Sodium Gradient-dependent Phosphate Transport in Renal Brush Border Membrane Vesicles EFFECT OF AN INTRAVESICULAR > EXTRAVESICULAR PROTON GRADIENT*

A H+ gradient (intravesicular > extravesicular), in the absence of a Na’ gradient (extravesicular > intra- vesicular) stimulated phosphate uptake by renal brush border membrane vesicles and provided the driving force to effect the transient accumulation of phosphate against its concentration gradient. The H+ gradient- dependent uptake of phosphate had an absolute re- quirement for Na+. The rates of uptake and peak accumulation were functions of the ApH and the concen- tration of H’ in the intravesicular medium. The H’ gradient-energized Na+-phosphate cotransport system was not affected by valinomycin- or carbonyl cyanide p-fluoromethoxyphenylhydrazone-induced ion diffu- sion potentials. Therefore, it was independent of the membrane potential, ie. an electroneutral process. Amiloride, which inhibited the H‘-Na’ exchange reaction and prevented the efflux of H+ from the intravesi- cular medium, enhanced the uptake of phosphate. A model is proposed by which the H+ gradient mediates the uphill transport of phosphate. It is suggested that a similar process may operate in more physiologically intact preparations and may provide one mechanism by which acid-base balance regulates renal phosphate transport.

Previous studies have shown that phosphate reabsorption by the kidney takes place predominantly, although not exclusively, in the proximal tubule (l), is highly dependent on the presence of Na' in the tubular fluid (2), and is transported actively against an electrochemical gradient (3). These physiological results become explicable with the demonstration of a Na'-phosphate cotransport system in the proximal tubule brush border membranes and with the finding that a Na' gradient, extravesicular r intravesicular ([Na'], > [Na'],), can energize the transient concentrative accumulation of phosphate into the membrane vesicles (4)(5)(6)(7)(8). Regulation of the renal reabsorption of phosphate has been found to be mediated by various physiological-pathological factors, including hormones, diet, metabolism, and acid-base status (1,9). The effect of pH on renal phosphate clearance is seemingly conflicting, however. Alkalinization of the fiitrate has been reported to decrease phosphate reabsorption by the tubule and increase phosphate excretion (10). On the other hand, rnicroperfusion studies indicate that an acidic and not an alkaline pH inhibits phosphate transport (11). Investigations of the effect of pH on phosphate uptake by membrane vesicles tend * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18  to support the latter conclusion (4, 8). When the pH of the intravesicular and extravesicular media is increased from 6.0 to 8.5, a 10-fold increase is found in the Na+ gradient-dependent rate of phosphate uptake (8). This enhancement is due to the pH dependency of the transport system per se and to an increase in the proportion of phosphate anions that are in the divalent form relative to those that are in the monovalent form (8). In this paper, the effect of pH on phosphate uptake is examined further and evidence is presented showing that a H+ gradient, [H'], > [H'],, stimulates phosphate transport and, in fact, can provide the driving force for the uphill uptake of phosphate in the absence of a Na' gradient. A preliminary account of part of this work has been presented in abstract form (12).

