, 77), contain a Diamox-sensitive catalyst of the hydration of CO₂. We assume this catalytic activity to be due to a carbonic anhydrase. The tissue disruption and extraction experiments strongly suggest that this enzyme is located intracellularly, within the gill epithelial cells, and is in a soluble, or at most loosely membrane-associated form. Our experiments do not distinguish different intracellular pools of this enzyme. Given the extensive infolding of gill epithelial tissues (Copeland, 1968; Taylor & Butler, 1978), the possibility that the enzyme is distributed among functionally distinct subcellular compartments must be kept in mind when speculating on the function of the enzyme.

The lack of any detectable activity in haemolymph, muscle, heart, hepatopancreas and male gonads does not prove the absence of the enzyme in these tissues; it is in principle impossible to prove the absence of an in vivo functional enzyme using in vitro techniques. This caveat about the extension of in vitro results to in vivo reality also concerns the interpretation of our values for the specific activity of the gill carbonic anhydrase; all that is significant about these data are that the specific activities of gill extracts are well above zero, and that this in vitro activity is subject to a dose-dependent inhibition by Diamox.

We suggest that Diamox inhibits gill carbonic anhydrase in vivo as well as in vitro. This suggestion is supported by our experiments with radiolabelled Diamox, showing rapid equilibration of the label between sea water and haemolymph at Diamox concentrations which are effective in totally inhibiting carbonic anhydrase in vitro and by our observations that haemolymph CO₂ is elevated while O₂ consumption (and by inference CO₂ production) is depressed. The use of O₂ consumption in estimating CO₂ production does, however, present some difficulties. A recent study on a land crab indicates that the ratio of CO₂ production to O₂ consumption (RQ) increases from 0·7–0·9 to above1·0 when the crabs exercise (Herreid, Lee & Shah, 1979). It is impossible to evaluate changes in RQ accompanying Diamox exposure from evidence presented here. However, in the present study, even a 25% increase in RQ in Diamox-exposed crabs would not lead to CO₂ production greater than in control crabs, since O₂ consumption declines in the presence of Diamox by 20–25%. This makes it unlikely that changes in tissue CO₂ production contribute to elevated haemolymph CO₂ contents (Fig. 2). The most parsimonious explanation of our observations of elevated haemolymph CO₂ is that the gill intracellular CO₂⇌HCO₃⁻equilibrium is: (1) critical to the CO₂ excretion rate, and (2) strongly affected by intracellular carbonic anhydrase.

Another possible interpretation of the elevated haemolymph CO₂ is that Diamox mobilizes CO₂ from some unknown source within the crab into the haemolymph. For example, some recent evidence suggests that crabs can use internal stores of CaCO₃, presumably from the shell, to buffer the haemolymph against severe decreases in pH which may occur when a crab is exposed to air (deFur, Wilkes and McMahon, 1981). Mobilization of internal CO₂ stores during Diamox exposure cannot be completely ruled out in the present study. However, a release of CaCO₃ stores associated with a decrease in haemolymph pH similar to that observed by deFur et al. is unlikely, since pH did not decrease during exposure to Diamox in any of our experiments.

In support of our interpretation that carbonic anhydrase facilitates CO₂ excretion we have recently found that Diamox inhibits the excretion of H¹⁴CO₃⁻ added to the artificial internal perfusate of an isolated gill of Cancer anthonyi in organ culture (Burnett et al. 1980b; Woodson et al. 1980). The posterior gills used in these studies were found to contain Diamox-sensitive carbonic anhydrase activity.

Our results with P. crassipes sharply contrast with those of Aldridge & Cameron (1979) who, using blue crabs (Callinectes sapidus) acclimated to fresh water, showed that Diamox injection had no effect on haemolymph pH or CO₂ content after six hours. In an effort to explain the discrepancies between the two studies, we conducted an experiment on P. crassipes using methods similar to those used by Aldridge & Cameron. We acclimated the crabs to 16‰ salinity water, the lowest salinity at which they can regulate haemolymph chloride, in order to approximate both the direction and degree of chloride regulation found in C. sapidus. Diamox was dissolved in a physiological saline (based on Prosser et al. 1955) approximating the ionic content of the crab’s haemolymph at 16%o salinity. Crabs were then injected with the solution to achieve a final circulating Diamox concentration in the haemolymph of 2·7 × IO⁻³M (after Aldridge & Cameron, 1979 and based on the assumption of haemolymph volume = 30% wet animal weight; Gleeson & Zubkoff, 1977). Haemolymph withdrawn after three hours in one group of crabs and after six hours in another group of crabs was analysed for pH, C0₂ and chloride content. The results (Table 4) are consistent with our findings of an elevation in haemolymph C0₂ content and inhibition of chloride regulation during Diamox exposure. Thus, the discrepancies between the results of the two studies are not likely to be due to the differences in dose or route of administration of Diamox.

