Effect of bradykinin on Na-K-2Cl cotransport and bumetanide binding in aortic endothelial cells.

Simultaneous measurements of potassium influx and binding of [3H]bumetanide were performed in endothelial cells cultured from bovine aortas to determine how bradykinin regulates Na-K-2Cl cotransport. [3H]Bumetanide displayed saturable binding and was displaced by low concentrations of unlabeled bumetanide. All three transported ions were required for binding and high concentrations of chloride inhibited binding, consistent with binding of bumetanide to the second chloride site of the transporter. Scatchard analysis of binding under maximal conditions (100 mM sodium, 30 mM potassium, 30 mM chloride) revealed a single class of binding sites with a binding constant of 112 nM and a density of 22 fmol/cm2 or approximately 122,000 sites/cells. Na-K-2Cl cotransport, measured as bumetanide-sensitive potassium influx, was stimulated 118 +/- 30% by bradykinin (p less than 0.01) at physiologic ion concentrations. Stimulation was inhibited by increased potassium or decreased external chloride concentrations and was not seen in conditions required for maximal binding of bumetanide. Simultaneous measurement of the binding of tracer [3H]bumetanide and its inhibition of potassium influx in medium containing 10 mM potassium and 130 mM chloride revealed a turnover number for the cotransporter of 293 +/- 68 s-1 which increased to 687 +/- 105 s-1 with bradykinin (p less than 0.001). There was no change in cell volume and only a 5.6 mM increase in intracellular sodium concentration associated with this stimulation. Bradykinin also increased the affinity of the cotransporter for bumetanide as indicated by a decrease in the Ki for potassium influx from 464 +/- 46 nM to 219 +/- 19 nM (p less than 0.005). Our results show that [3H]bumetanide can be used to quantitate Na-K-2Cl cotransporter sites in aortic endothelial cells and to determine the mechanism by which cotransport is regulated. The stimulation of cotransport in aortic endothelial cells by bradykinin is due to an increase in the activity of existing transporters rather than to an increase in the number of transporters. This, together with the increased affinity for bumetanide, strongly suggests that a change in cotransporter structure is occurring in response to bradykinin.

and Ehrlich ascites tumor cells (1,2), is now known to occur in a variety of cells. Its function may relate to its ability to mediate net transport of salt across the cell membrane since it is prominent in epithelia that absorb or secrete chloride (3). In other cells, net salt transport via Na-K-2Cl cotransport has been shown to mediate regulatory volume increase (2) and thus may be important in the regulation of cell volume. Despite the ubiquity of Na-K-2Cl cotransport and its prominent role in certain tissues, little is known about its structure and regulation.
Cells enriched in cotransport are often either difficult to study or are not suitable for regulatory studies. One exception to this is aortic endothelial cells, which exhibit a high level of Na-K-2Cl cotransport that is regulated by vasoactive hormones (4,5). Since these cells are easily cultured in large quantities, they provide an excellent opportunity to study the regulation of Na-K-2Cl cotransport. Specific inhibition by low concentrations of bumetanide has proven to be very useful in studying Na-K-2Cl cotransport, not only in the identification of Na-K-BCl-mediated ion fluxes but also in the quantification of cotransporters.
[3H] Bumetanide has been shown to bind specifically and saturably with high affinity to membranes from canine kidney (6) and to intact duck red cells (7), vascular smooth muscle cells (8), colonic adenocarcinoma cells (9), and HeLa cells (10). Recently we (11) and others (12) have demonstrated saturable binding of bumetanide to bovine aortic endothelial cells in culture. In this report we have used simultaneous measurement of potassium influx and [3H]bumetanide binding to determine the turnover number for the Na-K-2Cl cotransporter in these cells and to determine how cotransport is stimulated by bradykinin.
Preliminary accounts of this work have been presented in abstract form (11,13 All other reagents were obtained from Sigma.

RESULTS
Initial bumetanide binding studies were performed in a high potassium, low chloride medium (30 mM potassium, 30 mM chloride) as this was found to maximize binding in duck red cells (7). Preliminary studies indicated that specific binding of tracer [3H]bumetanide (25 nM) was rapid and complete by 20 min at 37 "C, and this incubation time was used for all assays. Specific binding of bumetanide was saturable with increasing concentrations of bumetanide and corresponded closely with the inhibition of potassium influx (Fig. 1A).  (7).
