Effects of heavy metals on osmoregulation in aquatic organisms have recently been reviewed by Bouquegneau & Gilles (1979). Whereas most of the data in the literature deal with the toxic effects of mercurials, only a few reports can be found on the toxicity of essential trace elements such as Cu and Zn. Rainbow trout exposed to lethal (Skidmore, 1970) and sublethal (Watson & Beamish, 1980) concentrations of Zn maintain a relatively constant internal ionic environment. On the other hand Lewis & Lewis (1971) showed, in the channel catfish, a decreased osmolarity of blood serum after treatment with either Cu or Zn, and Katz (1979) reported an increased Na efflux in freshwater teleosts exposed to heavy metals. An increased gill Na, K-ATPase activity was described in the Zn-treated rainbow trout (Watson & Beamish, 1980). Shephard & Simkiss (1978) showed, in the same species, an increase in the gill protein content of fish exposed to Cu and Zn. These data agree with reports from Cu-treated winter flounder (Baker, 1969) and from Zn-treated dogfish (Crespo, Soriano, Sam-pera & Balasch, 1981; Crespo, 1982) each of which shows an increase in the number of the gill chloride cells.
The opercular epithelium of the killifish, Fundulus heteroclitus, is a flat epithelium with a high density of chloride cells (Karnaky & Kinter, 1977) identical in ultrastructure to that of the gill chloride cells (Karnaky, Kinter, Kinter & Stirling, 1976). This preparation has been studied under short-circuit current conditions in a lucite chamber and proposed as an in vitro model for gill osmoregulatory function (Karnaky, Degnan & Zadunaisky, 1977; Karnaky, 1980). The short-circuit current (Isc) across the epithelium is attributed uniquely to the active transport of chloride ions from the blood to the seawater side of the epithelium (Karnaky et al. 1977 ; Degnan, Karnaky & Zadunaisky, 1977). There is a direct correlation between the number of chloride cells and the Isc, demonstrating that these cells are responsible for chloride secretion (Karnaky et al. 1979).
To investigate the effects of Cu and Zn on chloride transport across the opercular epithelium of 100% seawater-adapted Fundulus heteroclitus, electrical parameters were recorded following the addition of small aliquots (20–200 μl), of Cu (CUSO4. 5H2O) or Zn (ZnSO4. 7H2O) in Ringer’s solution either to both mucosal and serosal sides of the preparation (M + S) or to one side only. Procedures for the dissecting and mounting of the operculum, and descriptions of the Lucite chambers and Ringer’s solution are presented in detail elsewhere (Degnan et al. 1977). The epithelia used for the present study reached a steady-state within 30 min and thereafter exhibited an average spontaneous decay of 11 %/30min.
Table 1 shows that the addition of either Cu or Zn (4 × 10−5 M) to both sides of the preparation caused a statistically significant decrease in both short-circuit current (Isc) and transepithelial potential difference (P.d.). Transmural resistance (R), however, remained unaltered following exposure to either heavy metal. Concentrations as low as 10−5 M (M + S) caused a significant inhibition of the Isc(P< 0·05, n = 3). Initial Isc values were recovered after washing (×3) with Ringer’s solution. When heavy metals were added to the mucosal side only, no inhibition of K was detected, even at higher doses (5 × 10−4 M ; n = 6). After addition of 4 × 10−5 M-CU or Zn to the serosal side only, the pattern for Isc inhibition was the same as the one recorded for M + S exposure (Fig. 1). Zn caused a greater inhibition of the Isc (P<0·02) and P.d. (P <0·01) than did Cu (Table 1).
From our results it is apparent that low concentrations of Cu or Zn can affect chloride transport across the opercular epithelium of the seawater-adapted Fundulus heteroclitus. In contrast to our findings for Cu and Zn, Hg inhibits Isc across the killifish opercular epithelium after mucosal as well as serosal exposure (Degnan & Miller, 1980). The absence of effects from the mucosal side argues against the direct action of these heavy metals on the chloride cell apical membrane. Likewise, these data suggest that these heavy metals do not rapidly penetrate the opercular epithelium.
The fact that Cu and Zn are effective only when added to the serosal side suggests that these metals might interact with Na,K-ATPase located on the basolateral membrane of the chloride cell (Karnaky et al. 1976).
To test the effects of Cu and Zn on the Na pump we studied the in vitro inhibition of the specific activity of Na,K-ATPase. 25µl crude gill homogenates (2–3 mg protein/ml) treated with 0·001% Na deoxycholate were initially preincubated for 10 min at 37 °C in a final volume of 0·5 ml [0·4 mM-EGTA (pH 8·1), 10 mM-Na azide, 2·5mM-MgCl2, 104mM-NaCl, 16mM-KCl, 40mM-Tris-HCl (pH 7·2)] with or without heavy metals. The reaction was started by the addition of Mg-Tris-ATP (3 mM-ATP, pH 7·5) and stopped with 2% trichloroacetic acid. The inorganic phosphate liberated by ATP hydrolysis was determined according to the method of Fiske & Subarrow (1925). Na,K-ATPase activity was calculated as the difference between ATPase activity in the presence and absence of 0·3 mM-ouabain. Protein content of the homogenate was determined according to the method of Lowry, Rosebrough, Farr & Randall (1951).
Fig. 2 shows the dose response curves for Zn and Cu. A statistically significant (P<0·02) decrease in Na,K-ATPase specific activity was first detected following exposure to 10−5M-Zn and 10−4M-Cu, which caused inhibitions of 7 % and 20%, respectively. These results suggest that the inhibition of chloride secretion (Isc) across the opercular epithelium of seawater-adapted Fundulus heteroclitus following exposure to 4 × 10−5 M-Cu or Zn is only partially due to an inhibition of the Na pump. Moreover, Cu and Zn dose response curves for the inhibition of enzyme activity do not differ statistically from each other, yet Zn causes a greater inhibition of the Isc and P.d. than does Cu. The striking difference in responses to these heavy metals suggests that the Na pump is not the only mechanism involved in their toxicity.
According.to recent models of chloride cell function (Silva, Solomon, Spokes & Epstein, 1977 ; Ernst, Dodson & Karnaky, 1980) chloride is transported initially into chloride cells from the blood side by a Na-facilitated, neutral-coupled carrier at the basolateral interface. This ‘secondary’ active transport is driven by the low intracellular Na gradient established by the primary active transport of Na out of the cell by basolateral, plasmalemmal Na,K-ATPase. Clearly, Cu and Zn are potent inhibitors of chloride transport across the opercular epithelium. They inhibit Na,K-ATPase and may interact as well with the coupled NaCl carrier. Since the Isc across the killifish opercular epithelium is inhibited by a-adrenergic agents (Degnan et al. 1977; Mendelsohn, Cherksey & Degnan, 1981) Cu and Zn may also act directly on a-adrenergic receptors. Additionally, heavy metals are thought to alter permeability of the cell membrane (Rothstein, 1959; Kinter & Pritchard, 1977). Further work on this in vitro preparation may help elucidate the underlying mechanisms of Cu and Zn toxicity to electrolyte transport processes.
We would like to thank E. Bell for her technical assistance. We would also like to acknowledge L. Garretson, Dr R. Perez and Dr W. P. Dubinsky for their helpful comments. This work was supported by NIH grant GM 29099 to Dr Karnaky and C.I.R.I.T. grant (Generalitat de Catalunya) to Dr S. Crespo.