Amiloride-Sensitive Sodium Flux and Potentials in Perfused Carcinus Gill Preparations

The gills of marine and brackish-water organisms play a key role in the exchange of osmotically active substances between the environment and the internal ‘milieu’. The large surface area and well developed blood supply enable diffusion of the major ions and small organic molecules, as well as providing active ionic transport. Sodium transport mechanisms in the gills of Crustacea include diffusion, carrier-mediated exchange diffusion and active transport (Shaw, 1961 ; Evans, Cooper & Bogan, 1976; Lucu, 1977). The exchange diffusion mechanisms proposed for Na/Na, Na/H or Na/NH₄ exchanges are, however, still the subject of controversy.

Isolated gill preparations seem to be promising experimental models for the elucidation of sodium and chloride transport mechanisms. Koch (1965) has already demonstrated active Na and Cl absorption in the ligatured posterior gills of the euryhaline crab Eriocheir sinensis. The posterior gills have numerous large mitochondria located in the folded proximal cell membranes (Dehnel, 1974). Sodium fluxes in perfused isolated crustacean gill preparations were experimentally determined by Croghan, Curra & Lockwood (1965). Mantel (1967), King & Schoffeniels (1969) and Pequeux & Gilles (1978, 1981).

Sodium and chloride transport have been intensively studied in numerous tight epithelia such as frog skin (Moreno et al. 1973 ; Benos, Simon, Mandel & Cala, 1976) and colon (Bentley & Smith, 1975) and low resistance leaky epithelia such as rabbit gall bladder (Frederiksen, 1983) and toad urinary bladder (Bentley, 1968; Palmer, 1984).

Evidence from several isolated tissues indicates that amiloride has profound effects on sodium transport through conductive pathways of tight epithelia at concentrations less than 10⁻⁶mol l⁻¹ (Eigler, Ketler & Renner, 1967; Nagel & Dörge, 1970). However, in low resistance leaky epithelia, amiloride affects sodium–proton exchangers at concentrations higher than 10⁻⁶mol1^-1 at high external sodium concentrations.

The gills of decapods consist of a central stem which supports several flattened lamellae. Each lamella consists of sheets of epithelial cells covered by cuticle and separated by a blood space. Epithelia contain a transcellular route by which ions penetrate the apical and basal membranes, and paracellular pathways consisting mainly of tight junctions in series with the lateral spaces between epithelial cells (Cioffi, 1984).

Isolated perfused gill preparations should be investigated as a model for studies of ionic mechanisms. The current study was undertaken to investigate the effects of media hyposmotic and isosmotic to the perfusion solution on sodium fluxes and changes in TBP (transbranchial potentials) of isolated Carcinus gill epithelia. The effects of the specific Na transport inhibitor amiloride on sodium fluxes was examined.

Specimens of Carcinus mediterraneus Csrn. were collected in the Adriatic Sea on the west Istrian coast near Rovinj (Yugoslavia). Crabs for experiments in which the gills were under symmetrical conditions were previously acclimated for a few weeks in plastic aquaria containing 501 of sea water diluted with distilled water (14·5‰ salinity). Another group of crabs for experiments under asymmetrical conditions was kept in running seawater aquaria (38‰ salinity). Crabs were fed twice a week in a system with an internal aeration and filtration.

The posterior branchial gill pairs (gills no. 7 and 8) were severed from crabs of comparable sizes (carapace width 4·9 ± 0·3 cm) for these experiments. Immediately after dissection, the gills were immersed in the appropriate solutions for flushing away the blood from the circulatory system. Perfusion solution was delivered by a perfusion pump (Orion Co., USA) through the afferent blood vessel and collected in a flask from the efferent blood vessel by polyethylene capillary tubes. Before the start of the experiments, gills were fitted with capillary tubes and fixed by a clamp constructed from Plexiglas and covered by neoprene material. The preparation was immersed in 40 ml of bathing solution (constant aeration at 20°C). Sodium and chloride fluxes were measured in successive periods of 10 min. After influx measurements the preparations were washed for 10 min in non-radioactive medium and afterwards the effluxes were measured continuously. The branchial preparations were kept at least 15 min in each medium for equilibration. The flow rate of perfusate was 0·15 ml min⁻¹. For experiments in which gills were under asymmetrical conditions, the bathing solution was ordinary sea water (496mmol Nal⁻¹; pH8·l) or diluted sea water (DSW; 102mmol Nal⁻¹; pH 7·9) and the perfusion solution consisted of (in mmol l⁻¹): Na, 506; K, 10·6; Ca, 11·0; Mg, 57·3: HCO₃, 2·8; Cl, 589·8; SO₄, 30·3. In the experiments under symmetrical conditions, the medium on both sides of the gills was dilute ordinary sea water (450mosmol l⁻¹; 201 mmol Na l⁻¹ ; pH 7·9). A Ca-free solution contained (in mmol l⁻¹) Na, 201·8; K, 4·2; Mg, 22·8; HCO₃, 1·1; Cl, 234·0; choline, 4·4.

