Sodium extrusion by the sea-water-acclimated fiddler crab Uca pugilator: comparison with other marine crustacea and marine teleost fish

Although the basic pattern of NaCl balance by marine teleost fish has been intensively investigated (see reviews by Maetz, 1971, 1973, 1974; Maetz & Bomancin, 1975; Potts, 1975), the mechanisms of NaCl balance of hyporegulating marine Crustacea have been relatively little studied. Croghan (1958a) showed that Artemia salina (hereafter referred to as Artemia) maintains water balance in sea water by ingesting the medium as do teleost fish (Maetz, 1974) and this has now been confirmed for Artemia (Smith, 1969b) and demonstrated in Metapenaeus bennettae (Dall, 1967), Uca pugilator (Hannan & Evans, 1973) and Parartemia zietziana (Geddes, 1975b). Since the antennal excretory gland of marine Crustacea is unable to produce a hyper-osmotic urine (Lockwood, 1967; Prosser, 1973), it appears that extrarenal mechanisms exist to extrude NaCl to offset the sum of the oral influx and any net diffusional gain across the permeable surfaces of the animal. The site of this extrarenal NaCl extrusion in hyporegulating Crustacea has not been precisely determined. Croghan (1958b) proposed that NaCl extrusion is across the branchiae of Artemia, and Geddes (1975 b) has recently come to the same conclusion while studying ionic regulation in Parartemia zietziana. In support of this hypothesis, Copeland (1967) described mitochondria-rich cells in the metepipodite segments of the branchiae of Artemia. However, Dall (1967) has suggested that the gut is the site of NaCl extrusion by Metapenaeus bennettae, since the rate of efflux of ²²Na from the cephalothorax actually decreases after M. bennettae is salt-loaded.

Little is known about the actual mechanisms for the excretion of NaCl from hyporegulating Crustacea. Potts & Parry (1964) proposed that the prawn Palaemonetes varions, when acclimated to sea water, extrudes Na electrogenically. The resulting internal electronegativity maintains blood Cl concentration in electrochemical equilibrium with sea water despite the concentration gradient favouring net influx of Cl. Croghan (1958b, c) proposed that both Na and Cl are excreted by the branchiae of Artemia and that much of the Na and Cl efflux is via a Na/Na and Cl/Cl exchange diffusion mechanism which provides for no net extrusion of these ions. This proposition was supported by Croghan’s finding (1958c) that the efflux of ²²Na falls rapidly after Artemia is transferred from sea water to distilled water. Croghan (1958 c) also showed that substitution of K for Na in the sea-water medium results in a net loss of haemolymph Na from Artemia. This would seem to indicate that some Na efflux is via a Na/K exchange system and more recent determinations show that Artemia homogenates contain significant quantities of an ATPase which is activated by Na and K, and is sensitive to ouabain, and that the concentration of the enzyme increases when Artemia is raised in hypersaline solutions (Augenfeld, 1969). Thuet, Motais & Maetz (1968) supported the concept of Na/Na and Cl/Cl exchange diffusion across the branchiae of Artemia and hypothesized that a specific Na exchange diffusion carrier is present since the stimulation of Na efflux by external Na displays what appears to be Michaelis-Menten saturation kinetics. Smith (1969a,b) proposed that Na movements across Artemia branchiae are largely diffusive and that haemolymph Na is maintained in electrochemical equilibrium with sea water by the internal positivity of the organism (23 mV). In addition, he proposed that the internal positive potential is the result of a large Na permeability relative to Cl permeability. He also showed that the immediate fall in Na efflux after transfer of Artemia to Na-free solutions (choline sea water) could be accounted for by a depolarization of the electrical potential between the haemolymph and the medium to an inside negative potential (– 36 mV). Smith (1969b) also calculated that changes in the electrical potential across a membrane can give rise to changes in passive Na fluxes which display what appear to be Michaelis-Menten saturation kinetics, and he concluded that there is no need to postulate active Na extrusion by Artemia or a Na/Na exchange diffusion system. However, he proposed that Cl extrusion is indeed active and that a considerable proportion of the Cl efflux is through a Cl/Cl exchange diflfusion system.