EXPERIMENTAL PROCEDURES
Rabbit renal brush border membrane vesicles were prepared by the method previously reported (13). In experiments in which the intravesicular medium was varied, the vesicles were preloaded by resuspending and washing the 35,000 X g pellet initially in 300 mM mannitol buffered with 5 mM Hepesl/Tris, pH 7.3, followed by carrying out the remainder of the washing procedure in the described medium (13). The efficacy of the preloading procedure to vary the intraveiscular medium, thereby to establish gradients of Na', K', and H' across the intravesicular and extravesicular media or to equalize the distribution of these ions in the two media, was ascertained previously (8,13,14). The quality of the membrane preparations, evaluated by specific activities of enzyme markers, was the same as reported earlier (13 ,, in the absence of a Na' driving force, also effected the transient concentrative uptake of phosphate. When the intravesicular medium was pH 5.5 and the extravesicular medium was 7.3, the initial rate (the 10-s uptake value was taken to represent an approximation of initial velocity) was increased about 5 times compared to the uptake rate when both media were pH 7.3. Accumulation was maximal in 1 min. Thereafter, the amount of phosphate in the vesicles decreased, indicating efflux. The level of uptake in the presence and absence of a H+ gradient was identical after about 15 min, indicating that equilibrium had been achieved and the average intravesicular volumes of the vesicles in the presence and absence of the pH gradient were the same (16). Fig. 1 also shows that when Na+ was omitted from the incubation media the intravesicular > extravesicular H+ gradient had no effect. Thus, the increased uptake of phosphate due to a H+ gradient had an absolute requirement for Na+. The action of the H' gradient on phosphate uptake was specific for the anion. Na+-D-ghcose cotransport was only slightly affected by the intravesicular > extravesicular H' gradient ( Fig. 1). This minimal effect could be explained by the generation of a small membrane potential (inside negative) concomitant with the movement of H' down its electrochemical gradient, from the intravesicular to extravesicular medium (18). The inside negative membrane potential would enhance the electrogenic Na+-D-glucose cotransport, even in the absence of a Na+ gradient (16). This small hyperpolarization of the membrane would not influence Na+-phosphate cotransport, however, since the Na+-phosphate cotransport system was found to be electroneutral and thus independent of the membrane potential (8). Moreover, in other experiments (not illustrated) in which Na' was added to the media in the form of NaCl rather than Na-gluconate ( Fig. l), the slight H+ gradient stimulation of D-glUCOSe was not observed. Since the brush border membrane was found to be more permeable to C1-than to gluconate- (14), C1-could cross the membrane to compensate for the inside negative potential; thus, the development of a potential-induced uptake of D-glucose would be precluded. On the other hand, the substitution of NaCl for Na-gluconate did not affect the H+ gradient-dependent uptake of phosphate (data not shown).
The H+ gradient-dependent overshoot of phosphate uptake was completely inhibited by monensin (Fig. 2). The ionophore, which increased the electroneutral exchange of H' for Na', presumably dissipated the H+ gradient, thereby obviating the driving force for the concentrative uptake.
Increases in the H' gradient-dependent uptake of phosphate were found to be a function of the ApH. As shown in Fig. 3, both the initial rate of uptake and the accumulation of phosphate at the peak of the overshoot were increased as the intravesicular pH was decreased from 7.3 to 5.0 while the pH of the extravesicular medium was maintained at 7.3. The same correlation with H+ gradient was seen when the pH of the intravesicular medium was kept at 5.5 and the pH of the extravesicular medium was varied from 6.5 to 8.0 (Fig. 4). In addition to the ApH, the actual concentration of H' in the intravesicular medium was of importance to the uptake of phosphate. Table I shows that although the ApH was kept the same at 1.5 pH units, i.e. intravesicular pH of 5.5 and extravesicular pH of 7.0 compared to intravesicular pH of 7.0 and extravesicular pH of 8.5, the greater the internal [H'], the more rapid was the initial rate of phosphate uptake and the greater was the accumulation at 1 min.
The kinetic effect of the H+ gradient is shown in Fig. 5. A K, value of 126 IJM was calculated, when the intravesicular pH was 5.5 and the extravesicular pH was 7.3. This was virtually identical with a K, value of 131 IJM, when the pH of both intravesicular and extravesicular media was 7.3. The V,,, of the phosphate uptake system was greatly increased in the presence of the H' gradient.