Transbranchial potential (TBP) measurements were made with a Keithley Instruments 601 Electrometer, connected through Ag–AgCl reference electrodes to agar bridges (3 % agar in 3 mol l⁻¹ KC1 ; 3 mm diameter).

As a criterion for sodium and chloride fluxes under symmetrical conditions, we selected those preparations which showed a TBP more negative than −4 mV with respect to the bathing medium.

The sodium concentration was measured by conventional flame photometry, and the chlorides were measured by chloride titrator. Osmoconcentrations were measured using the Knauer vapour pressure osmometer modified for microsampling.

Amiloride was obtained from Merck, Sharp & Dohme, USA.

Sodium influxes and transbranchial potentials (TBP) of the isolated perfused gill preparation of Carcinus mediterraneus in sea water (SW; c_in/c_out = 0·98) and diluted sea water (DSW; c_in/c_out = 0-20) were measured. The effect of amiloride on the sodium influx in DSW was also examined (Table 1). Sodium influxes were reduced by 37 % when the gills were transferred from sea water to diluted sea water. Reduction of the sodium influx following dilution of the external medium has also been described in the isolated gill preparation of Carcinus maenas (King & Schoffeniels, 1969). In the present experiments, the transfer to diluted sea water was observed to produce a change in TBP from −0·9 mV to -27·0 mV.

In parallel, we investigated Na and Cl fluxes and TBP, in an external medium isosmotic to that being perfused through the gills (symmetrical conditions). As shown in Table 2, a sodium influx of 758 μmol g⁻¹ h⁻¹ and TBP of −7·6 ± 2·3 mV (inside negative polarity with respect to the outside medium) did not appear to arise solely from passive physico-chemical forces, because in the absence of any chemical gradient between the apical and basal side of the tissues it could be maintained for 6 h or more. The addition of 2·5 ×10⁻⁴ mol l⁻¹ amiloride to the external medium caused a rapid decrease in sodium influx from 758 to 232 μmolg⁻¹ h⁻¹ and TBP from −7·6 to −14·8 mV. Also, reduction of sodium effluxes were observed. However, chloride fluxes were not affected by amiloride. A tendency to partial restoration of the TBP (Fig. 1) was shown upon removal of amiloride.

Isolated perfused gill preparations showed reduced sodium influx after transfer from SW (slightly hyposmotic to the perfusion solution) into DSW (20% diluted sea water). In DSW the sodium influx was affected by externally applied amiloride, but the TBP, which is diffusional in origin, was unaffected (Table 1).

An active transport component could be masked by predominantly diffusive fluxes in these media which are hyposmotic to the perfusion solution.

Sodium and chloride fluxes and effects of amiloride on these fluxes were studied in detail under symmetrical conditions, i.e. with identical media on both sides of the gills. Under these conditions potentials are not diffusional and are produced by the activity of ion pumps in the branchial epithelia. According to our unpublished results, ouabain-sensitive sodium fluxes in Carcinus maenas gill preparations are located on the basal and not apical membrane side (D. Siebers, Č. Lucu & A. Winkler, in preparation). In the control (non-amiloride treated) group, experimentally obtained and calculated flux ratios (J_in/J_out) for Na and Cl are statistically different suggesting active transport processes (Table 2). However, under symmetrical conditions where amiloride (2·5 × 10⁻⁴ mol l⁻¹) was applied apically, the sodium influx fell and TBP was hyperpolarized. In the Ca²⁺-free medium, TBP was less hyperpolarized by amiloride (from −4·4 ± 0·6 to −5·7 ± 0·8 mV) than in the presence of calcium (Fig. 2). Study of toad preparations (Benos, Mandel & Balaban, 1979) has also indicated that the sensitivity of the amiloride response to the sodium influx depends on the presence of external calcium. Although the mechanism whereby calcium affects TBP is still unclear, Ca²⁺ seems to play an important role in ion and water transport across the gills of aquatic organisms (Ando, 1980).

