Roles of Na+ and K+ in alpha-aminoisobutyric acid transport by the marine bacterium Vibrio alginolyticus.

Effects of monovalent cations on alpha-aminoisobutyric acid (AIB) transport were examined in the marine bacterium Vibrio alginolyticus. In K+-containing cells, AIB was actively accumulated only in the presence of Na+, and the addition of K+ had essentially no effect. On the other hand, K+-depleted and Na+-loaded cells required K+ as well as Na+ for the accumulation of AIB against its concentration gradient. The characterization of the roles of Na+ and K+ in AIB transport was performed by manipulation of intra- and extracellular cation compositions. K+ concentration gradient (K+in greater than K+out) was not essential for the Na+-dependent AIB uptake. Na+ extrusion against its concentration gradient in Na+-loaded cells occurred only in the presence of K+(Rb+). Half-maximal stimulations of the Na+ extrusion and AIB uptake by K+ were observed at K+ concentration near apparent Km for K+ transport. Finally, in the presence of the Na+ electrochemical gradient (toward the inside), K+ was not necessary for AIB uptake. From these results, it was concluded that the Na+-dependent AIB uptake is driven by the Na+ electrochemical gradient across the membrane and that K+ is required for AIB uptake only for the generation of the Na+ electrochemical gradient.

halobium membrane vesicles is the electrochemical gradient of Na'. Since these bacteria have adapted to salt-rich environments, it may be more advantageous to use Na' than to use H' as a coupling ion. The recent discovery of a primary Na' pump, halorhodopsin (31-34), suggests that the role of Na+ in energetics is more fundamental in H . halobium than in nonhalophilic bacteria.
K' has also been reported to affect many transport systems (for a review, see Ref. 11) although its mechanism is yet to be clarified and may not be common to all systems. KC indirectly affects active transport systems by changing metabolisms in some cases and the charge-balancing function of K' has been speculated upon in other cases. The requirement for K' is known in some Na+-dependent active transport systems. Uptake of a-aminoisobutyric acid by the marine bacterium A. haloplanktis (35, 36) and uptake of glutamate by E. coli (37) do not function in the absence of K' when K' is depleted from cells by an osmotic shock treatment. Although Thompson and MacLeod (26) once concluded that neither a Na+ nor a K' concentration gradient is necessary for the AIB' uptake by A. haloplanktis, Niven and MacLeod (27) reached the conclusion from recent experiments that the AIB uptake is driven by the electrochemical gradient of Na+. However, the role of K' in this system is stiU ambiguous. Using membrane vesicles isolated from H. halobium, MacDonald et al. (29) have shown that many Na+-dependent amino acid transports are stimulated by K' . The stimulatory effect of K' has been attributed to its function as a counter ion permitting the overall electroneutral extrusion of Na' .
However, in a more recent paper (38), Lanyi et el. have shown that the permeability of membrane vesicles to K' is not sufficiently great to compensate for the loss of intravesicular Na'. Although the data described above indicate the necessity of K' in some transport systems, the role of K' in the active transport is still an open question. In order to characterize the role of K' in transport systems, the following precautions seem to be important. 1) K' depletion must be performed by an isoosmotic treatment. The osmotic treatment, as used in the case of A. haloplanktis, causes a plasmolysis of cells (39), and uptake of solutes is generally dependent on the size of intracellular space. Furthermore, K' uptake causes a deplasmolysis of such cells (39).
Therefore, the use of plasmolyzed cells may make it difficult to distinguish whether the stimulatory effect of K+ is derived Roles of Na+ and K' in A I 3 Transport by K alginolyticus from its osmotic effect or others. 2) So far reported, no membrane vesicles show the active accumulation of K' (38,(40)(41)(42). If the effect of K' involves the K' transport system, it should be examined in whole cells. 3) The technique which allows the manipulation of intracellular cation compositions is useful since the artificial ion gradients can be imposed across the membrane.
The technique which we have recently developed (43) fulf i the points discussed above and using this method the roles of Na' and K' in AIB transport by the marine bacterium Vibrio alginolyticus are presented in this paper.

MATERIALS AND METHODS
Growth of Cells-The marine bacterium V. alginolyticus 138-2 was grown aerobically on a synthetic medium (43) at 37 "C to the late logarithmic phase of growth. The cells were harvested by centrifugation at 4 "C.