The intertidal crab Uca pugilator maintains its blood NaCl concentration below that of sea water (Green et al. 1959) and ingests the medium to offset the osmotic loss of water (Hannan & Evans, 1973). The present study deals with the mechanisms of sodium extrusion by this species and compares these mechanisms with those described for Artemia and marine teleost fish.

Specimens of both sexes of Uca pugilator (Bose) (hereafter referred to as Uca) (1– 3 g) were collected in the intertidal region of Biscayne Bay, Florida, and maintained in a 30 gal aquarium containing a crushed coral bottom and with bottom filtration. They were acclimated to natural sea water (Table 1) (maintained by the addition of either distilled water or Instant Ocean Sea Salts) for at least one week before experimentation. The temperature of the acclimation tank and the experimental solutions was 23 ± 1 °C.

All experimental solutions were modifications of Pantin’s formulation for artificial sea water (Pantin, 1962) (Table 1).

Individuals were weighed on a Mettler Model 1200 balance and then homogenized in a known volume of distilled water in a Virtis Model 23 Homogenizer. The Na concentration of an aliquot was determined with an Instrumentation Laboratory Model 143 Flame Photometer with an internal lithium standard. The total body Na of the crab was calculated by correcting for the volume of the homogenate and the weight of the crab. To determine the Na concentration of the haemolymph, samples were withdrawn at the base of a walking leg with a microcapillary tube and immediately diluted with a known volume of 15 mm-Li solution. The Na concentration of the diluted sample was determined on the Flame Photometer. Total body Na is expressed in /tmole/g and haemolymph Na concentration in mm.

A procedure modified from Evans, Carrier & Bogan (1974) was used to determine the electrical potential (transepithelial potential or T.E.P.) between Uca haemolymph and various external solutions. The internal recording bridge (3 M-KCI, 2 % Agar-filled polyethylene, I.D. 0·011 in. tubing) was placed into haemolymph through a 19 gauge needle which had been inserted through the left anterior quarter of the dorsal carapace. The needle was left in place and the site of insertion was coated with silicone grease to avoid leakage of haemolymph and possible shorting of the measuring circuit. The animal was clamped in place with a laboratory clamp on a ring stand and was partially submerged in the appropriate solution so that the legs, mouth, anus and gills were submerged, while the majority of the dorsal carapace and the site of the insertion of the needle were above the water. This type of restraint of the experimental animal prevented short-circuiting of the T.E.P. and facilitated handling of the animal during changes of the experimental media without interfering with leg or respiratory movements. The recording bridge and a similar bridge in the experimental medium were inserted into separate 3 M-KCI, 2% Agar-filled Polyethylene I.D. 0·055 æ. tubing which terminated in separate 3 M-KCl-filled test tubes. The electrical potential between these bridge terminals was measured with matched calomel electrodes connected to either a Bausch and Lomb VOM recording multimeter (10⁶ Ω input impedance) or to a Keithley Model 616 digital multimeter (10¹³ Ω input impedance). No differences were noted between potentials measured with the different multimeters. Electrode asymmetry was measured by placing a 3 M-KCI, 2% agar-filled polyethylene, I.D. 0·055 1°-tube between the two test tubes containing the calomel electrodes. Bridge and tip asymmetry was checked by placing both internal recording bridge terminals in the same sea-water medium. All measured T.E.P.s were corrected for electrode and bridge asymmetry. All potentials are expressed as haemolymph relative to medium.

In order to determine the ‘transport numbers’ of Na, Cl and K across the epithelium of Uca, the T.E.P. was recorded in sea water and then measured again immediately after the external medium had been changed to either ‘50 %’ Na sea water, ‘50 %’ Cl sea water or ‘50 %’ K sea water (see Table 1). The T.E.P. in sea water was rechecked between each experimental solution, and the change in TJE.P. between that measured in the experimental medium and in the immediately preceding sea-water solution was noted. Using the Nemst equation, Δ V_i would be – 20 mV for ‘50%’ Na or K sea water and +20 mV for ‘50%’ Cl sea water, since these solutions were actually 46% of the natural sea-water concentration.