Since it was reported previously that Na+ gradient-dependent phosphate uptake, in the absence of a H+ gradient, was an electroneutral cotransport process (8), i.e. was not associated with the net transfer of electrical charge, the question was posed as to whether the same was true for the H+ gradientdependent system. To resolve this question, the effects of imposed membrane potentials were determined. Fig. 6 shows the effect of a valinomycin-induced K+ diffusion potential (interior negative) on the H' gradient-dependent uptake of phosphate in an experiment in which [Na+], = [Na+],. The membrane potential had no significant effect on phosphate uptake. In contrast, as reported previously (13), valinomycin greatly enhanced the overshoot when D-glucose was the transported solute. The ionophore presumably induced the efflux of K+ down its electrochemical gradient with concomitant generation of a membrane potential, interior negative, and, in the case of D-glucose uptake, the development of this potential accelerated the influx into the vesicles of the positively charged Na+ coupled to the transport of the uncharged sugar, even though intially there was no Na+ gradient. Fig. 7 illustrates how a membrane potential, interior negative, generated by a H+ diffusion potential rather than a K' diffusion potential, affected the uptake of phosphate and D-TIME IMlNl TIME IMlNl glucose. In this experiment, H+ diffusion potentials were induced by the mitochondrial uncoupler, FCCP. Again, Hf gradient-dependent Na+-phosphate cotransport was affected slightly, if at all, whereas the electrogenic Na+-D-ghcose co-mM Hepes, adjusted to the indicated pH with Tris. The extravesicular medium was 125 mM mannitol, 100 mM Na-gluconate, 75 mM Hepes/ Tris, pH 7.3, 25 p~ KzH"P04. FIG. 4 (right). The effect of the extravesicular pH on the H+ gradient-dependent uptake of phosphate. The intravesicular medium was the same as given in Fig. 2. The extravesicular media were 125 mM mannitol, 100 mM Na-gluconate, 75 mM Hepes, adjusted to the indicated pH with Tris. The concentration of KsH"P04 was 25 PM. transport was greatly stimulated by the interior negative potential, despite the absence of a Naf gradient. It was important to note that in these experiments the gluconate anion was used throughout. In the presence of this anion, to which the membrane was relatively electrophoretically impermeable, the FCCP-induced membrane potential (exterior positive) would oppose the efflux of H+ down its electrochemical gradient. Thus, in these experiments FCCP would presumably induce the inside negative membrane potential without appreciably dissipating the H+ chemical gradient. However, an alternate hypothesis might suggest that the H' gradient driving force for the concentrative uptake of phosphate was not concomitant with the downhill efflux of H' . To test this possibility, H+ gradient-dependent uptake of phosphate was examined when the movement of H+ from the intravesicular to the extravesicular medium was inhibited by amiloride, a drug previously reported to block H'-Na' exchange in renal brush border membranes (19).
First, it was necessary to show that amiloride would inhibit the exchange reaction when the conditions described here to measure phosphate uptake were employed, namely, the presence of a H' gradient but the absence of a Na' gradient and with a high concentration of Nat in the intravesicular and extravesicular media. Fig. 8 shows that amiloride decreased the initial rate of "Naf uptake about 90%. Thus, this finding indicated that amiloride did block the H'-Na' exchange reaction when driven by a Hf gradient.
Next, the H' gradient-dependent uptake of phosphate was examined in the presence of amiloride. As shown in Fig. 9A, uptake was markedly enhanced. This stimulation by amiloride was dependent on the presence of a H' gradient, for, in the absence of H' and Na' gradients, amiloride had no affect.