Preparation of Cells Loaded with Specified Cations-The technique developed recently (43) was modifled slightly and extended to prepare cells depleted of K' and/or Na' and loaded with various cations. In brief, the harvested cells were treated twice at 25 "C for 10 min with 50 m M diethanolamine-HC1, pH 8.5, containing 0.4 M desired cation(s) as a chloride salt. The cells loaded with the specified cation were washed twice with 50 nm HEPES, pH adjusted to 7.0 as specified, containing 0.4 M of the same salt as those used in the loading. In our previous paper (43), the K' depletion and Na' loading were performed at pH 9.3 and these conditions decreased neither the cellular activity to generate a proton motive force (43) nor the ability to take up AIB? However, for the loading with various alkali metal cations other than Na', pH during the loading was lowered to 8.5 since the former conditions significantly reduced the activity of AIB uptake even after washing the cells with the buffer at pH 7.0. For the loading with choline+, pH was lowered to 8.0. Detailed experimental results and mechanisms involved will be discussed elsewhere? The above modifications gave satisfactory results regarding the depletion of K+ and Na' and the loading with specified cations.
The cation-loaded cells were resuspended in the buffer (50 m M HEPES, 0.4 M salt) at pH 7.0 and kept on ice until use.
Assay of AIE Uptake-Unless otherwise specified, AIB uptake was performed after dilution of cells into 50 pl of 50 nm HEPES-Na, Materials-[3H]AIB and "hac1 were purchased from New England Nuclear. CCCP was obtained from Sigma. HEPES was a product of Nakarai Chemical Co. Ltd.

RESULTS
AIB Uptake in K+-containing V. alginolyticus- Fig. 1 represents the uptake of AIB, a nonmetabolizable substrate? examined in the presence of various cations using cells washed with 50 nm HEPES-choline, pH 7.0, containing 0.4 M choline chloride. Intracellular concentrations of K' and Na' in these cells were found to be 395 and 22 mM, respectively. Only Na+ (open circles) could support the uptake among ions tested, Li', frequently known to substitute for Na+ in Na+-dependent transport systems, had no effect (closed triangles). The addition of KC1 at a final concentration of 10 nm had essentially no stimulatory effect (closed circles). Moreover, 0.4 M KCl, choline chloride, or CsCl with or without 10 ~l l~ KC1 did not stimulate the uptake (results not shown). CCCP, a proton conductor, collapsed membrane potential (A*) and ApH under all the conditions shown in this paper and caused a significant inhibition of the Na'-dependent AIB uptake (open triangles). Although the results shown were obtained with the cells preincubated for 5 min, the same results as described above were obtained without the preincubation. The intracellular concentrations of K' and Na' after the preincubation were similar to those determined before the preincubation even in the presence of CCCP. The addition of CCCP after the start of AIB uptake immediately caused the efflux of accumulated AIB. A half-maximal stimulation of the uptake was obtained with about 80 mM of Na' and kinetic examinations revealed that both V, , , and apparent K , for AIB uptake were affected by the concentration of Na'. In the presence of 0.4 M NaC1, V, , and apparent K,,, for AIB uptake were 90 nmol/min/mg of cell protein and 43 prq respectively. AIB Uptake in K'-depleted and Nu+-loaded Cells- Fig. 2 shows the AIB uptake by K'-depleted and Na'-loaded V. alginolyticus that contained about 5 mM K' and 0.4 M Na'. In contrast to the results presented in Fig. 1, the Na+-loaded cells showed little AIB uptake even in the presence of 0.4 M NaCl (open circles). However, the preincubation of the Na+loaded cells with K' (closed circles) or Rb' (triangles) prior to the assay caused a dramatic stimulation of the AIB uptake.
The addition of K' at 0 time to the cells preincubated without K+ also gave a similar stimulation of the uptake. However, in this case, the uptake always occurred after a certain length of lag (see Fig. 7). Neither  In order to check degradation, ['HIAIB was recovered after transport experiments and analyzed on a paper chromatogram using the solvent system 1-butanol/acetic acid/water, 25:4:10. All the ra&activity was found in a position of authentic AIB.  Fig. 1, where the control cells were diluted into 0.4 M NaCl, also imposed the concentration gradients of both cations. Under these conditions, the active uptake of AIB was observed. On the other hand, Na+-loaded cells in the absence of K' could not generate the Na' concentration gradient and the level of K' in such cells was far below the steady state level. Such conditions did not allow the uptake of AIB. These results suggested that K' was required by Na+-loaded cells for AIB uptake to generate the concentration gradients of K' and/or Na+. Since the stimulatory effect of K' has been reported in animal systems (45-47), the effect of K' concentration gradient was next examined.
