Efflux of Potassium Induced by Dio-9, a Plasma Membrane ATPase Inhibitor in the Yeast Schizosaccharomyces pomhe*

The active uptake of L-leucine is 50% inhibited by 10 pg of Dio-S/ml in glycolyzing cells of the yeast Schizosaccharomyces pombe in which the respiration has been inhibited by antimycin. Similar inhibition is obsetied in the nuclear respiratory-deficient mutant RD32 which lacks mitochondrial ATPase activity. Therefore the mitochondrial ATPase is not involved in the inhibition by Dio-9 of the cellular uptake of metabolites. On the other hand, the ATPase activities of purified plasma membrane fractions isolated from the wild type or the mutant RD32 were both 50% inhibited by 10 pg of Dio-S/ml. Moreover, in the presence of glucose, Dio-9 induces an efflux of K+ simultaneous to an influx of H+ with a half-time for equilibration of 2 min. Uncouplers like carbonylcyanide m-chlorophenylhydrazone increases the velocity of the potassium and proton movements induced by Dio-9 by reduc- ing the half-time for equilibration to about 20 s. In the absence of glucose, Dio-9 does not induce cation movements unless uncouplers are present. The addition of CaCI, or NaCl decreases the H+ influx without modifying the Dio-9- induced K+ efflux, suggesting that Ca2+ and Na+

The active uptake of L-leucine is 50% inhibited by 10 pg of Dio-S/ml in glycolyzing cells of the yeast Schizosaccharomyces pombe in which the respiration has been inhibited by antimycin.
Similar inhibition is obsetied in the nuclear respiratory-deficient mutant RD32 which lacks mitochondrial ATPase activity. Therefore the mitochondrial ATPase is not involved in the inhibition by Dio-9 of the cellular uptake of metabolites. On the other hand, the ATPase activities of purified plasma membrane fractions isolated from the wild type or the mutant RD32 were both 50% inhibited by 10 pg of Dio-S/ml. Moreover, in the presence of glucose, Dio-9 induces an efflux of K+ simultaneous to an influx of H+ with a half-time for equilibration of 2 min. Uncouplers like carbonylcyanide m-chlorophenylhydrazone increases the velocity of the potassium and proton movements induced by Dio-9 by reducing the half-time for equilibration to about 20 s. In the absence of glucose, Dio-9 does not induce cation movements unless uncouplers are present. The addition of CaCI, or NaCl decreases the H+ influx without modifying the Dio-9induced K+ efflux, suggesting that Ca2+ and Na+ can substitute for H+ to balance the K+ efflux.
For increasing concentrations of Dio-9, the apparent stoichiometry of the K+ and H+ movements varies largely from negative to positive values finally tending toward 1.0 at saturating concentrations of the inhibitor. This variation reflects a dual action of Dio-9 on proton movements which can be discriminated by the effects of 80 mM external KC1 which totally abolishes H+ influx at Dio-9 concentrations higher than 50 pg/ml but allows the expression of H+ efflux at concentrations lower than 10 pg of Dio-g/ml. We conclude that Dio-9 elicits an electrogenic exit of K+ along the concentration gradient of this cation. The high electric membrane potential so generated must be balanced by the uptake of cations. Therefore in ATP-depleted cells, the ejection of K+ induced by Dio-9 can be observed only if a proton-conducting agent is added to the medium to balance the K+ efflux by an H+ influx. In the presence of glucose and Dio-9, protons or other cations such as Ca'+ or Na+ are taken up by the cell to neutralize the electric potential generated by the exit of K+. The Dio-g-induced K+ efflux is not second-* Publication 1277 of the EURATOM Biology Division. $ This work is part of a "Doctorat d'Etat en Sciences Naturelles" thesis submitted to the FacultQ des Sciences d'Orsay in 1975. CNRS Registration AO. 11.238. ary to the inhibition of the plasma membrane ATPase activity since it can be observed under conditions where the plasma membrane ATPase is either nonfunctional (absence of glucose) or not inhibited (low Dio-9 concentration). COB5 was grown in a medium containing 36 g of glycerol, 1 g of glucose, and 20 g of yeast extract (Difco)/liter. In these conditions, the respiration and other biochemical properties of this strain are identical with those of the wild strain (19). The nuclear respiratorydeficient mutant RD32 is unable to grow on nonfermentable carbon source and was grown in a medium containing 58 g of glucose and 20 g of yeast extract/liter.