Thus, the inhibition of the exchange reaction, which presumably prevented efflux of H' (19) and sustained a high concen-FIG. 6. The effect of valinomycingenerated K+ diffusion potential (interior negative) on the H+ gradientdependent uptake of phosphate. The intravesicular medium was 25 mM mannitol, 100 mM Na-gluconate, 50 mM Kgluconate, 75 mM Mes/Tris, pH 5.5. The extravesicular medium contained 125 mM mannitol, 100 mM Na-gluconate, 75 mM Hepes/Tris, pH 7.3, and either valinomycin (VAL) ( tration of intravesicular H+, augmented the uptake of phosphate. This finding would favor the hypothesis that the H+ gradient-dependent phosphate uptake was not obligatorily coupled to the efflux of H+. For comparison, the effect of amiloride on the transport of D-glUCOSe was determined, using the same incubation conditions as for the measurement of phosphate uptake. The uptake of D-glucose, in the absence of a Na' gradient, was also increased (Fig. 9B). However, the explanation for this action of amiloride on the uptake of the sugar was different from the one proposed for its effect on phosphate uptake. As noted in the experiment described in Fig. 1, the presence of the H+ gradient ([H+]i > [H+],) induced the generation of a small membrane potential (inside negative) concomitant with the electrophoretic efflux of H+ down its electrochemical gradient. Inhibition of the electroneutral H+-Na+ exchange reaction by amiloride would tend to preserve the H+ gradient, thereby resulting in a prolonged hyperpolarization and an enhanced electrogenic uptake of D-glucose. This explanation was supported by the observation that amiloride had little effect on D-ghCOSe uptake when the conductive anion C1-rather than gluconate was in the uptake media (data not illustrated).

DISCUSSION
The present results showed that a H' gradient (H+]i > [H+],), in the absence of a Na+ gradient ([Na+], > [Na+]i), provided the driving force to effect the uptake of phosphate into renal brush border membrane vesicles against its concentration gradient. The rate and peak accumulation of the H' gradient-dependent uptake were functions of both the ApH and the concentration of Hf in the intravesicular medium. The H' gradient-energized Na+-phosphate cotransport system, like the Na+ gradient-dependent phosphate cotransport system (B), was not affected by ionophore-induced K+ and H+ diffusion potentials and, therefore, was an electroneutral process. Moreover, although the H' gradient energized the uphill uptake of phosphate in the absence of membrane potential changes, and in the absence of Na' and other anion gradients, no evidence was obtained that the uptake of phosphate was coupled to the transmembrane flux of H+. In fact, when the efflux of H+ was inhibited and a high intravesicular concentration of H' was maintained, uptake of phosphate was further enhanced. These findings suggested that the H' gradient acted by accelerating the influx of phosphate while keeping the reducing equivalents within the membrane vesicle. This mechanism differed from the Na+-L-glutamate cotransport system in renal brush border membrane vesicles, in which a K+ gradient ([Kf]i > [K'],) energized the uphill uptake of the amino acid but in which a transmembrane flux of K+ was suggested (14).
A hypothetical model by which a H+ gradient ([H'li > .Phosphate Uptake [H+],) effected the uphill transport of phosphate is illustrated schematically in Fig. 10. In the absence of a Na+ gradient, the rate of uptake of phosphate into the membrane vesicle would be dependent, in part, on the relative concentrations of phosphate in the extravesicular and intravesicular media. But, the species of the phosphate anion should be considered, for in isotope experiments with 32P the anionic forms of phosphate were indistinguishable. In an extravesicular medium of pH 7.3, the predominant anion form of phosphate would be HP02'. Additionally, in the presence of both the divalent and monovalent species, HP02-was the probable preferred species for transport (8). Thus, relatively more HP04'-would be taken up by the membrane vesicle than HzP04-. With an intravesicular pH more acidic than the extravesicular pH, HC would convert HP02-to HzP04-. In this way, the extravesicular to intravesicular gradient of HP02-was largely maintained; consequently, HP02-uptake was not appreciably diminished. Intravesicularly, the HzP04-species accumulated, since efflux was considerably slower than influx (8). This resulted in the overshoot, or the transient accumulation of total phosphate in the intravesicular medium.
It is suggested that a similar mechanism may operate in more physiologically intact preparations, even in the presence of a Na' gradient, for any process that results in intracellular acidification will tend to increase the uptake of phosphate from the filtrate of the proximal tubular lumen. Thus, one mechanism by which acid-base balance may regulate renal phosphate transport is implied. In addition, this investigation is consistent with and further strengthens the view that the HP04'-anion is preferentially transported relative to the anion.