ABSTRACT
Active transport of Tl+ by frog skin was compared with transport of K+ and Rb+. Tl+ was transported actively (using a Na,K-ATPase) by the epithelium and the adrenalin-stimulated glands.
In the epithelium, K+, Tl+ and Rb+ competed for transport in the ratio 1:1 ·7: 0· 9; in the glands the ratio was 1:1:1.
INTRODUCTION
Thallous ions are very similar to potassium ions with respect to charge, hydrated radius and mobility in water. The two ions are also transported in similar fashion by muscle cells (Mullins & Moore, 1960).
In the midgut of the American silkworm, there is active transport of potassium but not of thallium, and the passive flux of thallium is ten times that of potassium (Zerahn & Koefoed, 1979). Rubidium ions also have similar properties to potassium ions, and are transported to a similar extent in many systems (Zerahn, 1980). Transport of Tl+, Rb+ and K+ by the frog skin are compared in this paper.
METHODS
Potential difference, short-circuit current and fluxes were measured in frog skin, using techniques essentially as previously described (Ussing & Zerahn, 1951). Skins were dissected from brown frogs, Rana temporaria, obtained from Robert Stein, Lauingen, W. Germany, and kept in shallow water at 4 °C. Frogs were obtained in October and experiments were carried out between the months of October and March. Ringer’s solution consisted of 115 mM-NaCl, 2 mM-KHCO3 and 1 mM-CaCl2. Thallium was added, as the nitrate, at a concentration of 0·1 mM. This concentration produces only a small inhibition of potential difference and short-circuit current. Rubidium and barium were added as chlorides; Rb at 0·1 mM or 1 M, and barium at 2 or 5 mM. The isotopes 42K, 86Rb and 204Tl were obtained from Risø (Denmark) and did not contain significant amounts of other radioisotopes. Antidiuretic hormone (ADH) in the form of arginine vasotocin was obtained from Calbiochem; ADH as vasopressin from Alfred Benzon, Denmark; adrenalin from Nordisk Droge & Kemikalie; amiloride from Merck, Sharp & Dohme, West Point, PA, and ouabain from Meco Benzon, Denmark.
Experiments were performed with paired half skins, either to compare fluxes in opposite directions, or compare experimental treatment with a control. The isotopes were added to one side of the skin and the samples taken from the other side at the desired time. Plastic tubing was used instead of rubber, because rubber may react with thallium.
When the flux experiments were finished the skin was usually removed, blotted on filter paper and weighed before it was placed in a vial with 1 ml 0· 3 M perchloric acid for measurement of 42K, 2O4T1 or 86Rb. The 42K was measured with a Nal crystal, ‘Selectronic’ amplifiers and scalers and a pulse height analyser, adjusted so that very little 86Rb and no 2O4T1 were counted. K+ was determined by flame photometry (Unicam SP90B). The extracellular space in the skins was assumed to be 50%. Inaccuracy in this estimate is insignificant in comparing the ratio between different isotopes. Later an aliquot was taken for measurement of 2O4T1 and 86Rb. When all three isotopes were used simultaneously the 42K samples of 30 min and 1 h periods were often measured with an error greater than ±2% (due to low count-rates), but accuracy of the later samples was always better than ±2%. The calculated net values for the later samples varied accordingly, from 5 to 10% for 42K samples. The 2O4T1 and 86Rb samples were measured with a Packard Tricarb liquid scintillator with an accuracy greater than 2%.
RESULTS
In the isolated short-circuited frog skin, Tl+ fluxes were as shown in Table 1. The net Tl+ flux was very small compared to the short-circuit current of 700 nequiv Na+ h−1 cm−2.
Effect of antidiuretic hormone
When the skin was treated with 0·04μg ml−1 vasotocin or 0·04 i.u. ml−1 vasopressin the Tl+ flux from inside to outside was increased (Table 2) indicating that the epithelium is involved in the active Tl+ flux across the skin. The skins contained 4·5 ± 0·9 μequiv Tl+ g−1 of cells when Tl+ was taken up from the inside solution, and only 0·045 ±0·008 μequivg−1 when taken from the outside (N = 5). When measuring outward flux a 7 cm2 skin of about 200 mg with 50% extracellular space will contain 0·45 μequiv Tl+. Thus it is clear that the flux of about 0·001– 0·002 μequivh−1 (Table 1), taking 1–2 h before equilibrium is obtained, must involve an appreciable exchange of Tl+ between cells and inside solution.
Effect of Ba2+
Addition of 2 mM-Ba2+ to the inside solution induced a transient inhibition of the active transport of Na+, reaching a minimum in 5-15 min, then increasing over several hours, as previously shown (Natochin & Skulskii, 1971; Nielsen, 1979) and increased the active outward transport of Tl+ by 2 to 3·5 times (Table 3). Uptake of Tl+ by skins was unaffected by barium, being 0·71 ± 0·11 μequivg−1 in 30min for 2mm-Ba2+ experiments and 0·85 ± 0·11 /tequivg−1 for controls (N = 3), (Fig. 1).