The media were brought to pH 4.5 with HCl. The cells were harvested in the exponential phase of growth.
For H+ and K+ movements, 5 x lo8 cells were incubated at 25" in 5 ml of 2 mM Tris-citrate, pH 4.5. Proton movements were determined by measuring the extracellular pH with a Radiometer pH combination electrode (GK 2321C) connected to a Watanabe recorder (model MC611).
The scale was calibrated after each assay by addition of known amounts of NaOH. In all cases, the pH variations were lower than 0.05 pH unit. The measurement of L-lU-'4Clleucine uptake was described in a previous article (11). For determinations of K+ movements, aliquots of 500 ~1 of the cell suspension were centrifuged for 15 s in an Eppendorf microfuge.
The supernatant was diluted five times with bidistilled water and the K+ content was determined by flame spectrophotometry.

Isolation
of Plasma Membranes ~ The following procedure for the isolation of plasma membrane from S. pombe was elaborated in our laboratory by Delhez (10). The cells (10 g) suspended in 20 ml of 50 rn~ Tris-acetate, pH 7.5, 250 rnM sucrose, and 10 rnM MgCl, were mixed with 10 g of glass beads (0.5 mm) and treated at maximal speed for 2 min with a refrigerated Braun homogenizer. The supernatant of a centrifugation for 10 min at 3,000 x g was centrifuged at 12,000 x g for 10 min. The pellet was discarded and the supernatant was further centrifuged at 25,000 x g for 20 min. Ten milliliters of the resulting supernatant were layered on 6 ml of sucrose at density 1.17 g/cm:% and centrifuged at 100,000 x g for 3 h. Since the density of the mitochondrial inner membrane fragments is 1.17 g/cm", the mitochondrial membranes did not penetrate into the lower layer. On the other hand, ribosomes and plasma membranes migrated to the bottom of the tube. The pellet and the 2 ml located just above the pellet were collected and suspended in 15 ml of 10 mrvr Tris-acetate, pH 7.5, 1 mM MgCl,, and centrifuged at 130,000 x g for 1 h (Step 1). The pellet was resuspended in 0.5 ml of 10 mM Tris-acetate, pH 7.5, and 1 rnM MgCl, and mixed with 2.5 ml of sucrose at density 1.34 g/ cm". On the top of this suspension, four successive bands of sucrose at densities 1.30, 1.25, 1.20, and 1.15 g/cm' were layered. The gradient was centrifuged at 100,000 x g for 15 h. The main purpose of this flotation discontinuous gradient was to separate the dense ribosomes recovered at the bottom from the plasma membranes which migrated up to density 1.20 g/cm3. The fraction at density 1.20 g/cm3 was collected, washed with 30 ml of buffer, and centrifuged at 130,000 x g for 2 h (Step 2).
Adenosi ne Triphosphatase -The reaction mixture (1 ml) contained 25 rnM Tris-acetate, pH 6.0, or 25 mM Tris-NaOH, pH 9.0, 6 rnM MgCl,, and 5 mM ATP. The reaction was started with about 50 +g of proteins and carried out at 30" for 8 min and stopped by 0. The nonrespiring nuclear petite mutant S. pombe RD32 was used to further demonstrate a nonmitochondrial action of Dio-9. In this mutant, the specific activity of the mitochondrial Dio-g-sensitive ATPase, measured at pH 9.0, in the cell homogenate was less than 5% that of the wild strain grown in the same conditions (20). Fig. 1 shows that the uptake of L-leucine in RD32 cells was 50% inhibited by 10 pg of Dio-S/ml of cell suspension.