Exposure of endothelial cells to bradykinin at physiologic ion concentrations (Earle's salts) increased bumetanide-sensitive potassium influx 118 * 30% (n = 4, p < 0.05) consistent with previous reports (4,5). However, when these studies were repeated in 30 mM potassium, 30 mM chloride, and 120 mM  sodium to obtain maximal bumetanide binding, there was no stimulation of potassium influx by bradykinin. Subsequent studies were therefore performed in 120 mM sodium, 10 mM potassium, 130 mM chloride, a medium that produced measurable binding without completely inhibiting the effect of bradykinin.
Scatchard analysis of the binding of [3H]bumetanide under these conditions yielded errors large enough to prevent distinction between changes in B,,, and changes in KD.
The effect of bradykinin on cotransporter turnover was therefore determined directly by measuring potassium influx and bumetanide binding simultaneously in the same cells. The data from a typical experiment are shown in Table I (Table II) by the maximum number of binding sites determined by Scatchard analysis of bumetanide binding under maximal binding conditions (Fig. 1B).
Bumetanide-sensitive potassium influx was 75 f 21% higher in the cells exposed to bradykinin, and this was associated with a 134 f 46% increase in cotransporter turnover. A small but significant increase in the specific binding of tracer [3H]bumetanide resulted from exposure of the cells to bradykinin.
An additional finding was that in the presence of bradykinin, tracer [3H]bumetanide inhibited a greater proportion of the total bumetanide-sensitive potassium influx (25 f 3% uerws 11 f 2% in control cells). This apparent increase in the affinity of the cotransporter for bumetanide was confirmed by subsequent measurements of potassium influx over a range of bumetanide concentrations (Fig. 3). A -iBK cells (12). Detailed Scatchard analyses revealed only one class --BK of binding sites for bumetanide and indicated the presence of approximately 122,000 binding sites/cell, in the same range as values reported previously in these cells (12), vascular smooth muscle cells (8), and HeLa cells (lo), but far greater than the number obtained in duck red cells (7). The significantly higher quantity of binding sites (2 X 106) reported in Ehrlich ascites cells (15) was associated with a much lower affinity and may represent binding to a Na-Cl cotransporter rather than the Na-K-2Cl cotransporter.  Because it is lipophilic, bumetanide could enter cells and bind to internal cotransporters that are not available for transporting ions. In previous studies, cells have been lysed after incubation with t3H]bumetanide to release intracellular bumetanide so that only bumetanide bound to the plasma membrane was measured (7,8,12). To what extent intracellular contents are removed by this treatment, however, is unclear. Our results, with binding performed and measured in intact cells, do not indicate the presence of an intracellular pool of cotransporters. Intracellular receptors would be exposed to an ionic milieu that is not influenced directly by extracellular ion concentrations. The absence of specific bumetanide binding in low sodium, low potassium, or low chloride extracellular medium argues against significant binding of bumetanide to intracellular cotransporters. Furthermore, the normally low intracellular sodium concentration should lower the affinity of cotransporters for bumetanide yet there was no evidence on Scatchard analysis for two binding sites with differing affinities for bumetanide. Lastly, our results are comparable to those obtained previously in aortic endothelial cells that were hypotonically lysed prior to measurement of bound ["Hlbumetanide (12).
half-maximal inhibitory concentration of bumetanide (KJ was determined for each experiment by linear interpolation between the points immediately above and below 50% inhibition. The mean K, decreased from 464 f 46 to 219 f 19 nM in the presence of bradykinin (p < 0.005). Although bumetanide and chloride compete for the same site (14), activation of cotransport by external chloride was unaffected by bradykinin (data not shown).
Cell volume and intracellular sodium were measured to determine if changes in these parameters were associated with the stimulation of cotransport by bradykinin (Table III). As shown in the table, there was no significant change in cell volume (measured as cell water) with bradykinin.
There was a small but significant increase in sodium content in cells treated with bradykinin.
This represented an increase in intracellular sodium concentration from 17.5 to 23.1 mM.