AIB Uptake in the Presence and Absence of a K' Concentration Gradient-In order to examine the effect of a K' concentration gradient toward the outside, the cells loaded with 0.2 M KC1 and 0.2 M CsCl were prepared as described under "Materials and Methods." Cs' was used as a counterpart of Na' since it had no effect on the Na'-dependent AIB uptake (see below). The uptake of AIB was determined in either 0.2 M KC1 and 0.2 M NaCl or 0.2 M Cscl and 0.2 M NaCl (Fig. 3). The cells were diluted about 1000-fold with the assay buffer to reduce a carryover of K' into the reaction mixtures. The cells diluted into KCl/NaCl (open circles), where no K+ KC1 (0) or 10 m M RbCl (A) was present in the reaction mixture. AIB uptake was started and determined as described in Fig. 1. FIG. 3 (rzght). Effect of K+ concentration gradient on Na'dependent AIB uptake. The cells loaded with K' and Cs' were prepared by the treatment with 50 m M diethanolamine-HC1, pH 8. concentration gradient was present initially, showed even better activity than the cells diluted into CsCl/NaCl (closed circles), where the concentration of intracellular K' was about 1000-fold higher than that of extracellular K' . Under both conditions, a Na' concentration gradient (Na+out > Na+in) was present, and CCCP completely inhibited the uptake (triangles). These results indicate that a K' concentration gradient (K+in > K+& is not essential for the Na+-dependent AIB uptake by V. alginolyticus. Extrusion of Na' by Na+-loaded Cells-Since the Na'electrochemical gradient was a possible candidate for the driving force for the Na'-dependent AIB uptake, movements of Na+ in Na+-loaded cells were examined in the presence and absence of K' in order to elucidate the mechanism of the K' requirement for AIB uptake in Na'-loaded cells (Fig. 4). In these experiments, the Na+-loaded cells equgibrated with "Na+ and kept on ice were transferred to 25 "C and the level of "Na+ retained by the cells was determined. If no additions were made, there occurred a slow decrease in the level of =Na+ (open circles). On the other hand, KC1 (closed circles) or RbCl (triangles) added at a final concentration of 10 mM immediately caused the extrusion of "Na'. The observed decrease in the level of "Na' represented the extrusion of Na' against its concentration gradient, i.e. the generation of the Na' chemical potential. The intracellular concentration of Na' reached a steady state level at about 2 min after the addition of K' and at about 5 min after the addition of Rb'. It should be noted that the steady state levels of Na' obtained with K' and Rb" were the same. Although results are not presented, choline chloride, LiC1, or CsCl added at 10 mM as a final concentration did not cause the extrusion of "Na+.
Effect of K' on the Extrusion of Nu'-Rates of Na' extrusion in the presence of various concentrations of KC were examined (Fig. 5). Because of technical difficulties in obtaining accurate rates, especially in the presence of limited concentration of K', the experimental conditions like those in Fig. 4 could not be used and were modified as follows. 1) The concentration of cells in the assay buffer was reduced and the volume of the assay buffer was increased to avoid a decrease in the external K' concentration due to the accumulation of K' by the cells. 2) In order to reduce the concentration of radioactivity in the assay system, the cells equilibrated "Na' were diluted into nonradioactive 0.4 M choline chloride instead of the transfer of cells to 25 "C without dilution. The level of background due to nonspecific binding to filters under such experimental conditions was much lower than that in Fig. 4.
Under the conditions described above and in Fig. 5, the Na' concentration gradient was imposed and the rate of Na' release from the cells in the absence of K' was faster than that obtained in Fig. 4 (compare open circles in Figs. 4 and 5). Nevertheless, the K'-dependent extrusion of Na' could be observed (Fig. 5). The stimulatory effect of K' exhibited a saturation kinetics when initial rates of Na' extrusion were determined as a function of K' concentration (Fig. 6), and the K' concentration required to produce a half-maximal stimulation is 3.5 k 0.7 mM (Fig. 6, inset) which is consistent with the apparent K , (3. 0 f 0.2 mw) for K' transport determined  . 6 (right). Effect of K+ on the rate of Na+ extrusion. The rates of "Na' extrusion determined as described in Fig. 5 are plotted as a function of K' concentration. The initial rates of Na' extrusion were determined at I min after the addition of KC1 and corrected for K'-independent efflux of Na' (open circles in Fig. 5). The values obtained are presented as a double reciprocal plot in the inset.