In this case, the virtually absent mitochondrial ATPase activity could not be the site of action of Dio-9, neither was it the production of glycolytic ATP which is not affected during the first minutes of incubation with Dio-9 (11). Inhibition ofPlasma Membrane ATPase by Dio-9 -Purified plasma membranes of S. pombe contain an oligomycin-insensitive ATPase with maximum activity at pH 6.0 and virtually no activity at pH 9.0 (10). On the other hand, purified mitochondrial membranes exhibit maximum oligomycin-sensitive ATPase activity at pH 8.5 to 9.0. Both plasma membrane and mitochondrial ATPases are sensitive to Dio-9 (10). In Table I, the specific activities of Dio-g-sensitive ATPase were measured both at pH 6.0 and at pH 9.0 in different steps of the purification of the plasma membranes of glycerol-grown COB5.
The specific activity of Dio-g-sensitive pH 6.0 ATPase increased by a factor of 6.4 during purification (from 0.37 pm01 of inorganic phosphate/min/mg of protein in the homogenate to 2.36 pm01 of inorganic phosphate in the final plasma membrane fraction).
The ratio of the activity of the pH 6.0 ATPase to that of the pH 9.0 ATPase increased from 0.7 in the homogenate to 26.0 in the purified plasma membrane fraction.  The specific activity of Dio-g-sensitive ATPase is the difference between the activities of ATPase in absence and presence of 25 pg of Dio-S/ml.
Step 1 and Step 2 are described under "Materials and Methods." About 100 Kg of proteins in 1 ml were used for the ATPase assav carried out as detailed under "Materials and Methods." FIG. 2. Inhibition by Dio-9 of plasma membrane pH 6.0 ATPase activity i n Schizosaccharomyces pombe COB5 and RD32. The twostep purification of the plasma membranes of both strains and the measurements of ATPase at uH 6.0 were described under "Materials Purification steps Dio-9-sensitive ATPase COB5 RD32 PH 6.0 pH 9.0 pH 6.0 pH 9.0

Homogenate
Step 1 Step 2 pm01 P, x min-' x mg protein+ 0 drial activity measured at pH 9.0, but significant amounts of Dio-g-sensitive ATPase with a specific activity of 0.09 pmol of phosphate/min/mg of protein were observed at pH 6.0. In purified plasma membranes, the specific activity of Dio-9sensitive pH 6.0 ATPase increased by a factor of 10 up to 0.9 pm01 of inorganic phosphatelminlmg of protein (Table I). These results indicate that even in a mutant which has lost the pH 9.0 mitochondrial ATPase activity, a Dio-g-sensitive pH 6.0 plasma membrane ATPase activity was still present.
The plasma membrane ATPase activities of both COB5 and RD32 were inhibited 50% by 10 pg of Dio-S/ml (Fig. 2). It must be noted that, in these in vitro assays, the ratio inhibitor to protein was high (0.1 mg of Dio-9/mg of protein) and that comparisons with concentrations used in vivo for uptake of Lleucine are not possible since the pH conditions are different and since it is not possible to estimate how much Dio-9 enters the cell. It was, however, verified that the concentration of Dio-9 required to inhibit the pH 6.0 ATPase activity was a linear function of the absolute concentration of the inhibitor and was independent of the inhibitor to protein ratio within the range of 10 to 250 pg of protein/ml.
Protons and Potassium Movements Induced by Dio-9-Mitchell's chemiosmotic hypothesis proposes that the osmotic work required for the active transport in bacteria is carried out at the expense of an electrochemical gradient generated by a plasma membrane proton-translocating ATPase (17). According to this hypothesis, the inhibition of the plasma membrane ATPase by Dio-9 should be accompanied by the interruption of proton extrusion. In S. pombe COB5 cells where ATP was limited by the absence of exogenous glycolytic substrates and by the presence of antimycin, a respiratory inhibitor, a slow inward movement of protons was observed (Fig. 3A). The external medium was more acidic (pH 4.5) than the interior of the cells which we evaluated to be at pH 6.2 according to the method of Riemersma and Alsbach (3). The protons therefore were moving slowly, down their concentration gradient, reflecting a very limited permeability of the cell membrane to protons. Similar movements have already been described in starved cells of Saccharomyces cerevisiae (3). The addition of 40 mM glucose induced glycolysis as well as an active extrusion of protons into the external medium, against their concentration gradient (Fig. 3A). Subsequent addition of Dio-9 (16 pg/ml) not only stopped the proton extrusion, as expected, but elicited an inward movement of these protons, simultaneously to an exit of K+, both ions moving along their concentration gradients (Fig. 3A). The time courses of potassium and proton on H+ and K+ movements across the cellular membrane of Schizosaccharomyces pombe COB5. Glycerol-grown cells of S. pombe COB5 were harvested in exponential phase of growth and transferred at the density of 10" cells/ml (equivalent to about 1 mg of protein/ml) into a glass vial containing 5 ml of 2 rnM Tris adjusted at pH 4.5 with citric acid. The cells were incubated for 2 min in presence of 10 /LM antimycin.