In order to determine the effect of rapid changes in the ionic concentration of the external medium on the sodium efflux, a flow system described by Evans, Mallery & Kravitz (1973) was used. The only modification of this system was that the total volume of the circulating bath was 140 ml. In these experiments the crabs were injected with ²²Na (3– 5 μ Ci per crab, carried in 3– 15 μ I of distilled water), placed into non-radioactive sea water for approximately 15 min, to allow equilibration of the isotope in the haemolymph, and then transferred to the experimental sea-water bath. After the initial period in sea water (5– 10 min), the crab was transferred into one or a series of ion-free or ion-substituted media as indicated in the Results and Discussion. The crabs remained in each of these experimental media for 5– 15 min.

The total body Na content of Uca in sea water (500 mii-Na) is 219 ± 11 μ-mole g⁻¹ (8) (mean ± standard error, number of animals in brackets). The sodium concentration of Uca haemolymph is 454 ±10 mm (11). Thus, Uca is able to maintain its haemolymph Na concentration some 46 mm below that of its environment. These data corroborate earlier work by Green et al. (1959), who showed that U. pugilator maintains its haemolymph at 328 mm-Na when acclimated to sea water containing 397 mm-Na. However, Uca is unable to maintain the substantial Na concentration gradient described for Artemia, which can maintain its haemolymph Na concentration at 172 mm when acclimated to sea water whose Na concentration is 469 mm (Smith, 1969a).

The electrical potential between the haemolymph of Uca and sea water (trans-epithelial potential or T.E.P.) is – 0·7±0·1mV (67). This slightly haemolymphnegative T.E.P. is in sharp contrast to that found in Artemia which maintains its haemolymph 23·4 mV positive to sea water (Smith, 1969 a). Given the blood : sea water Na gradient measured across Uca in the present study, one can calculate (using the Nernst equation) the Na equilibrium potential which could account for the observed Na gradient. This potential would be +2·3 mV. Although the measured T.E.P. is only 3·0 mV displaced from the sodium equilibrium potential, we observed a blood positive electrical potential across Uca in only 3 out of 67 determinations. Therefore on the basis of the inequality of the measured and Nemst Potential (Gutknecht, 1970), it appears that Uca is maintaining its blood Na concentration out of electrochemical equilibrium with sea water. In addition to the slight net gain of sodium down the electrochemical gradient, Uca must also extrude Na which is gained via oral ingestion of the medium. Hannan & Evans (1973) showed that Uca ingests the medium at a rate of 600 μ l 100 g⁻¹ h⁻¹. Since their sea water was also 500 mm-Na, this ingestion represents a net gain of 3 μ-moleNa g⁻¹ h⁻¹ which must be balanced by some type of extrusion mechanism.

In order to estimate the relative ionic permeabilities of the epithelium of Uca, we determined the ‘transport numbers’ for Na, K and Cl by reducing in turn the external concentration of each ion by approximately 50 % and noting the change in the T.E.P. (Fig. 1,Table 2). If our data are compared with similar determinations on Artemia (Smith, 1969a), two discrepancies are readily evident. In Artemia, T_Na is 0·65 whereas T_Cl is only 0·04, which indicates that the epithelium of Artemia is many times more permeable to Na than it is to Cl (Smith, 1969 a). Our data indicate that, contrary to the situation in Artemia, Uca maintains an approximately equal T_Na and T_Cl. In addition; while the sum of T_NaT_C1 and T_K of Artemia is 0·69, the sum of the ‘transport numbers’ for the same ions in our experiments is only 0·20. Croghan et al. (1965) showed that, in theory, the sum of the ‘transport numbers’ for the permeant ions should be 1·0. That this criterion is not met by either our study or Smith’s (1969a) indicates either that other permeant ions play a substantial role in the origin of the potential across Artemia and Uca, or that equation (2) does not define the origin of the T.E.P. across these species. Neither study can differentiate between these alternatives, but it is unlikely that equation (2) strictly defines the membrane potential, because of the constraints of the constant field theory and because it is possible that some electrogenic ionic transport may also be playing a role in the origin of the T.E.P. Thus, the ‘transport mumbers’ of complex epithelia should really be termed ‘apparent transport numbers’ (Croghan et al. 1965) and should be used only as an estimate of relative ionic permeabilities rather than quantitative measurements of actual permeabilities.

