Cell Sodium-Induced Recruitment of Na+-K+-ATPase Pumps in Rabbit Cortical Collecting Tubules Is Aldosterone-dependent*

The influence of intracellular sodium concentration ([Na+]i) on the number of Na’-K+-ATPase pumps was examined in cortical collecting tubules (CCD) of kidneys from rabbits in different aldosterone conditions. Specific [3H]ouabain binding was measured in isolated CCD with various [Na+]i. Experiments were performed on adrenalectomized rabbits receiving only a substitutive dose of dexamethasone and on adrenalectomized rabbits replete with aldosterone. In aldosterone-replete rabbits, the number of binding sites increased linearly with [Na+]i, from 16 fmol/nl tubular volume at I5 mM Na+i to 39 fmol/nl tubular volume at I40 mM Na+i. Neither actinomycin D (5 MM) nor cycloheximide (10 FM) prevented this [NaQdependent increase. In adrenalectomized rabbits, the number of ouabain-binding sites was reduced and did not increase with [Na+]i. These results are in favor of the presence of a “latent” pool of pumps in CCD, rapidly recruited under [Na+]i influence. Aldosterone appears to be required for the constitution and/or activation of this pool.

The influence of intracellular sodium concentration ([Na+]i) on the number of Na'-K+-ATPase pumps was examined in cortical collecting tubules (CCD) of kidneys from rabbits in different aldosterone conditions. Specific [3H]ouabain binding was measured in isolated CCD with various [Na+]i. Experiments were performed on adrenalectomized rabbits receiving only a substitutive dose of dexamethasone and on adrenalectomized rabbits replete with aldosterone.
In aldosterone-replete rabbits, the number of binding sites increased linearly with [Na+]i, from 16 fmol/nl tubular volume at I5 mM Na+i to 39 fmol/nl tubular volume at I40 mM Na+i. Neither actinomycin D (5 MM) nor cycloheximide (10 FM) prevented this [NaQdependent increase. In adrenalectomized rabbits, the number of ouabain-binding sites was reduced and did not increase with [Na+]i. These results are in favor of the presence of a "latent" pool of pumps in CCD, rapidly recruited under [Na+]i influence.
Aldosterone appears to be required for the constitution and/or activation of this pool.
Na+-K'-ATPase pumps present in the membrane maintain the low sodium, high potassium content of the cell (1). In several tissues, the number of pumps is regulated by hormones (2-6). In addition, it has been shown that intracellular sodium per se could exert a direct effect on this number (7-10). The cortical collecting tubule (CCD)' is an aldosterone-sensitive transporting epithelium (11) which adjusts sodium reabsorption within the kidney. Aldosterone-mediated increase in sodium reabsorption involves a rapid increase in cell sodium entry via apical channels coupled to an increase in Na'-K'-ATPase-dependent basolateral sodium extrusion (12). It has been shown that aldosterone regulates maximum Na'-K'-ATPase activity and the maximum number of ouabain-binding sites (13). Furthermore, aldosterone increases the biosynthesis of mRNA encoding for Na+-K+-ATPase pumps in the A6 renal cell line (14). Since aldosterone modifies apical sodium entry, the question arises whether modulation of the number of basolateral pumps could be secondary to aldosterone-induced changes in cell sodium load (15,16). On the other hand, in view of the fast adaptation of Na+-K+-ATPase activity to very large variations of the sodium load in these cells, it has been proposed that this adaptation involves the recruitment or activation of latent pumps to the membrane (16, 17). * 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 USC. Section 1734 solely to indicate this fact.
In in vitro cell preparations, variations of the sodium load are currently realized by changing the intracellular sodium concentration (10,19). In this study, we established the relationship between intracellular sodium concentration and the number of basolat-era1 pumps, as measured by specific [3H]ouabain binding, in isolated rabbit CCD. The influence of aldosterone on this relationship was studied using animals in various aldosterone conditions. Results show that, in the presence of aldosterone, [Na+]i induces a rapid increase in the number of Na+-K'-ATPase pumps that could be ascribed to the recruitment of pumps from a "latent" pool. In the absence of aldosterone, no sodium-dependent increase is observed, suggesting that this hormone is required for the constitution and/or activation of this pool.

MATERIALS
AND METHODS* Experiments were performed on female New Zealand rabbits. Kidneys were prepared for microdissection of isolated CCD. Five ml of blood was collected from aorta to measure plasma aldosterone concentration (radioimmunoassay, Compagnie ORIS Industrie, Saclay, France).