The effect of amiloride on sodium influxes was a concentration-dependent process, with incomplete inhibition at the highest concentration used (2·5 × l0⁻⁴mol l⁻¹). From the variation of inhibition under the various amiloride concentrations in the 201·8 mmol Nal⁻¹ external medium, identical to the perfusion solution, the half-maximal inhibition (ID₅₀) of the sodium influxes by amiloride was approximately at 7 × l0⁻⁵mol l⁻¹ (Fig. 3). Since the amiloride concentration required for inhibition of sodium influxes on the apical membrane side of gills was rather higher than 1 μmol l⁻¹ it seems likely that the molecule responsible for this transport is capable of Na—proton counter transport (Na/H antiport) rather than conductive Na movement through the sodium channels (Kinsella & Aronson, 1981; Erikson & Spring, 1982). Inhibition of the electrogenic system described as transport through the sodium channels requires amiloride concentrations that are at least 1 or 2 orders of magnitude lower. The inhibitory effect of amiloride is exerted at separate and distinct regions of the transporting sites depending on the ionic composition of the external solution (Benos et al. 1976). It appears that in our case sodium transport mechanisms showed a low amiloride affinity, typical of low resistance, leaky epithelia.

In the presence of 2·5 × l0⁻⁴mol l⁻¹ amiloride a large hyperpolarization of TBP occurred (basal side electronegative to the apical side). Furthermore, as shown in Table 2, amiloride inhibits only sodium fluxes and has no effect on chloride fluxes. A similar effect has been observed by Bentley (1968) and Kirschner, Greenwald & Kerstetter (1973). The amiloride-sensitive protein responsible for the Na⁺ exchange mechanism had no effect on Cl⁻ fluxes. This phenomenon might be explained at least partially by separate paths for chloride and sodium through the branchial membrane, or by the passage of chloride by the same carrier without any amiloride interaction.

In the case of electrogenic sodium influxes, the internally more negative TBP which developed after amiloride application would stimulate sodium influxes. On the contrary, sodium influxes were markedly decreased. Therefore, it may be suggested that the effect of amiloride upon the Na influx was not due to the effect upon the TBP. Assumed Na/H exchange appears to be an electroneutral process. In freshwater animals hydrogen ion excretion is coupled with Na exchange through the gill epithelia, to maintain electrical neutrality (Kirschner et al. 1973). Amiloride could inhibit acidification (H⁺ secretion) and hence oppose the movement of protons from the basal side of the membrane (Benos, 1981). A direct interaction of amiloride with proteins responsible for the Na/H exchange was also suggested, by La Belle & Eaton and Paris & Pouyssegur (1983). Salako & Smith (1970), Cuthbert & Shum (1974) and Cuthbert (1981) have reported an effect of amiloride on Na transport in the abdominal skin of Rana temporaria.

In the isolated gill epithelia, amiloride might produce the hyperpolarization of the TBP by reducing Na fluxes with no effect upon Cl flux. If an Na/H exchanger was blocked, intracellular acidification would occur, and pH would decrease (Benos, 1981). However, if the H⁺ permeabilities across the paracellular pathways exceed that for Na, this could also explain the hyperpolarization of the TBP. These hypotheses should be experimentally tested in the future.

The authors are indebted to the Self Management Community of Interest for Scientific Research in Croatia for financial support. The research is also part of programme of the German—Yugoslav scientific cooperation which help is greatly appreciated. We are grateful to Drs R. S. Fisher and S. Lewis for critical reading of the manuscript, Mrs M. Mareëii and Mr B. Jagió for excellent technical assistance.