Using equation (3) we can estimate the relative permeabilities of Uca epithelium to Na, Cl and K. Thus P_NS/P_CI = 0·97 and P_NA/P_K = 2·25 and P_Na:P_CI: P_K, is therefore 1·0:1·0:0·4. It appears, therefore, that the relative Na and Cl permeabilities of Uca epithelium are distinctly different from those of Artemia. Smith (1969 a) found that in Artemia the ratios are 1·0:0·03:0·6. It should be noted that since approaches zero, our calculated relative P_K must be regarded as a gross approximation. Nevertheless, it appears that, like Artemia (Smith, 1969a), Uca maintains a P_K that is lower than P_Na. Evidence in teleosts suggest that in this group P_K is substantially greater than P_Na (Potts & Eddy, 1973; Kirschner, Greenwald & Sanders, 1974).

Since Na/K exchange is involved in active sodium transport across a variety of tissues (Ussing & Thorn, 1973; Askari, 1974) and in particular hyporegulating marine teleosts (Maetz & Bomancin, 1975), it is of interest to investigate the effect of removal of K from sea water on the efflux of Na from Uca. If Na extrusion by Uca is via a Na/K exchange mechanism then removal of external K should result in a reduction of Na efflux as is found in some species of teleost (Maetz, 1969; Motais & Isaia, 1972; Evans et al. 1973). In experiments with ²²Na-loaded animals (long-term effluxes) transfer into K-free SW after 2 h in sea water increases the rate constant of the unidirectional Na efflux from 0·1510·04 h⁻¹ (8) to 0·37 ± 0·04 h⁻¹ (8). Using the rapid transfer technique we found that within 1–15 min after the transfer of ²²Na-injected Uca to K-free SW (Fig. 2) the unidirectional efflux of Na increases to 602 ±66% (40, range 163%–1666%) of the sea-water efflux.

This ‘K-free sea-water effect’ (stimulation of unidirectional Na efflux) does not at first seem to support the model for Na extrusion mediated via a Na/K exchange system. However, studies on squid axon (Caldwell et al. 1960; Baker et al. 1969; DeWeer, 1974), erythrocytes (Garrahan & Glynn, 1967a, b, c, d,Garrahan & Garay, 1974), frog striated muscle (Keynes, 1966; Sjodin, 1971) and marine teleosts (Maetz, 1969; Motais & Isaia, 1972) indicate that under certain conditions the carrier system mediating Na/K exchange can be transformed into a carrier with a variety of ionic transfer modes such as a reversed mode (K/Na exchange which generates ATP), Na/Na exchange, K/K exchange and uncoupled electrogenic Na transport. The supporting evidence for this model is extensive (for a recent review see Glynn & Karlish, 1975). The proposition that Na/K to Na/Na transformation takes place is based upon the finding that (under certain conditions), in squid axon, erythrocyte and frog muscle, unidirectional Na efflux continues unabated (or actually increases) after removal of K from the ‘trans’ side of the membrane and that this Na efflux is dependent upon ‘trans’ Na and is inhibited by external application of ouabain (see references above). In addition, Maetz (1969) showed that the K-stimulation of Na efflux from the flounder, Platichthys flesus (which is seen immediately after transfer of the sea-water-acclimated fish to K-enriched tap water), is reduced in the presence of Na. Motais & Isaia (1972) found that transfer of the eel Anguilla anguilla to K-free SW does not alter the efflux of Na even though the unidirectional Na efflux in normal sea water is reduced by 16% when ouabain is added. In addition, they showed that ouabain is much more effective in reducing the K-stimulated Na efflux in K-enriched tap water than the Na efflux in sea water and that it has no effect on Na efflux when A. anguilla is placed in K-free SW. They postulated that, under the conditions of normal sea water, unidirectional Na efflux across the teleost gill is via a carrier system which mediates both Na/K and Na/Na exchange. Whereas the former component is ouabain-sensitive, the latter is apparently ouabain-insensitive. When K is removed from sea water, the carrier transforms to the single, ouabain-insensitive Na/Na mode; in Na-free solutions (K-enriched tap water) the carrier mediates only ouabainsensitive Na/K exchange (Motais & Isaia, 1972).