Microdissection of Isolated
CCD-Microdissection of CCD (see Miniprint) was performed as previously described (20). After incubation of kidney pyramids with collagenase, CCD were dissected at 4 "C in two different solutions depending upon the experiment: a saline solution was used for experiments on sodium-loaded tubules, and a sucrose solution was utilized in experiments on CCD in the absence of a sodium load. CCD were dissected in their "light" portion. loo-150 pieces of CCD (~1 mm length) were obtained in each experiment.
Concentration Dependence of Ouubain Binding-Experiments were done on three adrenalectomized and three sham-operated animals (see Miniprint).
CCD were dissected in sucrose solution (to avoid intracellular sodium accumulation) and transferred into the cavity of a microscope slide containing 50 ~1 of sucrose solution with various concentrations of [3H]ouabain (Amersham Corp., 15-31 Ci/mmol, 4.1 x lo-@ to 5.0 x lo-' M) in the presence or absence of a loo-fold excess of unlabeled ouabain. Slides were covered to prevent evaporation, bubbled with air, and placed on a thermostated plate at 20 "C. Incubation was performed for 60 min (see Miniprint). At the end of the incubation time, tubules were rinsed in choline solution (containing 150 mM choline chloride, 2.5 mM MgSO,, 1.2 mM CaC12, 10 mM HEPES/Tris, pH 7.4) at 4 "C (see Miniprint). After 60 min in this solution, the radioactivity of CCD was counted using a liquid scintillation p-counter (Pharmacia LKB Biotechnology Inc. Model 1217).

Effect of Cell Sodium on pH]Ouabain
Binding-Experiments were performed on six adrenalectomized rabbits supplemented with a physiological dose of dexamethasone and five adrenalectomized rabbits supplemented with both dexamethasone and 3 rg of aldosterone/ day/100 g of body weight (see Miniprint). [3H]Ouabain binding was ' Portions of this paper (including part of "Materials and Methods," Figs. 7-9, and Footnotes 3-6) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press.
After 75 min of incubation, CCD were rinsed in choline solution at 4 "C for 15-30 min (see miniprint), and double isotopic counting (**Na and 3H) was performed.
Since both cell sodium and [3H]ouabain binding were determined in single CCD by double isotopic counting, simultaneous measurements of extracellular sodium contamination ([Na+],) and nonspecific [3H]ouabain binding (NSB) were not possible on the same samples. Thus, for each experiment, [Na+], and NSB were measured on separate samples. [Na+], was determined by incubation of -lo-mm CCD in saline solution in the presence of **Na and the extracellular marker [3H]sorbitol (Du Pont-New England Nuclear; 24 Ci/mmol, 35 pCi). NSB was measured by incubation of =lO-15-mm CCD in saline solution in the presence of 2 x 1O-5 M [3H]ouabain and 2 X 10m3 M unlabeled ouabain. The mean value of [Na+], in all experiments was 21.0 + 3.7 peq/nl tubular volume and that of NSB was 1.1 +-0.2 fmol/ nl tubular volume. For each individual CCD in which the total sodium content and total [3H]ouabain binding were simultaneously measured, the values of NSB and [Na'], determined in the same experiment were subtracted to obtain [Na']; and specific [3H]ouabain binding. The effects of actinomycin D (5 FM) and cycloheximide (10 PM) were tested in three adrenalectomized rabbits receiving both dexamethasone and aldosterone: in two rabbits, either actinomycin D or cycloheximide was added to the incubation medium, and in one supplementary rabbit, both inhibitors were tested on separate tubular samples from the same kidney.
Expression of Rest&-In each experiment, CCD were photographed after rinsing in choline solution at 4 "C using a camera placed on a stereomicroscope to determine the tubular volume of each sample (see Miniprint). Ouabain binding was expressed as femtomoles/nanoliter tubular volume. [Na+], was expressed as picoequivalents/nanoliter tubular volume, i.e. as millimolar. The rational for expressing our results as a function of tubular volume are the following. First, it has been shown that cell volume influences ouabain binding (22); second, it is well-known that the steroid condition modulates cell volume (23,24); third, we observed that, even within the same kidney, large variations in tubular volume of CCD were present. Fig. 1

Specific [3H]ouabain binding is expressed per
Upper, control sham-operated rabbits; lower, adrenalectomized millimeter tubular length and plotted against the tubular volume per (ADX) rabbits. Each point is the mean value f S.E. of 4-38 individual unit length. Each point is the mean value + S.E. of 9-44 CCD. CCD CCD from three sham-operated and three adrenalectomized rabbits. were obtained from six kidneys of adrenalectomized rabbits, a con-CCD were incubated at 20 "C in sucrose solution for 60 min with [3H] dition in which sodium concentration does not affect specific ouabain ouabain in the presence or absence of a IOO-fold excess of unlabeled binding (see "Results" and binding per unit tubular length: ouabain binding per millimeter tubular length under the same experimental condition is highly dependent on tubular volume.

RESULTS
Plasma aldosterone concentration was 564 f 123 pg/ml (n = 10) in sham-operated rabbits. For adrenalectomized rabbits (n = 9), all values of plasma aldosterone concentration were below the limit of sensitivity of the method (~50 pg/ml). The mean value was 675 & 292 pg/ml in adrenalectomized rabbits receiving aldosterone (n = 5).