in the presence of choline chloride.2 Kinetics of K' Requirement in Nu'-dependent AIB Uptake-The attempts to find the relationship between initial rates of the Na+-dependent AIB uptake in Na'-loaded cells and the concentration of K' were unsuccessful. Once the uptake started, the initial rates were essentially identical over the range of K' concentrations examined. Instead, the K' concentration dependence was clearly observed in the length of lag required for the start of AIB uptake as shown in Fig. 7. The lower that a K' concentration was, the longer a lag became. On the other hand, the linear uptake of AIB after the lag proceeded in the same speed in all cases.5 The rates obtained after the lag were the same as that obtained after a 5-min preincubation with the saturating concentration of K' (closed circles). When reciprocals of the length of lag were plotted as a function of K' concentration, the Michaelis-Menten-type kinetics was obtained as shown in Fig. 8. A halfmaximal effect of K' was obtained with 1.2 & 0.1 mM KC1 (inset) which is consistent with the apparent K, (1.65 2 0.5 mM) for K' transport determined in the presence of NaC1/ choline chloride. The minimum value of a lag time was 2.3 min which was very close to the time required in the presence of a saturating concentration of K' to lower the intracellular Na' to the steady state level (Fig. 4, closed circles). Similarly, in the presence of a saturating concentration of Rb', the Although the transient phase of the increase in the rate of AIB uptake is obscure in Fig. 7, the rate examined in the presence of 0.9 mM Rb' started to increase at about 18 min after the start of assay and then, at about 20 min, became constant which was similar to those shown in the figure.
by guest on March 24, 2020 http://www.jbc.org/ Downloaded from minimum value of a lag time (about 5 min, results not shown) was essentially the same as the time necessary to reach the steady state level (Fig. 4, triangles). As described in Fig. 4, the levels of Na' at the steady state obtained with K' and Tlmo (mlnl Rb' were the same. The correlation between the lag time and the time required to reach the steady state level of Na' was also observed in the presence of 1 m~ K' . Furthermore, the level of Na' at the steady state was approximately 0.1 M which was similar to those obtained with saturating concentrations of K+ and Rb' . These results suggested that the lag period corresponded to the time required to reach the steady state level of Na+ and that the generation of a certain magnitude of Na' chemical potential was necessary for AIB uptake. The fact that the steady state levels of Na' were essentially the  Roles of Nu' and K' in AIB Transport by V. alginolyticus same under the conditions examined may be the reason why the AIB uptake after various lengths of lag proceeded at the identical speed including the case of Rb' (Fig. 2). AIB Uptake in the Presence of Nu' Concentration Gradients-The results described above revealed that the concentration gradient of Na' but not of K' is necessary for AIB uptake and that K' is required only for the generation of a Na' chemical potential. However, since the Na' extrusion is tightly coupled to K' uptake under the conditions employed, it is not clear whether the intracellular accumulation of K' is essential or not. In order to clarify this, the cells loaded with cations other than K' or Na' were prepared and assayed for AIB uptake in the presence of Na+ (Fig. 9). Since these cells contained less than 3 mM intracellular Na' , the dilution of cells into 0.4 M NaCl imposed a large concentration gradient of Na' across the membrane. The intracellular concentration of Na' was found to be less than 0.1 M even after a 10-min incubation of Cs' -or choline+-loaded cells with 0.4 M NaC1, and K' retained by these cells was negligible in all cases.
The uptake determined in the absence of K+ with (2s'loaded cells (Fig. 9A, open circles) was comparable to that in the control cells ( Fig. 1) and to that in the Na+-loaded cells assayed in the presence of K' (Fig. 2). Furthermore, the addition of KC1 had essentially no effect (closed circles). Under both conditions, the uptake was abolished by CCCP (triangles). The Cs'-loaded cells generated no ApH and about -100 mV of A?lr that was completely collapsed by CCCP. The addition of K' was without effect on the magnitude of A* and ApH generated by the &'-loaded cells. It is also noteworthy that no detectable lag was observed in the AIB uptake by the cells. Choline+-loaded cells assayed after a 5-min preincubation in 0.4 M NaCl also accumulated AIB in the absence of K' (Fig. 9B). Similar results were obtained without the preincubation. It should be pointed out that Cs' -or choline'-loaded cells accumulated little K' under similar conditions to those shown in Fig. 9.