The ‘K-free sea-water effect’ on Na efflux from Uca may therefore be the result of a similar type of Na/K to Na/Na transformation of the carrier system (see appendix for alternative propositions). To test this we examined the sensitivity of the ‘K-free seawater effect’ to external Na. Uca were transferred to K and Na-free SW, and rapid changes in Na efflux and electrical potentials were noted in separate experiments. In K and Na-free SW the electrical potential increases from –0·6±0·2 mV to– 7·9 ± 0·7 mV (15) while the flux falls to 45 ± 3 % (32) of the sea-water efflux (Figs. 3 and 4). Using equation (5) (see Appendix) one can calculate that this measured increase in the internal negativity of Uca would result in a 15% reduction in passive Na efflux, far less than that measured. It is therefore obvious that the ‘K-free sea-water effect’ is indeed dependent upon external Na (exclusive of potential effects) and is presumably the result of a stimulation of Na/Na exchange after the removal of external K. Since the Na efflux in K and Na-free SW is less than 50% of the Na efflux in sea water, it appears that either Na/Na exchange is already taking place in sea-water-acclimated Uca (parallel to Na/K exchange) and is stimulated by the removal of K (and cessation of Na/K exchange), or else a considerable portion of theNa efflux in sea-water-acclimated Uca is running through a Na/K carrier system that is transformed into a Na/Na carrier when external K is removed. One cannot differentiate strictly between these two alternative models but evidence presented below indicates that Na/K exchange is a relatively small component of Na efflux under the conditions of normal sea water. However, it should be noted that both models suppose that in the presence of external K the rate of Na/Na exchange is limited-the maximal rate is only possible when external K is removed. Thus, functionally, K is inhibiting Na/Na exchange under normal conditions.

Although the above data indicate that Na/K to Na/Na transformations are possible in Uca epithelium, they are not direct evidence for a Na/K exchange system itself. If Na/K exchange is one of the components of Na efflux from Uca then re-addition of K after both Na and K have been removed from the external medium should stimulate Na efflux. To test this we transferred Uca to 25 mm-K, Na-free SW (see Table 1) after an initial transfer from sea water to K and Na-free SW. The efflux and potential changes were studied in separate experiments. In 25 mm-K, Na-free SW the electrical potential across Uca is – 6·0 ± 0·9 mV (8 animals) while the Na efflux is 72 ± 7 % (26) of the control, sea-water Na efflux (Fig. 3). Although the slight measured drop in internal negativity could account for minor stimulation (5 %) of Na efflux from the rate seen in K and Na-free SW (see above), it is obvious that the re-addition of K to the external medium stimulates Na efflux (P ⪡ 0·005 compared to K and Na-free SW efflux) much more than can be accounted for by changes in the electrical potential. These data support the proposition that the epithelium of Uca is capable of carrying on Na/K exchange. However, since addition of 25 mm-K (2·5 times sea-water K concentration) results in an Na efflux that is still less than the Na efflux in sea water, it is apparent that Na/K exchange is a relatively minor component of normal Na efflux across Uca epithelium. Thus, one must conclude that the substantial drop in Na efflux seen in K and Na-free SW (see above) is secondary to the cessation of both Na/Na and Na/K exchange and that Na/Na exchange does take place in Uca epithelium even in the presence of external K.