Concentration
Dependence of Ouabain Binding- Fig.  2 shows the concentration dependence of total and nonspecific binding of [3H]ouabain in sham-operated and adrenalectomized rabbits. In sham-operated rabbits, total binding saturates at -10 fmol/nl tubular volume for 10e5 M ouabain. The apparent Kd was =2 X lOme M. Nonspecific binding was very low and did not saturate up to 5 x 10m5 M ouabain. Similar patterns were observed in adrenalectomized rabbits: the apparent Kd was -3.3 X 10e6 M. However, maximum total ouabain binding (~7.0 fmol/nl tubular volume) was lower than that in sham-operated rabbits. Nonspecific binding did not differ from that obtained in sham-operated rabbits. These results indicate that the number of specific binding sites is decreased by adrenalectomy, whereas the apparent affinity of Na'-K'-ATPase pumps for ouabain is not modified. Mean values f SE. of specific ouabain binding are given at 5 mM Na'; intervals. Each point represents 3-44 individual CCD. Specific ouabain binding was lower than that in aldosterone-replete animals (Fig. 3). No significant increase of specific ouabain binding with [Na'], was apparent (y = 0.03x + 6.39, n = 12, r = 0.59, not significant).

DISCUSSION
In CCD, an aldosterone-sensitive transporting epithelium which precisely adjusts the final reabsorption of sodium in the kidney (ll), aldosterone enhances transepithelial sodium transport by increasing both sodium entry via apical channels and basolateral sodium extrusion via Na'-K'-ATPase pumps (12,25). Since one of the initial effects of aldosterone consists of an increase in sodium entry into cells (12, 26) and since very large and rapid changes in Na+-K+-ATPase activity occur in this epithelium (31), we examined the possibility that the recruitment of latent pumps at the basolateral membrane could be dependent on cell sodium. For this purpose, both [Na+li and specific [3H]ouabain binding were measured in isolated CCD from adrenalectomized rabbits with or without aldosterone infusion.
In the absence of external sodium (incubation in sucrose solution), concentration dependence curves of [3H]ouabain binding show that the N,,,,, of Na+-K'-ATPase is reduced in adrenalectomized rabbits as compared to controls. This result is in accordance with previous reports (13). However, the affinity of Na'-K'-ATPase pumps for ouabain in control animals (&(app) = 2 X 10e6 M) is at variance with that reported by Doucet and Barlet (1.2 X 10m7 M) in rabbit CCD composed of inactive pumps in the membrane or of intracel-(27). Differences in experimental conditions, such as a higher lular pumps warrants further studies. In aldosterone-depleted incubation temperature (37 "C) and the use of vanadate, in rabbits, the AJ,,,,, of specific ouabain binding is reduced, and the study of Doucet and Barlet may be the source of the no [Na+li-dependent recruitment of pumps is observed. This difference in the observed affinity of Na'-K'-ATPase pumps suggests that aldosterone is necessary for the constitution of for ouabain. this latent pool and/or its activation. Whereas adrenalectomy reduced the N,,, of Na'-K'-ATPase pumps, the affinity of basolateral pumps for ouabain was unchanged by adrenalectomy, as evidenced by the similar &(app) values obtained in CCD from sham-operated and adrenalectomized rabbits.

2.
The simultaneous determination of specific [3H]ouabain binding and [Na'], in individual CCD clearly demonstrates that the number of basolateral pumps closely depends on [Na+]i in animals receiving aldosterone.
Cell sodium loading is a method currently used to examine the effects of sodium on Na'-K'-ATPase in in vitro cell preparations (10,19). Such studies are based on the assumption that [Na+]; reflects the sodium load that interacts with the pump. In transporting epithelia, such as CCD, the sodium load may vary within a large range, depending on the sodium delivery at the luminal membrane, as reflected by changes in transepithelial sodium transport (18). In our experiments, the number of pumps increased two to three times between 15 and 140 mM Na+, (Fig. 3). This increase was not dependent on RNA or protein synthesis since neither actinomycin D nor cycloheximide suppresses it (Figs. 4 and 5). The relatively short delay for the [Na+],-dependent increase in the number of pumps, associated with the absence of an effect of inhibitors of both RNA and protein synthesis, strongly supports the notion of a recruitment of pre-existent latent pumps under the influence of [Na'],. Such a notion was recently evoked in view of the increase of ouabain binding after incubation of rat CCD with the ionophore nystatin, even in the presence of actinomycin D (8). Skou, J. C. (1965) Physiol. Reu. 45, 596-617 Barlet-Bas, C., Khadouri, C., Marsy, S., and Doucet, A. (1988) Proc. Natl. Acad. Sci. U. S. A. 65, 1707-1711Grinstein, S., and Erlij, D. (1974 Nature 251, 57-58 Ismail-Beigi, F., Haber, R. S., and Loeb, J. N. (1986) Endocrinol-ogy119,2527-2536 In adrenalectomized animals without aldosterone repletion, specific ouabain binding at 15 mM Na', was lower than in aldosterone-replete animals. This is in accordance with data obtained after incubation in sucrose solution (Fig. 2). In addition, an increase in [Na'], up to 140 mM did not result in a recruitment of basolateral pumps, suggesting that the effect of [Na']i requires the presence of aldosterone.
In conclusion, this study indicates that, in rabbit CCD, a pool of latent pumps is present in the cell. These pumps are rapidly mobilized at the basolateral membrane under the influence of intracellular sodium. Whether the latent pool is 20.