Proton and potassium movemenk were initiated at 25" with 40 rnM glucose. Dio-9 and carbonylcyanide m-chlorophenylhydrazone, diluted in ethanol, were added at the final concentrations of 16 pg/ml and 50 ELM, respectively. Proton and potassium movements were expressed in microequivalents per ml of cell suspension. movements were hyperbolic as shown by the linear plots of the reciprocal of the variations of the concentrations of protons (A[H+l) or potassium (AlK+]) in the external medium versus the reciprocal of the time after the addition of Dio-9 (Fig. 4). As determined graphically, the half-time for the Dio-g-induced equilibration across the cell membrane was 132 s for H+ influx and 120 s for K+ outflux. These results raise the questions whether these movements were related to the inhibition of the plasma membrane ATPase and which ion was primarily moved by Dio-9.
Effect of Uncouplers -The addition of uncouplers such as carbonylcyanide m-chlorophenylhydrazone (or 2,4-dinitrophe-no11 prior to Dio-9 greatly increased the rate of both cation movements induced by the antibiotic (Fig. 3C) and reduced the half-time for equilibration of H+ and K+ across the cell membrane to about 20 s. Dio-9 has thus an action distinct from that of an uncoupler. In the absence of Dio-9, carbonylcyanide mchlorophenylhydrazone induced a much slower K+ efflux and H+ influx with a half-time for equilibration across the cell membrane of about 20 min (Fig. 3B). Similar K+ efflux induced by proton-conducting agents have already been described in S. cerevisiae (24). These results indicate that in the absence of uncouplers, the rate of potassium efflux is limited by the permeability of the cell membrane to protons. In other terms, Dio-9 does not make the cell membrane freely permeable to protons but seems to induce a leakage of K+ balanced by the uptake of protons.
Specificity of Dio-9 toward Cations-If indeed the influx of to that observed in the absence of CaCl, (Fig. 3A) which was 0.39 peq x 10" cells-' x 3 mini. However, the proton influx was reduced to 0.12 peq x lox cells-' x 3 min-' (compared to 0.30 peq x 10R cells' x 3 min') and 0.085 peq x 10" cells-l x 3 min ml of ?a'+ was simultaneously taken up. These data indicate that Ca"+ can substitute, at least partly, for the proton influx. Fig. 5A shows that the addition of 40 mM NaCl after Dio-9 stopped immediately the inward movement of H+ without modification of the K+ efflux. Even though the movements of Na+ were not measured in that experiment it is quite likely that Na+ was taken up instead of H+ in exchange of the K+ efflux.
It can be concluded that the inward movement of cations elicited by Dio-9 is not restricted to H+, but that either monovalent cations such as Na+ or bivalent cations such as Ca'+ can participate for balancing the extrusion of K+. Effect of Glucose -A surprising aspect of this study was that in the absence of glucose, Dio-9 had very little effect on the movements of K+ and H+ (Fig. 6A). However, the addition of a /A Die-9 FIG. 5. Specificity of Dio-9 toward cation influx in Schizosaccharomyces pombe COB5. A, effect of NaCl on the movements of protons and potassium induced by Dio-9. The experiment was carried out as in Fig. 3A except that 40 mM NaCl was added 0.8 min after the addition of Dio-9. B, uptake of Ca'+ and effects of CaCl, on the movements of protons and potassium induced by Dio-9. Glycerolgrown cells of S. pombe COB5 harvested in exponential phase of growth were transferred into 2 rnM Tris-citrate buffer, pH 4.5, 10 pM antimycin, and 40 rnM glucose. Then 1 rnM VaCl, (0.094 mCi/ml, 10" cpm) was added 4 min after the addition of glucose and 2.5 min before the addition of 20 PM Dio-9. The movements of H+ and K+ were measured as described under "Materials and Methods." The measurements of calcium uptake are described elsewhere (25). Glycerol-grown cells of S. pombe COB5 harvested in exponential phase of growth were transferred into 2 mM Tris-citrate buffer, pH 4.5, in the presence of 10 pM antimycin. When indicated, 65 WM carbonylcyanide m-chlorophenylhydrazone (B) or 80 mM raffinose (C) or 40 rnM 2-n-deoxyglucose (D) and Dio-9 (22 pg/ ml final concentration) were added.