DISCUSSION
Aortic endothelial cells in culture exhibited saturable, high affinity binding of [3H]bumetanide with a dissociation constant of 112 nM which corresponded well with the inhibition of Na-K-2Cl cotransport by bumetanide and which is similar to values reported for binding to other cells (7)(8)(9)(10). Further evidence that bumetanide was binding directly to Na-K-2Cl cotransporters was provided by the ionic requirements for binding. All three transported ions were required, and high concentrations of chloride were inhibitory. This indicates that bumetanide is binding to the second chloride site of the transporter as indicated previously by kinetic data (14) and consistent with results obtained by other investigators in kidney membranes (6), duck red cells (7), and endothelial The high affinity and ionic conditions of binding and the linearity of the Scatchard analyses provide strong evidence for 1:l stoichiometric binding of bumetanide to the cotransporter, indicating that binding can be used to quantitate cotransporters.
Such quantification in conjunction with ion transport measurements has yielded turnover numbers for Na-K-2Cl cotransport of 300 s-' in aortic endothelial cells (12), 70 s-' in vascular smooth muscle cells (8), 60 s-' in HT29 colonic adenocarcinoma cells (9), 700 s-l in HeLa cells (lo), and 4000 s-' in duck red cells (7). In none of these studies, however, were ion transport and bumetanide binding measured simultaneously, and only in duck red cells were they measured under identical conditions. Simultaneous measurement of [3H]bumetanide binding and potassium influx in this study yielded a turnover of 293 s-'. Although there can be large errors associated with the method used to determine turnover, the error in determining B,,, from Scatchard analyses can be equally large. The actual error in determining turnover was small, and further evidence for the validity of this method was its close agreement with the value for turnover calculated using the B,,, derived from the Scatchard analyses of binding under maximal conditions, and its close agreement with the value of 300 s-' obtained previously in aortic endothelial cells (12). Our data demonstrate that stimulation of Na-K-2Cl cotransport by bradykinin is associated with an increase in turnover number. Although cotransporter number could not be measured directly, the fact that the increase in turnover exceeded the increase in overall cotransport implies that cotransporter number did not increase and may have actually declined. The latter is also suggested by the fact that affinity for bumetanide doubled yet binding of tracer [3H]bumetanide increased only 65%. One possible explanation of our results could be consistent with translocation of intracellular cotrans-porters that bind bumetanide but cannot transport ions, to the plasma membrane where they can transport ions. The finding that regulation of Na-K-2Cl cotransport can affect its affinity for bumetanide in addition to altering its activity must be taken into account when using the binding of bumetanide to quantitate cotransporters. Unless concentrations of [3H]bumetanide substantially greater than saturating concentrations are used, changes in binding measured with just one concentration of [3H]bumetanide could reflect a change in affinity and not necessarily a change in transporter number. Thus, previous findings that CAMP, cGMP, norepinephrine, phorbol ester, and atriopeptin decreased binding to aortic endothelial cells (12), all determined with a single concentration of t3H]bumetanide, may not necessarily indicate changes in the number of cotransporters. The increase in maximal binding of bumetanide to duck red cells after exposure to hypertonicity or norepinephrine cannot be explained by changes in affinity and indicate an increase in cotransporter number (7). Scatchard analyses of bumetanide binding were performed in HT29 colonic adenocarcinoma cells and the decrease in Na-K-2Cl cotransport in response to phorbol esters was ascribed to a decrease in the number of cotransporters. This result, however, was accompanied by a decrease in turnover number, particularly after prolonged exposure to phorbol esters, suggesting a second mechanism for phorbol ester-induced decrease in cotransport (10).
Although there was a small increase in intracellular sodium after incubation with bradykinin, it is unlikely that the increased turnover of Na-K-2Cl cotransport was due to a change in internal substrate ions. Studies in red cells have not provided evidence for trans effects of substrate ions (16). In red cells the cotransporter also mediates a potassium-potassium exchange that is stimulated by internal sodium (16), but no such exchange has been found in aortic endothelial cells, as indicated by the lack of chloride-dependent potassium influx in the absence of external sodium (4,5), including cells treated with bradykinin (4). Measurement of cell water revealed a small decrease that, although not significant, may be of importance since cell shrinkage is a potent stimulus for Na-K-2Cl cotransport in endothelial cells. ' In summary, our results demonstrate that the stimulation of Na-K-2Cl cotransport by bradykinin occurs through a change in the turnover number of cotransporters which is associated with an alteration in the affinity for bumetanide. In the absence of evidence for substrate effects, this suggests that bradykinin is inducing a structural change in the Na-K-2Cl cotransporter. Whether this is the primary mechanism for regulation of this transporter and how it occurs remain to be determined.