- Email: [email protected]
Evidence for a Physiological Role of NH+4Transport on the Secretory Na+-K+-2Cl−Cotransporter
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
245, 301–306 (1998)
RC988428
Evidence for a Physiological Role of NH4/ Transport on the Secretory Na/-K/-2Cl0 Cotransporter Richard L. Evans*,† and R. James Turner* *Membrane Biology Section, Gene Therapy and Therapeutics Branch, National Institute of Dental Research, National Institutes of Health, Bethesda, Maryland 20892; and †Department of Dental Research, University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave., Box 611, Rochester, New York 14642
Received February 20, 1998
The secretory Na/-K/-2Cl0 cotransporter in salivary acinar cells is responsible for driving the transepithelial Cl0 fluxes that give rise to fluid secretion. We demonstrate that the application of the muscarinic agonist carbachol to rat parotid acini results in an intracellular acid load that can be blocked by bumetanide, a specific inhibitor of the cotransporter. One component of this bumetanide-sensitive acid load is ouabain-sensitive while a second is dependent on the presence of sub-millimolar concentrations of NH/ 4 in our media. Our data indicate that this latter effect arises from / NH/ 4 entry on the cotransporter operating in a Na 0 NH/ -2Cl cotransport mode and that at physiological 4 NH/ 4 levels in the rat (Ç0.1 mM), 10-15% of the acinar Cl0 entry occurs via this route. We suggest that Na/0 NH/ 4 -2Cl cotransport may also play a significant physiological role in other cell types and that this mode of operation of the secretory Na/-K/-2Cl0 cotransporter could account for the currently unexplained presence of this protein in a number of tissues. q 1998 Academic Press Key Words: salivary fluid secretion; Na/-K/-2Cl0 cotransport; exocrine glands; NH/ 4 transport; stimulussecretion coupling.
It is now well established that salt and water movement across a number of secretory and absorptive epithelia is due to net transepithelial Cl0 transport. In many of these tissues the active step in this process, cellular Cl0 entry against its electrochemical gradient, is mediated by Na/-K/-2Cl0 cotransport (1). Recent molecular cloning studies have demonstrated that there are, in fact, distinct secretory and absorptive isoforms of the Na/-K/-2Cl0 cotransporter targeted to the basolateral and apical membranes, respectively, of appropriate epithelial tissues (2). The absorptive isoform has thus far only been found in the kidney where it is responsible for salt reabsorption in the thick ascending limb of Henle’s loop. In contrast, evidence for expres-
sion of the secretory isoform as been found in a number of epithelia and in some non-epithelia as well, e.g., salivary glands, stomach, colon, kidney, heart, brain, lung, trachea and skeletal muscle(3,4). In some locations, such as salivary glands, colon and trachea, this transporter has a well-documented role in salt and water secretion (1, 5). However, its somewhat unexpected appearance in many of these other tissues remains to be clarified. In the present paper we document a previously unappreciated feature of the involvement of the secretory Na/-K/-2Cl0 cotransporter in salivary fluid production. Specifically, we show that, at physiological NH4/ levels (Ç0.1 mM), functionally significant Cl0 transport, and hence fluid secretion, by rat parotid acinar cells can be driven by Na/-NH4/-2Cl0 cotransport, i.e., with NH4/ substituting for K/ on the Na/-K/-2Cl0 cotransporter. This component of Cl0 transport is necessarily accompanied by a significant influx of NH4/ and an associated intracellular acidification. Based on these observations we suggest that the physiologically relevant mode of operation of the secretory cotransporter in some tissues is at least partially as a Na/NH4/-2Cl0 transporter and that its functional role in these cases incorporates the movement of NH4/ and/or H/ in addition to that of Cl0. This expanded view of the Na/-K/-2Cl0 cotransporter may help to account for its currently unaccounted for expression in some of the tissues mentioned above. MATERIALS AND METHODS Solutions. Experiments were carried out in a physiological salt solution (PSS) containing 135 mM NaCl, 5.8 mM KCl, 1.8 mM CaCl2 , 0.8 mM MgSO4 , 0.73 mM NaH2PO4 , 20 mM Hepes (pH 7.4 with NaOH), 11 mM glucose, 0.01 % BSA and either 2 mM glutamine (PSS/Gln) or 2 mM alanine (PSS/Ala). In some experiments a 0 HCO0 3 -replete PSS/Ala (PSS/Ala/HCO3 ), in which 25 mM NaCl was replaced with 25 mM NaHCO3 , was employed. Acinar preparation and measurement of intracellular pH (pHi ). Dispersed acini were prepared from the parotid glands of male Wistar
301
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
rats (Harlan Sprague-Dawley Inc., Indianapolis IN) by collagenase digestion and loaded with the fluorescent pH indicator BCECF (2*,7*bis[2-carboxyethyl]-5[6*]-carboxyfluorescein) as previously described (6). Following loading the acinar preparation was washed twice with PSS/Gln and maintained at 307C until use. For experiments carried out in PSS/Ala or PSS/Ala/HCO30 acini were washed once in PSS/Ala and maintained in PSS/Ala for at least 30 min at 307C before use. Throughout the entire preparation and subsequent experimental periods acini were kept continuously agitated (90 cycles/min) and gassed with 100%O2 or 95%O2/5%CO2 as appropriate. Intracellular BCECF fluorescence was monitored in ratio mode at 377C using a Spex ARCM fluorimeter (Spex Industries, Edison NJ) as previously described (6). At the end of each experimental run acini were lysed with 0.25% Triton X-100 and the BCECF fluorescence ratio of the released dye determined in PSS (pH 7.4). In a separate series of experiments the BCECF signal was calibrated by titrating the lysate with NaOH and/or HCl. The resulting calibration curve was linear with slope 0.263{0.006 pH units/fluorescence ratio unit (nÅ3) over the pH interval 6.45-7.45. For BCECF traces within this pH range (which includes all the results illustrated in the paper) it was assumed that the fluorescence ratio of the intracellular dye at pHi 7.4 would be the same as that of the released dye from the same run and the fluorescence trace was converted to pHi using the above slope of the linear portion of the calibration curve. Using this method the resting pHi of the rat parotid acinar preparation was found to be 7.31 { 0.02 (n Å3 6). In order to confirm the validity of the above method we also determined the resting acinar pHi independently using a recently published ‘‘null technique’’ (7). Briefly, acini were exposed to combinations of a weak base (ammonium) and a weak acid (propionate) at concentrations such that if the acinar pHi were at a certain predetermined ‘‘null pH’’ value the intracellular alkalinization resulting from the former would be exactly balanced by the intracellular acidification produced by the latter (7). Combinations of sodium propionate (held constant at 10 mM) and NH4Cl (varied from 16.21 to 4.07 mM) corresponding to null pH’s from 7.5 to 7.2 were employed in order to bracket the resting pHi . A plot of the intracellular alkalinizations or acidifications resulting from these challenges vs. the corresponding null pH’s was linear (not shown) allowing determination of the resting acinar pHi by interpolation (7.31 { 0.02 in PSS/Gln; n Å 3). The excellent agreement of this value with that obtained above is consistent with the assumption that the fluorescence properties of intracellular BCECF are similar to those of the released dye from Triton X-100 treated cells. Determination of acinar intracellular buffering power. Acini in PSS/Gln were challenged by the addition of NH4Cl (10 or 20 mM) or sodium propionate (10, 20 or 30 mM) to the extracellular medium. The resulting acute intracellular alkalinizations or acidifications were then determined by extrapolation of the associated BCECF traces back to the time of NH/ 4 or propionate addition. The intrinsic intracellular buffering power (bi) calculated from each of these challenges was assigned to the mid point of the associated alkalinization or acidification. All of these determinations of bi were within experimental error of each other (Ç10%) indicating that bi is constant over the pHi interval 7.1 to 7.5. The average value of bi over this pHi range was 27.6{0.8 mmol/liter per pH unit. In HCO0 3 -containing PSS the total intracellular buffering power is the sum of the intrinsic buffering power plus the intracellular CO2 0 buffering power given by bCO2 Å 2.3 [HCO0 3 ]i, where [HCO3 ]i is the intracellular HCO0 3 concentration (8). Ammonium determination. Medium ammonium levels were determined with the Sigma Diagnostic Kit #171-UV. Data analysis and presentation. Reported experimental values are means { S.E.M. for three or more independent determinations performed under the same conditions on different acinar cell preparations. P values õ 0.05 (Student’s t test) were taken to represent statistically significant differences.