proton-conducting agent can substitute for glucose and under these conditions, Dio-9 induced a rapid exit of K+ and influx of H+ (Fig. 6B) in exponential phase of growth were transferred into 2 rn~ Tris-citrate buffer, pH 4.5, in the presence of 10 PM antimycin A and 30 rn~ glucose for 5 min before adding Dio-9. The movements of H+ and K+ were measured during the first 3 min after Dio-9 addition.
In Fig. 6A It must be mentioned that no effect of K+ has been observed so far on the S. pombe ATPase in vitro (10). Moreover the ATP content of S. pombe COB5 incubated in the absence of substrates and in the presence of antimycin is very low (11) and no proton ejection is observed (Fig. 6B) suggesting that under these conditions the ATPase activity is negligible. Nevertheless, Dio-9 induces the efflux of cellular K+ in these ATP-depleted cells, provided that uncouplers be present (Fig.  6B). It appears thus that Dio-9 might exert an action on K+ efflux in the absence of appreciable ATPase activity.
In an attempt to dissociate distinct effects of Dio-9 on the plasma membrane ATPase and ionic movements, we have examined the effects of Dio-9 according to its concentration. Fig. 7A  The inhibition by Dio-9 of both L-leucine and plasma membrane ATPase could suggest that plasma membrane ATPase plays a role in the active uptake of nutrients in the yeast S.

pombe. However
Dio-9 is known to inhibit yeast mitochondrial ATPase (28,29) and uptake experiments carried out in the presence of glucose and antimycin, an inhibitor of the respiration, cannot exclude the possibility that mitochondrial ATPase is indirectly involved in active transport across the cell membrane.
Specific mitochondrial ATPase inhibitors, however, such as venturicidin (ll), oligomycin, and aurovertin (data not shown) do not inhibit the cellular uptake of L-leucine. Moreover the uptake of L-leucine is 80% inhibited by 50 pg of Dio-9/ ml of cell suspension in the wild type as well as in the respiratory-deficient mutant RD32 lacking the Dio-g-sensitive mitochondrial ATPase activity measured at pH 9.0. In this mutant, however, the purified plasma membrane possesses a high Diog-sensitive ATPase activity measured at pH 6.0 similar to that of the wild strain.
These results eliminate the possibility that Dio-9 acts on cellular transport by inhibiting the mitochondrial ATPase. They do not demonstrate however the involvement of the plasma membrane ATPase in active cellular transport in yeast.
If Dio-9 inhibits a plasma membrane proton-translocating ATPase in S. pombe, one could expect that the antibiotic stops the active extrusion of protons. This was indeed observed but at external pH 4.5, Dio-9 elicited also a rapid exit of K+ and an entry of H+, down the concentration gradient of both ions, with a half-time for equilibration across the cell membrane of about 2 min. These data raise the questions whether the flux of K+ and H+ is related to the inhibition of the plasma membrane ATPase and what is the first event induced by Dio-9. In ATP-depleted cells (no glucose and presence of antimycin to inhibit the respiration), Dio-9 does not induce significant movements of protons and potassium. However, if an uncoupler which makes the cell membrane freely permeable to H+ is added prior to the introduction of Dio-9 in the incubation medium, a fast K-+ exit is observed.
In the presence of glucose, Dio-9 induces uptake of H+ simultaneously to K+ exit. However, even in the presence of glucose, uncouplers greatly increase the rate of K+ uptake induced by Dio-9. The half-time for equilibration of the cation across the cell membrane