K-stimulated Na efflux has also been described in teleosts (Maetz, 1969; Motais & Isaia, 1972; Evans et al. 1973; Potts & Eddy, 1973; Greenwald, Kirschner & Sanders, 1974); however, only Evans et al. (1974) were able to show that this stimulation was exclusive of changes in the electrical potential. In addition, Thuet et al. (1968) showed that the sodium efflux from Artemia in 1 M-KCI is 93 % of the efflux in sea water.

It should be noted that since our data show that Na/Na exchange is one component of Na flux across Uca epithelium, our calculation of a P_Na of 2·45 ×10⁻⁸ cm/s (see above) is obviously an overestimate and that Uca is therefore even less permeable to Na than proposed. The true P_Na is difficult to estimate since we have not measured the Na influx in K and Na-free SW; it is probably of the order of 1·5 – 2·0 ×10⁻⁸ cm/s since the efflux declines by 30 % (exclusive of electrical potential) when Uca is placed in K and Na-free SW (when Na/K and Na/K exchanges cease).

Although our data support the proposition that both Na/K and Na/Na exchange are taking place in Uca epithelium when the animal is in sea water, the carrier system must have a much greater affinity for K than for Na since Na/K exchange is apparently taking place in the presence of a relatively huge concentration of the potentially competing ion, Na (500 mm-Na vs. 10 mm-K). To test this proposition directly, in experiments parallel to those just described, Uca were transferred from sea water to K and Na-free SW and then to K-free, 25 mm-Na SW (Table 1). In the latter solution, the potential decreases from – 7 ·9 mV to – 5·6 ± 0·9 mV (7) while the Na efflux actually declines slightly from 46 % of the sea-water efflux to 42 ± 3 % (6) of the sea-water efflux (Fig. 4). Thus, 25 mm-Na does not stimulate Na efflux whereas 25 mm-K produces a significant stimulation. One can only conclude that the carrier has a much greater affinity for external K than Na. The addition of 25 mm-Na to K and Na-free SW depolarizes the electrical potential to the same degree as addition of 25 mm-K to K and Na-free SW. These data indicate that P_NB approximates P_K, rather than being greater than P_K (section 3).

(8) The effect of ouabain on the ‘K-free sea-water effect’

Since Na/Na exchange is ouabain-sensitive in squid giant axons (Baker et al. 1969), erythrocytes (Garrahan & Glynn, 1967a) and frog striated muscle (Sjodin, 1971) but ouabain-insensitive in teleosts (Motais & Isaia, 1972) we investigated the sensitivity to ouabain of the ‘K-free sea-water effect’ of Uca. Using the long-term efflux technique (Fig. 5) with ²²Na-injected crabs and the rapid transfer technique we could find no diminution of the ‘K-free sea-water effect’ when concentrations of ouabain from 10⁻⁴ M to 10⁻³ M were added to the external medium. It appears then that either Na/K activated ATPase is not involved in the ‘K-free sea-water effect’ or that the enzyme has become ouabain-insensitive during the transformation to Na/Na exchange.

Our data indicate that unlike Artemia (Smith, 1969 a, b) and marine teleosts (Potts & Eddy, 1973; Kirschner, Greenwald & Sanders, 1974), Uca maintains a relatively similar permeability to Na and Cl across its epithelium. The small potential across Uca is a result of this lack of differential permeability as well as the absence of significant electrogenic active transport of either ion. Like some teleosts (Evans et al. 1973, 1974) but unlike Artemia (Smith, 1969a, i) and other teleosts (Kirschner et al. 1974), Uca is faced with a net diffusional gain, down their respective electrochemical gradients, of both Na and Cl, which must be extruded along with ingested NaCl transported across the gut. Our data indicate that this extrusion of Na is via a Na/K exchange carrier. In addition, a substantial portion of the unidirectional Na efflux from Uca is coupled to Na influx in a Na/Na exchange diffusion system. There are indications that similar Na/K and Na/Na exchanges can take place (under certain circumstances) in squid giant axons, erythrocytes, and frog muscle cells (Glynn & Karlish, 1975); in these cases it appears that Na/K exchange is the normal mode but that the ouabain-sensitive carrier transforms into a carrier mediating Na/Na exchange if no K is present. Although we have implied that a similar type of carrier transformation could account for the ‘K-free sea-water effect’ that we have described for Uca, in fact, we have no direct evidence for this proposition. It could also be proposed that the Na/K and Na/Na exchanges that we have described are via parallel, functionally linked, but separate carrier systems. However, it is important to note that in either case Na/Na exchange is limited by the presence of Na/K exchange and is therefore substantially stimulated when K is removed from the sea water.