RESULTS AND DISCUSSION In situ, the muscarinic agonist acetylcholine is the main fluid secretory stimulus for salivary glands. Past studies have shown that both Na/-K/-2Cl0 cotransport (6) and Na//H/ exchange activities (9,10) in rat parotid acinar cells are rapidly and dramatically upregulated (Ç20- and Ç3-fold, respectively) following stimulation by the acetylcholine analogue carbachol. In addition, measurements of Na/ uptake into carbachol-stimulated acini have demonstrated large components of Na/ entry due to both the cotransporter and the exchanger (11-13). As already mentioned, a major role for the Na/K/-2Cl0 cotransporter in salivary Cl0 secretion is welldocumented (5, 14). There is also good evidence that the Na//H/ exchanger contributes to Cl0 secretion by acting in parallel with a basolateral Cl0/HCO30 exchanger (5, 15) to effect acinar Cl0 entry. Interestingly, however, a significant component of carbachol-stimulated Na/ entry via the Na//H/ exchanger persists in HCO30-free media (12, 13) suggesting additional involvement of this transporter in the muscarinic response. The present studies were begun to further explore the roles and interactions of the Na/-K/-2Cl0 cotransporter and the Na//H/ exchanger during fluid secretion. Fig. 1 summarizes a number of our initial observations. In these experiments the fluorescent pHi indicator BCECF was used to monitor pHi (see Methods) following carbachol (Cch; 10mM) stimulation of rat parotid acini suspended in glutamine-containing PSS (PSS/ Gln; see Methods), a medium we (6, 9, 15-17) and others (10, 13, 18) have employed in a number of earlier studies. As illustrated in the lower trace of Fig. 1A, when otherwise untreated acini are exposed to carbachol they rapidly alkalinize (õ0.1 pH unit). This effect has been noted in several earlier studies (11, 13, 16) and is presumably due to the agonist-induced upregulation of Na//H/ exchange activity mentioned above. Fig. 1A also demonstrates that this rise in pHi is enhanced in the presence of the Na/-K/-2Cl0 cotransporter inhibitor bumetanide (10mM). In these and past experiments (6, 17) we have confirmed that the application of bumetanide in the absence of carbachol is without effect on pHi (data not shown); thus this enhanced alkalinization suggests the presence of a stimulus-induced, bumetanide-sensitive component of intracellular acidification. This effect is confirmed and shown more directly in Fig. 1B where the Na//H/ exchanger has been blocked by the addition of its potent inhibitor DMA (5-(N,N-dimethylamiloride; 20 mM). Under these conditions the acini acidify following the application of carbachol and this acidification is blunted by bumetanide, again consistent with the presence of a stimulus-induced, bumetanide-sensitive intracellular acid load. The results in Figs. 1A and 1B have been quantitated in the upper 4 bars of Fig. 1C (see Figure caption).
302
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Effects of inhibitors of Na/-K/-2Cl0 cotransport, Na//H/ exchange and Na//K/ ATPase on the response of rat parotid acinar pHi to muscarinic stimulation. The results shown in panels A and B are representative pHi traces (see Methods) from BCECF-loaded acini suspended in PSS/Gln and stimulated with 10 mM carbachol (Cch) in the presence or absence of 10 mM bumetanide (bumet; applied 50 s prior to the application of Cch) and/or 20 mM DMA as indicated. These and additional results have been quantitated in panel C by measuring the difference between the pHi found 100 s after the application of carbachol and that observed before carbachol addition (DpHi). In these experiments ouabain was used at 1 mM (applied 60 s before carbachol) and acini in PSS/Ala were prepared as described in Methods. The point labeled ‘PSS/Ala / bumet*’ was obtained with a slightly different protocol. Here acini were incubated with carbachol for 250 s (at which point DpHi was not significantly different from that found after 100 s incubation with carbachol; DpHi was 0.083 { 0.008 250 s after carbachol vs. 0.079 { 0.004 100 s after carbachol) then bumetanide (10 mM) was added and the DpHi from that observed before carbachol addition was determined 100 s later. The average results { S.E.M. of 3 or more independent experiments under each condition are shown.