The uptake of AIB by Li'-loaded cells in the absence of K' lasted for only a short period (Fig. 9C, closed circles). Although the initial rate determined at 30 s (about 16 m o l / min/mg of protein) was comparable to that determined after the preincubation with K+ (dashed line, 20 nmol/min/mg of protein), the accumulation of AIB stopped within 2 or 3 min. When the Li'-loaded cells were preincubated for 5 min in the absence of KC1 after the dilution into Na' (open circles), the uptake of AIB was significantly reduced and became indistinguishable from that in the Na'-loaded cells assayed in the absence of K+ (triangles). On the contrary, the AIB uptake by the Cs+-loaded cells assayed after the preincubation was similar to that shown in Fig. 9A. These results suggested the possibility that the internal Li' rapidly exchanges with the external Na' and, as a result, the Na' concentration gradient was collapsed within a few minutes. Indeed, the dilution of Li+-loaded cells into Na' or Na'-loaded cells into Li' caused much faster efflux of internal cations than the dilution of these cells into Cs' or choline'. Although results are omitted from the figures, none of these cells showed AIB uptake in the absence of Na' .

DISCUSSION
The results presented in this paper led us to the conclusions that the driving force for AIB uptake is the electrochemical potential gradient of Na' across the membrane and that K' is required only for the generation of Na' chemical potential.
In a previous paper (43), we have reported that the K'depleted and Na'-loaded cells generate a large A\lr (-145 mV, negative inside, at pH 7.0) and only a small ApH (15 mV, alkaline inside, at pH 7.0) in the absence of K' and that by the addition of K' to such cells the A* is partially collapsed to -100 mV with the concomitant generation ofApH (40 mV). Therefore, it may be possible that K' is required for AIB uptake to generate ApH. However, the facts that AIB uptake by Na'-loaded cells assayed at pH 8.0, where no ApH is present (43), still required K" and that Cs'-loaded cells accumulated AIB in the absence of ApH exclude the generation of ApH as the explanation for the K' requirement. Therefore, the strong inhibitory effect of CCCP under all conditions examined (for example, Figs. 1,3, 9A, and 9B), indicates that AIB uptake is coupled to A*.
The examination of Na' extrusion clearly demonstrated the K+ requirement for the active extrusion of Na' , namely the generation of the Na' chemical potential. The stimulatory effect of K' appeared to involve the K+ transport system since a half-maximal stimulation of Na' extrusion was obtained with K' at the concentration near K, for K' transport. When AIB uptake in Na'-loaded cells was examined as a function of K' (Fig. 7), the length of lag was inversely dependent on K' concentration. Since Cs'-loaded cells showed no lag whether in the presence or absence of K' , the intracellular accumulation of K' cannot be the reason for the lag. Instead, the lag is likely to represent the time required for the extrusion of a certain amount of intracellular Na' . The results presented in Figs. 5 and 7 are indicative of the gating effects on AIB uptake brought about by the magnitude of the Na+ chemical potential or by the intracellular concentration of Na' . The effect of Na' chemical potential as an essential factor for AIB uptake can be seen in the case of Li'-loaded cells (Fig. 9C). Since the presence of Li' inside the cells facilitates the equilibration of Na' across the membrane, AIB uptake lasted for only a short period and the preincubation of cells inhibited the uptake significantly. From these points discussed above, it is clear that the Na' chemical potential as well as A?lr is necessary for AIB uptake.
The function of K' as a counter ion for Na' extrusion has been demonstrated in a wide variety of microorganisms (48-52; for a review, see Ref. 53). Such a mechanism maintaining overall electroneutrality is indispensable to the generation of ion gradients. Although the role of K' in the Na'-dependent AIB uptake by V. alginolyticus was clearly characterized in this paper, it is still equivocal whether or not the K' requirement reported in A. haloplanktis (35, 36) can be attributed solely to the function of K' as a counter ion. The K'-dependent deplasmolysis may also play an important role in the Na+dependent AIB uptake by osmotically shocked A. haloplanktis. Lanyi et al. (38) have shown that K' does not serve as a counter ion for Na' extrusion in membrane vesicles isolated from H. halobium. Although it was shown that Kt serves as a counter ion for Na+ extrusion in whole cells of H. halobium (52), further investigations are necessary to reveal the role of K' in Na'-dependent amino acid transports in the membrane vesicles.
Our attempts to detect AIB-dependent uptake of Na+ under various conditions as an evidence for Na+/AIB co-transport were unsuccessful. Such a failure does not exclude the possibility of co-transport since it may be difficult to detect the rather small portion of Na' movement in a high concentration of Na' and with cells possibly having a high capacity to bind Na' .