Despite our inability to differentiate between the same or parallel carrier systems for Na/K and Na/Na exchange across Uca epithelium, it is interesting to propose a model for each alternative:

If the two ionic exchange pathways are running through the same carrier then we must propose that the presence of a K ion at a ‘potassium site’ on the carrier molecule somehow inhibits more than one ‘Na site’ so that, when K is removed, not only is the former ‘K site’ transformed into a ‘Na site’, but many formerly shielded ‘Na sites’ also become functional. This would account for the substantial stimulation of Na efflux after removal of external K. In this case one would propose that the ‘K-free sea-water effect’ would be ouabain-sensitive.

If the two efflux pathways for Na are parallel, but not identical, then we must propose that somehow Na/K exchange acts as a ‘brake’ on Na/Na exchange. Removal of external K (and therefore cessation of Na/K exchange) would remove the ‘brake’ and allow full expression of the rapid Na/Na exchange. Perhaps in this case, Na/Na exchange would not be ouabain-sensitive. Motais & Isaia (1972) showed that sodium efflux from the teleost, Anguilla anguilla, in potassium-free sea water is insensitive to ouabain, and our data indicate that the ‘K-free sea-water effect’ in Uca is also insensitive to even high concentrations of ouabain (10⁻³ M).

While the ouabain-insensitivity of the Uca ‘K-free sea-water effect’ would seem to support our second model (separate carrier systems for Na/K and Na/Na exchange), one could also propose that transformation of a Na/K carrier into a Na/Na carrier would abolish the carrier’s sensitivity to ouabain. Our data, unfortunately, do not allow us to distinguish between these two possibilities.

It should be noted that if Na/K and Na/Na exchanges are via separate carrier systems then our description of the relative affinities of Na and K for the carrier (see (7) above) must apply to their respective carriers rather than to a single carrier capable of mediating Na/K and Na/Na exchange.

This research was supported by NSF Grants GB 36423 and BMS 75-00091 to D.H.E.

(1) Depolarization of the T.E.P.

Smith (1969 a,b), Potts & Eddy (1973), Evans et al. (1974), House & Maetz (1974), and Kirschner et al. (1974) have shown that changes in the T.E.P. can result in substantial changes in unidirectional Na efflux from Artemia and various species of teleosts. However, our data indicate that, in Uca, the ‘K-free sea-water effect’ is not the result of changes in diffusional loss of Na secondary to depolarization of the T.E.P. We recorded the changes in the T.E.P. during transfers from sea water to K-free SW and found that the potential did not change (– 0 ·4 ±0 ·2 mV in sea water and – 0 ·6± 0 ·1 mV in K-free SW, 17 animals). In fact one can calculate from equation (5) the voltage change necessary to account for a 6-fold increase in unidirectional, passive Na efflux. This amounts to a change in T.E.P. of more than 5 V.

(2) Increase of

To account for a 6-fold increase in Na efflux as a result of an increase in one would have to postulate that the haemolymph concentration increased 6-fold (to approximately 3 M) within 15 minutes after the removal of external K. This seems highly unlikely although the uncoupling of a Na/K exchange system could result in a decline in Na extrusion, which in turn could give rise to a slight increase in blood Na concentrations. However, our data indicate that Na/K exchange is a relatively small component of the unidirectional Na flux.

(3) Increase of P_Na

P_Na would also have to increase 6-fold to account for the ‘K-free sea-water effect’ which we have observed in Uca. Although we have no direct evidence that an increase of P_Na does not take place, the dependence of the ‘K-free sea-water effect’ on external Na supports the proposition of an increase in Na/Na exchange rather than an increase in P_Na.