We reasoned that the above effect of bumetanide might be due to a Na/-K/-2Cl0 cotransporter-induced metabolic acid load due to the increased Na//K/ATPase turnover required to extrude the Na/ entering the cell via the cotransporter during secretion. Consistent with this hypothesis the carbachol-induced intracellular acidification observed in the presence of DMA was blunted by the Na//K/-ATPase inhibitor ouabain (Fig. 1C). However, a significant effect of bumetanide remained even in the presence of ouabain (Fig. 1C), indicating that this explanation could not completely account for the bumetanide-sensitive component of acidification. Moreover, when DMA-treated acini were stimulated with carbachol in the absence of extracellular K/, a maneuver expected to inhibit both Na//K/ATPase and Na/-K/-2Cl0 cotransport, acinar acidification was enhanced (Fig. 1C) rather than blunted. After some consideration of this latter result it occurred to us that the apparent effects of Na/-K/-2Cl0 cotransport inhibition illustrated in Fig. 1C might be accounted for by the presence of NH4/ in our medium. We hypothesized that the bumetanide-sensitive component of intracellular acidification could be due to the
Na/-K/-2Cl0 cotransporter operating in a Na/-NH4/2Cl0 mode, with the NH4/ possibly arising from the spontaneous (19) or acinar-induced breakdown of the glutamine present in PSS/Gln. It is well established that NH4/ can substitute for K/ on the cotransporter (6, 17). The NH4/ entry generated in this way would result in an intracellular acid load because of the dissociation of the accumulated intracellular NH4/ into NH3 plus a H/, allowing the permeant species NH3 to diffuse out of the cell, leaving the H/ behind. This hypothesis can account for the enhanced intracellular acidification observed in the absence of K/ (Fig. 1C) since under these conditions there would be increased NH4/ entry, and thus increased intracellular acidification, owing to the elimination of the competition between K/ and NH4/ for the cotransporter. As an initial test of this hypothesis we measured the alkalinization induced by carbachol in cells incubated in glutamine-free medium (PSS/Ala; see Methods). In these experiments a significantly higher alkalinization was observed than in the presence of glutamine and the effect of bumetanide addition was significantly reduced (Fig. 1C), consistent with the predicted effect of elimi-
303
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
0 0 FIG. 2. Quantitative estimation of the contribution of Na/-NH/ 4 -2Cl cotransport to cellular Cl entry during carbachol-induced fluid secretion. In the experiment illustrated in panel A BCECF-loaded rat parotid acini suspended in PSS/Ala were stimulated with 10 mM carbachol then challenged with 200 mM NH4Cl followed by 10 mM bumetanide as indicated. The protocols for the experiments illustrated in panels B and C were similar except that the medium was PSS/Ala/HCO0 3 and NH4Cl, at the concentrations indicated, was added 180 s before stimulation with carbachol. In panel D the initial rate of alkalinization following bumetanide addition is plotted vs. [NH/ 4 ] for acini suspended in PSS/Ala (open circles) and PSS/Ala/HCO0 3 (closed circles). The averaged results { S.E.M. of 4 or more independent determinations at each [NH/ 4 ] are illustrated.
nation of glutamine-derived NH4/ from the medium. We next determined the NH4/ content of our media directly. These results confirmed the presence of 12 { 3 mM NH4/ in PSS/Gln alone and 24 { 4 mM NH4/ in PSS/Gln plus acini (the experimental conditions of Fig. 1 were employed). Much lower levels of NH4/ (3 { 1 mM) were found in PSS/Ala and there was no detectable increase in the presence of acini (4 { 1 mM). Since NH4/ levels in rat plasma are Ç0.1 mM (20) the above results suggest that a substantial flux of H/equivalents and thus a substantial ion flux could be carried via Na/-NH4/-2Cl0 cotransport at physiological NH4/ concentrations. In order to explore this possibility in more detail we carried out the experiments illustrated in Fig. 2. In trace A of Fig. 2, acini in PSS/Ala were first treated with 10 mM carbachol, then challenged with 200 mM NH4Cl and finally exposed to 10 mM bumetanide, as indicated. This experiment confirms directly that sub-millimolar concentrations of NH4/ can produce a significant intracellular acid load in carbachol-stimulated rat parotid acini. The reversal of this effect by bumetanide is again consistent with the involvement of Na/-NH4/-2Cl0 cotransport. In control experiments (not shown) we have demonstrated
that the application of up to 1 mM NH4Cl to acini before carbachol treatment produces no significant effect on pHi. Thus the effect of NH4/ illustrated in Fig. 2A only occurs after the dramatic upregulation (Ç20-fold) of Na/-K/-2Cl0 cotransport activity resulting from muscarinic stimulation (6). The initial rate of alkalinization observed after bumetanide addition in Fig. 2A provides a measure of the bumetanide-sensitive intracellular acid load present at this concentration of NH4/. In Fig. 2D we summarize the results of a series of experiments of this type where we have determined the magnitude of this acid load as a function of NH4/ concentration in (HCO30-free) PSS/ Ala (open circles). From these data we find that the bumetanide-sensitive, NH4/-dependent intracellular acid load at 0.1 mM NH4/ (i.e., the increment in the bumetanide-sensitive intracellular acid load due to the presence of 0.1 mM NH4/) is Ç0.04 pH units/min. From the known buffering power of the acinar cytoplasm (27.6{0.8 mmol/liter per pH unit; see Methods) and the intracellular water content of the parotid (Ç45%; 21) this value can be converted into a rate of acinar NH4/ entry/gram tissue (Ç0.5 mmoles/g-min). Assuming that this bumetanide-sensitive NH4/ entry occurs via
304
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Na/-NH4/-2Cl0 cotransport, this value can be converted to a rate of acinar Cl0 entry (Ç1 mmole/g-min). As already indicated, salivary fluid secretion is thought to be primarily due to acinar Cl0 secretion driven in large part by Cl0 entry across the basolateral membrane via the Na/-K/-2Cl0 cotransporter. In previous experiments from our laboratory carried out under similar conditions to those presented here we have determined that the rate of sustained Cl0 secretion from carbacholstimulated rat parotid acini is Ç7 mmoles/g-min (14). Hence the above results demonstrate that at physiological NH4/ levels Na/-NH4/-2Cl0 cotransport could account for a significant component (Ç15%) of the observed secretion of Cl0 , and thus fluid, from the rat parotid. Figs. 2B and 2C show the results of similar experiments to those in Fig. 2A, but carried in the presence of 25 mM HCO30 (for practical reasons, in these experiments NH4Cl was added before carbachol; see Fig. 2 caption). Here the effect of bumetanide is blunted, due at least in part to HCO30 buffering, but a clear effect of NH4/ is still seen (cf., Figs. 2B and 2C). From the summary of these results in Fig. 2D we find that the bumetanide-sensitive intracellular acid load due to 0.1 mM NH4/ in the presence of HCO30 is Ç0.01 pH units/min. The resting acinar pHi in HCO0 3 -replete PSS was 7.14{0.01 (nÅ12). Thus the total acinar intracellular buffering power in the presence of HCO30 is 59.1{0.8 mmol/liter per pH unit (see Methods). Using the same method of calculation outlined above we find that the NH4/-dependent rate of Cl0 entry at physiological NH4/ levels in the presence of HCO30 is Ç0.5 mmoles/g-min. The above difference between the estimated contributions of Na/-NH4/-2Cl0 cotransport to fluid secretion in the presence and absence of HCO30 can be at least partially accounted for by the fact that our experimental procedures would be expected to underestimate the contribution of NH4/-dependent Cl0 entry in HCO30-containing medium. The reason for this is that the blockade of the cotransporter by bumetanide in carbachol-stimulated cells will lower intracellular Cl0 levels due to decreased Cl0 entry. This would, in turn, lead to increased acinar HCO30 loss via Cl0/HCO30 exchange as well as via an apical electrodiffusional HCO30 exit pathway (16, 22). The intracellular acidification resulting from these two components of increased HCO30 loss would blunt the alkalinization induced by bumetanide in the experiments of Fig. 2, consequently leading to an underestimation of the magnitude of the Na/-NH4/-2Cl0 cotransport flux in the presence of HCO30. In Figure 3 we illustrate a proposed model for the involvement of Na/-NH4/-2Cl0 cotransport in salivary fluid secretion based on the results presented here. According to this model the NH4/ cotransported into the cell along with Cl0 is recycled across the plasma membrane as NH3 , which is lost by diffusion, and a H/,