To test whether the decreased Isc was due to a decreased K+ permeability (Sperelakis, Schneider & Harris, 1967; Nagel, 1979) K+ flux and K+ content of skins were measured with 42K in the inside solution. Determinations were made after 30 min to ensure that 42K was far from being in equilibrium with the skin K+. With 2mm-Ba2+, there was a decrease of 10–20% in K+ outflux (Table 4). This flux represents an exchange rather than an uptake, since the K+ contents of the skins were not changed significantly. K+ concentration in the cell (with an arbitrary value of 50% for extracellular space) was 22·7 ± 1·4mmoll−1 in the presence of Ba2+ and 22·1 ± 1·0 mmol l−1 in the control.
Dependence of Tl+ transport on Na + transport
Table 5 shows that, in skins treated with 2mm-Ba2+ in the inside solution to stimulate Tl+ flux the addition of 10−6M amiloride to the outside solution produced a fall in the Isc from 100 μA to less than 10 μA in 1–2 min. Tl+ flux was affected more slowly and was significantly decreased after 1-2 h (Table 5). Only the active transport of Tl+ was affected, while the flux from outside to inside seemed constant.
Competition between Tl+, K+ and Rb +
The competition for active transport between Tl+, K+ and Rb+ was studied by adding 42K, 86Rb and 204T1 to the solution bathing the skins in which the Tl+ flux was stimulated by adding 2mm-Ba to the inside solution. All fluxes decreased with the length of time that frogs had been kept in storage; on average Tl+, K+ and Rb+ competed in the ratio 1·7:1:0·9. (Table 6).
Adrenalin
Application of adrenalin to the inside of the short-circuited frog skin causes an active transport of Cl− ion by the glands, amounting to several μequiv h−1 (Koefoed-Johnsen,Levi & Ussing, 1952) and was found to stimulate Tl+, K+ and Rb+ transport about equally (Tables 7 and 8). Competition for transport by the glands is therefore about 1:1:1. 10−6M ouabain decreased the active Tl+ flux from 0·0092 ± 0·0010 to 0·0019 ± 0·0010 μequiv h−1 (N = 3) after 30 min. Amiloride at a concentration of 10−6M on the outside of the skin had no significant effect: Tl+ flux before amiloride treatment was 0·0082 ± 0·0006 μequivh−1 and was 0·0093 ± 0·0002 μequiv h−1 (N = 3) after 30 min of treatment.
DISCUSSION
Active Tl+flux in relation to active Na +flux
Tl+, like K+, is transported actively in the outward direction across the frog skin (Table 1). This transport is increased by the action of antidiuretic hormone (Table 2) indicating that the transport is performed by the epithelium. The flux was about 100 times smaller than the Na+ flux (Table 1) and reduction of the Na+ flux by amiloride had no immediate effect on Tl+ flux (Table 5). However, after 1 h Tl+ flux was decreased, so the active transport of Tl+, K+ and Rb+ is apparently indirectly dependent on Na+ transport.
Effect of Bai+ on Tl+ flux
Tl+ transport was increased (and the Isc was decreased) when the skin was treated with 2mm-Baz+ on the inside (Table 3). Barium reduces the K+ permeability of cell membranes (Sperelakis et al. 1967; Nagel, 1979), which may explain the reduction in Isc, but cannot explain the altered Tl+ flux, since the flux remained high when the Isc recovered. In the presence of Ba2+, the exchange for K+ was from 0 to 20% lower. In addition it was found that Ba2+ increased the flux of a skin pretreated with antidiuretic hormone (ADH) but the hormone had no effect on a 2 mM-Ba2+-treated skin, which suggests that either Ba2+ and ADH increase the permeability of the outside membrane of the skin, with Ba2+ having the strongest effect, or that ADH can have no effect in the presence of Baz+.
An alternative explanation for the effect of Ba2+ is that the reduction in membrane potential, from about −95 to −40 mV (Nagel, 1979) increases the leak of Tl+, K+ and Rb+ to the outside. Both explanations fit with the model given by Koefoed-Johnsen & Ussing (1958) and with the model of Ussing & Windhager (1964).
Competition between K+, Tl + and Rb + in epithelium and skin glands
Competition between Tl+, K+ and Rb+ in the presence of Ba2+ has been shown to be in the ratio 1·7:1: 0·9 in the epithelium (Table 6) and in the ratio 1:1:1 in the glands (Tables 7 and 8). The transport mechanism in the glands is probably a K,Na-ATPase driven system since the transport was inhibited by ouabain.
The Na+ transport of the epithelium is also considered to be driven by a Na,K-ATPase, so the different affinity of the three ions to the transport enzyme is not explained by a difference in the kind of enzyme. In contrast the enzyme used for K+ transport in the midgut of the American silkworm is different, not being inhibited by ouabain. Tl+ is not actively transported and the competition between active K+ and Rb+ flux averages 1·0:0·6 (Zerahn, 1980).
ACKNOWLEDGEMENTS
I thank Professor H. H. Ussing, Dr V. Koefoed-Johnsen and Professor S. O. Andersen for reading and discussing the paper, Dr Leon Pape for correcting the language, Mrs Hanne Capion Nielsen and Mrs Susanne Munk Jensen for excellent technical assistance.