We would like to point out, however, that Diamox, a basic compound, has been observed to produce a non-specific effect on carbonic anhydrase activity when used more than millimolar concentrations (Maren, 1977). In addition, repeated haemolymph sampling produces slight haemolymph acidosis and lowering of CO₂ content (McMahon, Butler & Taylor, 1978). Multiple haemolymph sampling (six times within 12 h) on C. sapidus, a crab which is normally very aggressive and excitable, may have shifted haemolymph pH and CO₂ content in directions opposite to that which we observed in P. crassipes. Thus, both the high Diamox concentration and multiple haemolymph sampling in the Aldridge & Cameron (1979) study may have masked differences in haemolymph pH and CO₂ content before and after treatment.

The results of the Aldridge & Cameron (1979) study can be explained if there was a decrease in both CO₂ production and CO₂ excretion. O₂ uptake (and thus CO₂production) measured in two P. crassipes treated with Diamox in a manner similar to that used with C. sapidus showed no significant change after five hours (unpublished results of ours). However, the possibility of a Diamox-induced change in CO₂ production in C. sapidus was not explored. Finally, the contrasting results of the two studies may be related to (1) species-specific differences in the function of carbonic anhydrase in crab gills or (2) the possibility that six hours exposure to Diamox is insufficient time to measure a slowly accumulating effect on haemolymph CO₂ content or pH.

Diamox-sensitive chloride transport (Fig. 1) has been observed in other epithelia: gills of the goldfish (Maetz & Garcia-Romeu, 1964) and rainbow trout (Kerstetter & Kirschner, 1972), intact toadfish (Kormanik & Evans, 1979), the crayfish Astacus leptodactylus (Ehrenfeld, 1974), and frog skin (Ehrenfeld & Garcia-Romeu, 1978; Watlington, Jessee & Baldwin, 1977). A major observation of the present work is that Diamox affects haemolymph CO₂ content most strongly in hyposaline water (Fig. 2). Thus CO₂ excretion and regulation of haemolymph chlorinity under hyposaline conditions may be limited together by the rate at which either CO₂ or HCO₃⁻ can be supplied to (or removed from) certain intracellular transport sites within the gill. This suggestion is really a multipart theory, and we will now consider its various components.

First, it must be admitted that in addition to its known effect as an inhibitor of carbonic anhydrases from numerous different species, Diamox may be a specific inhibitory ligand of the chloride inward-transporting mechanism under hyposaline conditions. We ignore this possibility only because there is lack of evidence for it, there is plenty of evidence for the carbonic anhydrase inhibition role, and finally because we do not know how to settle the problem experimentally.

Given the assumption that the sole site of action of Diamox is carbonic anhydrase, we must then consider that the enzyme is known to catalyse both the hydration of CO₂ and the dehydration of HCO₃⁻ (H₂CO₃ is probably not the substrate for the dehydration reaction; DeVoe & Kistiakowsky, 1961; Khalifah, 1971). Furthermore Diamox appears to be a competitive inhibitor of the hydration reaction but a noncompetitive inhibitor of the dehydration reaction (Maren, 1967). The present data say nothing about the in vivo substrate or prevailing direction of catalysis. The possibility that there are different subcellular pools of the enzyme, and some catalysing one reaction some the other, cannot be ignored either (e.g. the enzyme favours the dehydration reaction in acid media; Khalifah, 1971). Thus, we must consider that Diamox may be making the rate of supply or removal of either (or both) CO₂ or HCO₃⁻ to intracellular sites limiting, and thereby impairing the inward transport of chloride. Clearly the number of possibilities is unpleasantly large. All we can assert, remembering our assumption about the sole site of action of Diamox, is that under hyposaline conditions the inward movement of chloride is coupled to the outward movement of CO₂in such a way that the gill intracellular CO₂⇌ HCO₃⁻ interconversion rate, as controlled by carbonic anhydrase, is critical to CO₂ and Cl⁻ translocation processes.

The inhibition of chloride ion regulation during Diamox exposure at low salinity is associated with an increase in haemolymph pH (Fig. 2). The alkalosis is probably an effect of the breakdown of ion regulation (and thus osmotic regulation), and not a result of carbonic anhydrase inhibition per se. Mangum & Towle (1977) have attributed the haemolymph alkalosis associated with the migration of Callinectes sapidus to dilute waters to a cellular volume regulation mechanism, in which intracellular free amino acid pools are reduced to accommodate the dilution of the haemolymph. The deamination of the free amino acid pool produces ammonia which passes into the haemolymph and picks up a proton, raising the pH. A similar situation is likely to exist in P. crassipes when its haemolymph suddenly becomes dilute due to a Diamox-induced inhibition of Cl⁻ ion regulation.

This research was supported by University of San Diego College of Arts and Sciences Faculty Research Grants to L.E.B. and P.B.J.W., and by a USD Associated Student Body Student Research Grant to V.C.V. Diamox was the kind gift of Lederle, Inc. We wish to thank David and Mary Dorman for the synthesis of radiolabelled acetazolamide and preliminary work on the enzyme assay. Thanks are due to Dr Robert Corbeil for help with statistical analyses. Finally, we wish to recognize the many students who collect and maintain crabs and help out in so many ways.