0 FIG. 3. Proposed model by which Na/-NH/ 4 -2Cl cotransport contributes to salivary fluid secretion.
which is extruded via the Na//H/ exchanger. Note that Cl0 entry in this way results in equal components of Na/ flux via Na//H/ exchange and Na/-NH4/-2Cl0 cotransport The net energetic cost for Cl0 secretion by the model shown in Fig. 3 is 3 Cl0 ions secreted per ATP molecule hydrolyzed. This is the same energetic cost for Cl0 secretion encountered when Cl0 entry is mediated via parallel Na//H/ and Cl0/HCO30 exchangers (23). By contrast, when Cl0 entry occurs via Na/-K/-2Cl0 cotransport, 6 Cl0 ions are secreted per ATP consumed (23). Although the operation of the cotransporter is less energy efficient for driving Cl0 secretion when functioning in a Na/-NH4/-2Cl0 mode, this mechanism of operation has the advantage of circumventing the large intracellular to extracellular K/ chemical gradient that thermodynamically opposes Cl0 entry via Na/-K/-2Cl0 cotransport. This effect undoubtedly contributes to the sizable Cl0 fluxes apparently driven via Na/-NH4/-2Cl0 cotransport even at sub-millimolar concentrations of NH4/ (Fig. 2). The absorptive isoform of the Na/-K/-2Cl0 cotransporter has a well established role in NH4/ handling in the renal thick ascending limb of Henle’s loop (24). In addition, several groups have speculated that the secretory isoform of the cotransporter may play a role in NH4/ handling and/or acid secretion in the distal nephron (25, 26). However, the present studies suggest that Na/-NH4/-2Cl0 cotransport may be of considerable physiological importance at the NH4/ levels found in serum, which are one to two orders of magnitude lower than those found in the kidney. Recently Soybel et al. (27) have shown that the secretory Na/-K/-2Cl0 cotransporter plays a previously unsuspected role in HCl secretion by Necturus gastric mucosa. Acid secretion by this epithelium was dramatically blocked by sodium removal, or bumetanide addition, to the serosal solu-
305
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
tion (27). These authors proposed not only that the cotransporter plays a major role in driving Cl0-dependent H/ secretion in Necturus stomach but that it may also account for a previously unexplained furosemidesensitive component of gastric H/ secretion in mammals (28). This hypothesis is consistent with the high levels of secretory Na/-K/-2Cl0 cotransporter mRNA found in both amphibian and mammalian stomach (3,4,27). But Na/-K/-2Cl0 cotransporter-mediated Cl0 entry into the gastric cell is clearly not sufficient for HCl secretion, a source of intracellular protons is required as well. We propose that the physiologically relevant mode of operation of the cotransporter under these circumstances is predominantly as a Na/NH4/-2Cl0 cotransporter so that H/ entry occurs along with that of Cl0 (cf., Fig. 3). This mechanism for gastric acid secretion may be particularly relevant to water dwellers. In these species blood HCO30 levels are low (Ç5 mM) owing to the low ambient PCO2 . At low [HCO30 ] the effectiveness of the Cl0/HCO30 exchanger, which is the dominant mechanism for HCl entry in mammalian gastric cells (29), will be considerably reduced. Consistent with this hypothesis, plasma ammonium concentrations in fish and water-dwelling amphibia are rather high (0.2-0.5 mM; 30). Ammonium levels in the plasma of land-dwellers are typically Ç30100 mM, (30) and higher ammonium concentrations are found in certain tissues such as muscle, liver and kidney. Based on the above results and discussion we propose that Na/-NH4/-2Cl0 cotransport could play a significant physiological role in a variety of cell types and that this mode of operation of the secretory Na/-K/-2Cl0 cotransporter could account for the currently unexplained presence of this protein in some tissues. ACKNOWLEDGMENTS We thank Drs. Bruce J. Baum, Mark A. Knepper, Philip C. Fox, Robert S. Balaban and Yoshito Matsumoto for discussions and encouragement concerning this project.
REFERENCES 1. Haas, M. (1994) Am. J. Physiol. 267, C869–C885. 2. Kaplan, M. R., Mount, D. B., and Delpire, E. (1996) Annu. Rev. Physiol. 58, 649–668.
3. Delpire, E., Rauchman, M. I., Beier, D. R., Hebert, S. C. and Gullans, S. R. (1994) J. Biol. Chem. 269, 25677–25683 4. Payne, J. A., Xu, J. C., Haas, M., Lytle, C. Y., Ward, D., and Forbush, B. (1995) J. Biol. Chem. 270, 17977–17985. 5. Cook, D. I., and Young, J. A. (1989) in Handbook of Physiology, Section 6, The Gastrointestinal System, Vol. 3 (Forte, J. G., Ed.), pp.1–23, Oxford University Press, New York. 6. Evans, R. L., and Turner, R. J. (1997) J. Physiol. 499, 351–359. 7. Boyarsky, G., Hanssen, C., and Clyne, L. A. (1996) Am. J. Physiol. 271, C1131–C1145. 8. Roos, A., and Boron, W. F. (1981) Physiol. Rev. 61, 296–434. 9. Manganel, M., and Turner, R. J. (1991) J. Biol. Chem. 266, 10182–10188. 10. Robertson, M. A., Woodside, M., Foskett, J. K., Orlowski, J., and Grinstein, S. (1997) J. Biol. Chem. 272, 287–294. 11. Soltoff, S. P., McMillian, M. K., Cantley, L. C., Cragoe, E. J., and Talamo, B. R. (1989) J. Gen. Physiol. 93, 285–319. 12. Dissing, S., and Nauntofte, B. (1990) Am. J. Physiol. 259, G1044–G1055. 13. Robertson, M. A., and Foskett, J. K. (1994) Am. J. Physiol. 267, C146–C156. 14. Melvin, J. E., Kawaguchi, M., Baum, B. J., and Turner, R. J. (1987) Biochem. Biophys. Res. Commun. 145, 754–759. 15. Melvin, J. E., and Turner, R. J. (1992) Am. J. Physiol. 262, G393–G398. 16. Melvin, J. E., Moran, A., and Turner, R. J. (1988) J. Biol. Chem. 263, 19564–19569. 17. Paulais, M., and Turner, R. J. (1992) J. Clin. Invest. 89, 1142– 1147. 18. Foskett, J. K. (1990) Am. J. Physiol. 259, C998–C1004. 19. Meister, A. (1956) Physiol. Rev. 36, 103-127 20. Good, D. W., Caflisch, C. R., and DuBose, T. D. (1987) Am. J. Physiol. 252, F491–F500. 21. Schneyer, L. H., and Schneyer, C. A. (1960) Am. J. Physiol. 199, 649–652. 22. Lau, K. R., Elliott, A. C., and Brown, P. D. (1989) Am. J. Physiol. 256, C288–C295. 23. Turner, R. J. (1993) Ann. N Y Acad. Sci. 694, 24–35. 24. Good, D. W. (1994) Annu. Rev. Physiol. 56, 623–647. 25. Wall, S. M., Trinh, H. N., and Woodward, K. E. (1995) Am. J. Physiol. 269, F536–F544. 26. Ginns, S. M., Knepper, M. A., Ecelbarger, C. A., Terris, J., He, X., Coleman, R. A., and Wade, J. B. (1996) J. Am. Soc. Nephrol. 7, 2533–2542. 27. Soybel, D. I., Gullans, S. R., Maxwell, F., and Delpire, E. (1995) Am. J. Physiol. 269, C242–C249. 28. Alan, A., Corcia, A., Klemperer, G., and Caplan, S. R. (1980) Am. J. Physiol. 239, G532–G535. 29. Hersey, S. J., and Sachs, G. (1995) Physiol. Rev. 75, 155–189. 30. Prosser, C. L. (1973) in Comparative Animal Physiology, W. B. Saunders Co., Philadephia, PA.
306
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
245, 301–306 (1998)
RC988428
Evidence for a Physiological Role of NH4/ Transport on the Secretory Na/-K/-2Cl0 Cotransporter Richard L. Evans*,† and R. James Turner* *Membrane Biology Section, Gene Therapy and Therapeutics Branch, National Institute of Dental Research, National Institutes of Health, Bethesda, Maryland 20892; and †Department of Dental Research, University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave., Box 611, Rochester, New York 14642
Received February 20, 1998
The secretory Na/-K/-2Cl0 cotransporter in salivary acinar cells is responsible for driving the transepithelial Cl0 fluxes that give rise to fluid secretion. We demonstrate that the application of the muscarinic agonist carbachol to rat parotid acini results in an intracellular acid load that can be blocked by bumetanide, a specific inhibitor of the cotransporter. One component of this bumetanide-sensitive acid load is ouabain-sensitive while a second is dependent on the presence of sub-millimolar concentrations of NH/ 4 in our media. Our data indicate that this latter effect arises from / NH/ 4 entry on the cotransporter operating in a Na 0 NH/ -2Cl cotransport mode and that at physiological 4 NH/ 4 levels in the rat (Ç0.1 mM), 10-15% of the acinar Cl0 entry occurs via this route. We suggest that Na/0 NH/ 4 -2Cl cotransport may also play a significant physiological role in other cell types and that this mode of operation of the secretory Na/-K/-2Cl0 cotransporter could account for the currently unexplained presence of this protein in a number of tissues. q 1998 Academic Press Key Words: salivary fluid secretion; Na/-K/-2Cl0 cotransport; exocrine glands; NH/ 4 transport; stimulussecretion coupling.
It is now well established that salt and water movement across a number of secretory and absorptive epithelia is due to net transepithelial Cl0 transport. In many of these tissues the active step in this process, cellular Cl0 entry against its electrochemical gradient, is mediated by Na/-K/-2Cl0 cotransport (1). Recent molecular cloning studies have demonstrated that there are, in fact, distinct secretory and absorptive isoforms of the Na/-K/-2Cl0 cotransporter targeted to the basolateral and apical membranes, respectively, of appropriate epithelial tissues (2). The absorptive isoform has thus far only been found in the kidney where it is responsible for salt reabsorption in the thick ascending limb of Henle’s loop. In contrast, evidence for expres-
sion of the secretory isoform as been found in a number of epithelia and in some non-epithelia as well, e.g., salivary glands, stomach, colon, kidney, heart, brain, lung, trachea and skeletal muscle(3,4). In some locations, such as salivary glands, colon and trachea, this transporter has a well-documented role in salt and water secretion (1, 5). However, its somewhat unexpected appearance in many of these other tissues remains to be clarified. In the present paper we document a previously unappreciated feature of the involvement of the secretory Na/-K/-2Cl0 cotransporter in salivary fluid production. Specifically, we show that, at physiological NH4/ levels (Ç0.1 mM), functionally significant Cl0 transport, and hence fluid secretion, by rat parotid acinar cells can be driven by Na/-NH4/-2Cl0 cotransport, i.e., with NH4/ substituting for K/ on the Na/-K/-2Cl0 cotransporter. This component of Cl0 transport is necessarily accompanied by a significant influx of NH4/ and an associated intracellular acidification. Based on these observations we suggest that the physiologically relevant mode of operation of the secretory cotransporter in some tissues is at least partially as a Na/NH4/-2Cl0 transporter and that its functional role in these cases incorporates the movement of NH4/ and/or H/ in addition to that of Cl0. This expanded view of the Na/-K/-2Cl0 cotransporter may help to account for its currently unaccounted for expression in some of the tissues mentioned above. MATERIALS AND METHODS Solutions. Experiments were carried out in a physiological salt solution (PSS) containing 135 mM NaCl, 5.8 mM KCl, 1.8 mM CaCl2 , 0.8 mM MgSO4 , 0.73 mM NaH2PO4 , 20 mM Hepes (pH 7.4 with NaOH), 11 mM glucose, 0.01 % BSA and either 2 mM glutamine (PSS/Gln) or 2 mM alanine (PSS/Ala). In some experiments a 0 HCO0 3 -replete PSS/Ala (PSS/Ala/HCO3 ), in which 25 mM NaCl was replaced with 25 mM NaHCO3 , was employed. Acinar preparation and measurement of intracellular pH (pHi ). Dispersed acini were prepared from the parotid glands of male Wistar
301
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
rats (Harlan Sprague-Dawley Inc., Indianapolis IN) by collagenase digestion and loaded with the fluorescent pH indicator BCECF (2*,7*bis[2-carboxyethyl]-5[6*]-carboxyfluorescein) as previously described (6). Following loading the acinar preparation was washed twice with PSS/Gln and maintained at 307C until use. For experiments carried out in PSS/Ala or PSS/Ala/HCO30 acini were washed once in PSS/Ala and maintained in PSS/Ala for at least 30 min at 307C before use. Throughout the entire preparation and subsequent experimental periods acini were kept continuously agitated (90 cycles/min) and gassed with 100%O2 or 95%O2/5%CO2 as appropriate. Intracellular BCECF fluorescence was monitored in ratio mode at 377C using a Spex ARCM fluorimeter (Spex Industries, Edison NJ) as previously described (6). At the end of each experimental run acini were lysed with 0.25% Triton X-100 and the BCECF fluorescence ratio of the released dye determined in PSS (pH 7.4). In a separate series of experiments the BCECF signal was calibrated by titrating the lysate with NaOH and/or HCl. The resulting calibration curve was linear with slope 0.263{0.006 pH units/fluorescence ratio unit (nÅ3) over the pH interval 6.45-7.45. For BCECF traces within this pH range (which includes all the results illustrated in the paper) it was assumed that the fluorescence ratio of the intracellular dye at pHi 7.4 would be the same as that of the released dye from the same run and the fluorescence trace was converted to pHi using the above slope of the linear portion of the calibration curve. Using this method the resting pHi of the rat parotid acinar preparation was found to be 7.31 { 0.02 (n Å3 6). In order to confirm the validity of the above method we also determined the resting acinar pHi independently using a recently published ‘‘null technique’’ (7). Briefly, acini were exposed to combinations of a weak base (ammonium) and a weak acid (propionate) at concentrations such that if the acinar pHi were at a certain predetermined ‘‘null pH’’ value the intracellular alkalinization resulting from the former would be exactly balanced by the intracellular acidification produced by the latter (7). Combinations of sodium propionate (held constant at 10 mM) and NH4Cl (varied from 16.21 to 4.07 mM) corresponding to null pH’s from 7.5 to 7.2 were employed in order to bracket the resting pHi . A plot of the intracellular alkalinizations or acidifications resulting from these challenges vs. the corresponding null pH’s was linear (not shown) allowing determination of the resting acinar pHi by interpolation (7.31 { 0.02 in PSS/Gln; n Å 3). The excellent agreement of this value with that obtained above is consistent with the assumption that the fluorescence properties of intracellular BCECF are similar to those of the released dye from Triton X-100 treated cells. Determination of acinar intracellular buffering power. Acini in PSS/Gln were challenged by the addition of NH4Cl (10 or 20 mM) or sodium propionate (10, 20 or 30 mM) to the extracellular medium. The resulting acute intracellular alkalinizations or acidifications were then determined by extrapolation of the associated BCECF traces back to the time of NH/ 4 or propionate addition. The intrinsic intracellular buffering power (bi) calculated from each of these challenges was assigned to the mid point of the associated alkalinization or acidification. All of these determinations of bi were within experimental error of each other (Ç10%) indicating that bi is constant over the pHi interval 7.1 to 7.5. The average value of bi over this pHi range was 27.6{0.8 mmol/liter per pH unit. In HCO0 3 -containing PSS the total intracellular buffering power is the sum of the intrinsic buffering power plus the intracellular CO2 0 buffering power given by bCO2 Å 2.3 [HCO0 3 ]i, where [HCO3 ]i is the intracellular HCO0 3 concentration (8). Ammonium determination. Medium ammonium levels were determined with the Sigma Diagnostic Kit #171-UV. Data analysis and presentation. Reported experimental values are means { S.E.M. for three or more independent determinations performed under the same conditions on different acinar cell preparations. P values õ 0.05 (Student’s t test) were taken to represent statistically significant differences.
RESULTS AND DISCUSSION In situ, the muscarinic agonist acetylcholine is the main fluid secretory stimulus for salivary glands. Past studies have shown that both Na/-K/-2Cl0 cotransport (6) and Na//H/ exchange activities (9,10) in rat parotid acinar cells are rapidly and dramatically upregulated (Ç20- and Ç3-fold, respectively) following stimulation by the acetylcholine analogue carbachol. In addition, measurements of Na/ uptake into carbachol-stimulated acini have demonstrated large components of Na/ entry due to both the cotransporter and the exchanger (11-13). As already mentioned, a major role for the Na/K/-2Cl0 cotransporter in salivary Cl0 secretion is welldocumented (5, 14). There is also good evidence that the Na//H/ exchanger contributes to Cl0 secretion by acting in parallel with a basolateral Cl0/HCO30 exchanger (5, 15) to effect acinar Cl0 entry. Interestingly, however, a significant component of carbachol-stimulated Na/ entry via the Na//H/ exchanger persists in HCO30-free media (12, 13) suggesting additional involvement of this transporter in the muscarinic response. The present studies were begun to further explore the roles and interactions of the Na/-K/-2Cl0 cotransporter and the Na//H/ exchanger during fluid secretion. Fig. 1 summarizes a number of our initial observations. In these experiments the fluorescent pHi indicator BCECF was used to monitor pHi (see Methods) following carbachol (Cch; 10mM) stimulation of rat parotid acini suspended in glutamine-containing PSS (PSS/ Gln; see Methods), a medium we (6, 9, 15-17) and others (10, 13, 18) have employed in a number of earlier studies. As illustrated in the lower trace of Fig. 1A, when otherwise untreated acini are exposed to carbachol they rapidly alkalinize (õ0.1 pH unit). This effect has been noted in several earlier studies (11, 13, 16) and is presumably due to the agonist-induced upregulation of Na//H/ exchange activity mentioned above. Fig. 1A also demonstrates that this rise in pHi is enhanced in the presence of the Na/-K/-2Cl0 cotransporter inhibitor bumetanide (10mM). In these and past experiments (6, 17) we have confirmed that the application of bumetanide in the absence of carbachol is without effect on pHi (data not shown); thus this enhanced alkalinization suggests the presence of a stimulus-induced, bumetanide-sensitive component of intracellular acidification. This effect is confirmed and shown more directly in Fig. 1B where the Na//H/ exchanger has been blocked by the addition of its potent inhibitor DMA (5-(N,N-dimethylamiloride; 20 mM). Under these conditions the acini acidify following the application of carbachol and this acidification is blunted by bumetanide, again consistent with the presence of a stimulus-induced, bumetanide-sensitive intracellular acid load. The results in Figs. 1A and 1B have been quantitated in the upper 4 bars of Fig. 1C (see Figure caption).
302
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Effects of inhibitors of Na/-K/-2Cl0 cotransport, Na//H/ exchange and Na//K/ ATPase on the response of rat parotid acinar pHi to muscarinic stimulation. The results shown in panels A and B are representative pHi traces (see Methods) from BCECF-loaded acini suspended in PSS/Gln and stimulated with 10 mM carbachol (Cch) in the presence or absence of 10 mM bumetanide (bumet; applied 50 s prior to the application of Cch) and/or 20 mM DMA as indicated. These and additional results have been quantitated in panel C by measuring the difference between the pHi found 100 s after the application of carbachol and that observed before carbachol addition (DpHi). In these experiments ouabain was used at 1 mM (applied 60 s before carbachol) and acini in PSS/Ala were prepared as described in Methods. The point labeled ‘PSS/Ala / bumet*’ was obtained with a slightly different protocol. Here acini were incubated with carbachol for 250 s (at which point DpHi was not significantly different from that found after 100 s incubation with carbachol; DpHi was 0.083 { 0.008 250 s after carbachol vs. 0.079 { 0.004 100 s after carbachol) then bumetanide (10 mM) was added and the DpHi from that observed before carbachol addition was determined 100 s later. The average results { S.E.M. of 3 or more independent experiments under each condition are shown.
We reasoned that the above effect of bumetanide might be due to a Na/-K/-2Cl0 cotransporter-induced metabolic acid load due to the increased Na//K/ATPase turnover required to extrude the Na/ entering the cell via the cotransporter during secretion. Consistent with this hypothesis the carbachol-induced intracellular acidification observed in the presence of DMA was blunted by the Na//K/-ATPase inhibitor ouabain (Fig. 1C). However, a significant effect of bumetanide remained even in the presence of ouabain (Fig. 1C), indicating that this explanation could not completely account for the bumetanide-sensitive component of acidification. Moreover, when DMA-treated acini were stimulated with carbachol in the absence of extracellular K/, a maneuver expected to inhibit both Na//K/ATPase and Na/-K/-2Cl0 cotransport, acinar acidification was enhanced (Fig. 1C) rather than blunted. After some consideration of this latter result it occurred to us that the apparent effects of Na/-K/-2Cl0 cotransport inhibition illustrated in Fig. 1C might be accounted for by the presence of NH4/ in our medium. We hypothesized that the bumetanide-sensitive component of intracellular acidification could be due to the
Na/-K/-2Cl0 cotransporter operating in a Na/-NH4/2Cl0 mode, with the NH4/ possibly arising from the spontaneous (19) or acinar-induced breakdown of the glutamine present in PSS/Gln. It is well established that NH4/ can substitute for K/ on the cotransporter (6, 17). The NH4/ entry generated in this way would result in an intracellular acid load because of the dissociation of the accumulated intracellular NH4/ into NH3 plus a H/, allowing the permeant species NH3 to diffuse out of the cell, leaving the H/ behind. This hypothesis can account for the enhanced intracellular acidification observed in the absence of K/ (Fig. 1C) since under these conditions there would be increased NH4/ entry, and thus increased intracellular acidification, owing to the elimination of the competition between K/ and NH4/ for the cotransporter. As an initial test of this hypothesis we measured the alkalinization induced by carbachol in cells incubated in glutamine-free medium (PSS/Ala; see Methods). In these experiments a significantly higher alkalinization was observed than in the presence of glutamine and the effect of bumetanide addition was significantly reduced (Fig. 1C), consistent with the predicted effect of elimi-
303
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
0 0 FIG. 2. Quantitative estimation of the contribution of Na/-NH/ 4 -2Cl cotransport to cellular Cl entry during carbachol-induced fluid secretion. In the experiment illustrated in panel A BCECF-loaded rat parotid acini suspended in PSS/Ala were stimulated with 10 mM carbachol then challenged with 200 mM NH4Cl followed by 10 mM bumetanide as indicated. The protocols for the experiments illustrated in panels B and C were similar except that the medium was PSS/Ala/HCO0 3 and NH4Cl, at the concentrations indicated, was added 180 s before stimulation with carbachol. In panel D the initial rate of alkalinization following bumetanide addition is plotted vs. [NH/ 4 ] for acini suspended in PSS/Ala (open circles) and PSS/Ala/HCO0 3 (closed circles). The averaged results { S.E.M. of 4 or more independent determinations at each [NH/ 4 ] are illustrated.
nation of glutamine-derived NH4/ from the medium. We next determined the NH4/ content of our media directly. These results confirmed the presence of 12 { 3 mM NH4/ in PSS/Gln alone and 24 { 4 mM NH4/ in PSS/Gln plus acini (the experimental conditions of Fig. 1 were employed). Much lower levels of NH4/ (3 { 1 mM) were found in PSS/Ala and there was no detectable increase in the presence of acini (4 { 1 mM). Since NH4/ levels in rat plasma are Ç0.1 mM (20) the above results suggest that a substantial flux of H/equivalents and thus a substantial ion flux could be carried via Na/-NH4/-2Cl0 cotransport at physiological NH4/ concentrations. In order to explore this possibility in more detail we carried out the experiments illustrated in Fig. 2. In trace A of Fig. 2, acini in PSS/Ala were first treated with 10 mM carbachol, then challenged with 200 mM NH4Cl and finally exposed to 10 mM bumetanide, as indicated. This experiment confirms directly that sub-millimolar concentrations of NH4/ can produce a significant intracellular acid load in carbachol-stimulated rat parotid acini. The reversal of this effect by bumetanide is again consistent with the involvement of Na/-NH4/-2Cl0 cotransport. In control experiments (not shown) we have demonstrated
that the application of up to 1 mM NH4Cl to acini before carbachol treatment produces no significant effect on pHi. Thus the effect of NH4/ illustrated in Fig. 2A only occurs after the dramatic upregulation (Ç20-fold) of Na/-K/-2Cl0 cotransport activity resulting from muscarinic stimulation (6). The initial rate of alkalinization observed after bumetanide addition in Fig. 2A provides a measure of the bumetanide-sensitive intracellular acid load present at this concentration of NH4/. In Fig. 2D we summarize the results of a series of experiments of this type where we have determined the magnitude of this acid load as a function of NH4/ concentration in (HCO30-free) PSS/ Ala (open circles). From these data we find that the bumetanide-sensitive, NH4/-dependent intracellular acid load at 0.1 mM NH4/ (i.e., the increment in the bumetanide-sensitive intracellular acid load due to the presence of 0.1 mM NH4/) is Ç0.04 pH units/min. From the known buffering power of the acinar cytoplasm (27.6{0.8 mmol/liter per pH unit; see Methods) and the intracellular water content of the parotid (Ç45%; 21) this value can be converted into a rate of acinar NH4/ entry/gram tissue (Ç0.5 mmoles/g-min). Assuming that this bumetanide-sensitive NH4/ entry occurs via
304
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Na/-NH4/-2Cl0 cotransport, this value can be converted to a rate of acinar Cl0 entry (Ç1 mmole/g-min). As already indicated, salivary fluid secretion is thought to be primarily due to acinar Cl0 secretion driven in large part by Cl0 entry across the basolateral membrane via the Na/-K/-2Cl0 cotransporter. In previous experiments from our laboratory carried out under similar conditions to those presented here we have determined that the rate of sustained Cl0 secretion from carbacholstimulated rat parotid acini is Ç7 mmoles/g-min (14). Hence the above results demonstrate that at physiological NH4/ levels Na/-NH4/-2Cl0 cotransport could account for a significant component (Ç15%) of the observed secretion of Cl0 , and thus fluid, from the rat parotid. Figs. 2B and 2C show the results of similar experiments to those in Fig. 2A, but carried in the presence of 25 mM HCO30 (for practical reasons, in these experiments NH4Cl was added before carbachol; see Fig. 2 caption). Here the effect of bumetanide is blunted, due at least in part to HCO30 buffering, but a clear effect of NH4/ is still seen (cf., Figs. 2B and 2C). From the summary of these results in Fig. 2D we find that the bumetanide-sensitive intracellular acid load due to 0.1 mM NH4/ in the presence of HCO30 is Ç0.01 pH units/min. The resting acinar pHi in HCO0 3 -replete PSS was 7.14{0.01 (nÅ12). Thus the total acinar intracellular buffering power in the presence of HCO30 is 59.1{0.8 mmol/liter per pH unit (see Methods). Using the same method of calculation outlined above we find that the NH4/-dependent rate of Cl0 entry at physiological NH4/ levels in the presence of HCO30 is Ç0.5 mmoles/g-min. The above difference between the estimated contributions of Na/-NH4/-2Cl0 cotransport to fluid secretion in the presence and absence of HCO30 can be at least partially accounted for by the fact that our experimental procedures would be expected to underestimate the contribution of NH4/-dependent Cl0 entry in HCO30-containing medium. The reason for this is that the blockade of the cotransporter by bumetanide in carbachol-stimulated cells will lower intracellular Cl0 levels due to decreased Cl0 entry. This would, in turn, lead to increased acinar HCO30 loss via Cl0/HCO30 exchange as well as via an apical electrodiffusional HCO30 exit pathway (16, 22). The intracellular acidification resulting from these two components of increased HCO30 loss would blunt the alkalinization induced by bumetanide in the experiments of Fig. 2, consequently leading to an underestimation of the magnitude of the Na/-NH4/-2Cl0 cotransport flux in the presence of HCO30. In Figure 3 we illustrate a proposed model for the involvement of Na/-NH4/-2Cl0 cotransport in salivary fluid secretion based on the results presented here. According to this model the NH4/ cotransported into the cell along with Cl0 is recycled across the plasma membrane as NH3 , which is lost by diffusion, and a H/,
0 FIG. 3. Proposed model by which Na/-NH/ 4 -2Cl cotransport contributes to salivary fluid secretion.
which is extruded via the Na//H/ exchanger. Note that Cl0 entry in this way results in equal components of Na/ flux via Na//H/ exchange and Na/-NH4/-2Cl0 cotransport The net energetic cost for Cl0 secretion by the model shown in Fig. 3 is 3 Cl0 ions secreted per ATP molecule hydrolyzed. This is the same energetic cost for Cl0 secretion encountered when Cl0 entry is mediated via parallel Na//H/ and Cl0/HCO30 exchangers (23). By contrast, when Cl0 entry occurs via Na/-K/-2Cl0 cotransport, 6 Cl0 ions are secreted per ATP consumed (23). Although the operation of the cotransporter is less energy efficient for driving Cl0 secretion when functioning in a Na/-NH4/-2Cl0 mode, this mechanism of operation has the advantage of circumventing the large intracellular to extracellular K/ chemical gradient that thermodynamically opposes Cl0 entry via Na/-K/-2Cl0 cotransport. This effect undoubtedly contributes to the sizable Cl0 fluxes apparently driven via Na/-NH4/-2Cl0 cotransport even at sub-millimolar concentrations of NH4/ (Fig. 2). The absorptive isoform of the Na/-K/-2Cl0 cotransporter has a well established role in NH4/ handling in the renal thick ascending limb of Henle’s loop (24). In addition, several groups have speculated that the secretory isoform of the cotransporter may play a role in NH4/ handling and/or acid secretion in the distal nephron (25, 26). However, the present studies suggest that Na/-NH4/-2Cl0 cotransport may be of considerable physiological importance at the NH4/ levels found in serum, which are one to two orders of magnitude lower than those found in the kidney. Recently Soybel et al. (27) have shown that the secretory Na/-K/-2Cl0 cotransporter plays a previously unsuspected role in HCl secretion by Necturus gastric mucosa. Acid secretion by this epithelium was dramatically blocked by sodium removal, or bumetanide addition, to the serosal solu-
305
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
tion (27). These authors proposed not only that the cotransporter plays a major role in driving Cl0-dependent H/ secretion in Necturus stomach but that it may also account for a previously unexplained furosemidesensitive component of gastric H/ secretion in mammals (28). This hypothesis is consistent with the high levels of secretory Na/-K/-2Cl0 cotransporter mRNA found in both amphibian and mammalian stomach (3,4,27). But Na/-K/-2Cl0 cotransporter-mediated Cl0 entry into the gastric cell is clearly not sufficient for HCl secretion, a source of intracellular protons is required as well. We propose that the physiologically relevant mode of operation of the cotransporter under these circumstances is predominantly as a Na/NH4/-2Cl0 cotransporter so that H/ entry occurs along with that of Cl0 (cf., Fig. 3). This mechanism for gastric acid secretion may be particularly relevant to water dwellers. In these species blood HCO30 levels are low (Ç5 mM) owing to the low ambient PCO2 . At low [HCO30 ] the effectiveness of the Cl0/HCO30 exchanger, which is the dominant mechanism for HCl entry in mammalian gastric cells (29), will be considerably reduced. Consistent with this hypothesis, plasma ammonium concentrations in fish and water-dwelling amphibia are rather high (0.2-0.5 mM; 30). Ammonium levels in the plasma of land-dwellers are typically Ç30100 mM, (30) and higher ammonium concentrations are found in certain tissues such as muscle, liver and kidney. Based on the above results and discussion we propose that Na/-NH4/-2Cl0 cotransport could play a significant physiological role in a variety of cell types and that this mode of operation of the secretory Na/-K/-2Cl0 cotransporter could account for the currently unexplained presence of this protein in some tissues. ACKNOWLEDGMENTS We thank Drs. Bruce J. Baum, Mark A. Knepper, Philip C. Fox, Robert S. Balaban and Yoshito Matsumoto for discussions and encouragement concerning this project.
REFERENCES 1. Haas, M. (1994) Am. J. Physiol. 267, C869–C885. 2. Kaplan, M. R., Mount, D. B., and Delpire, E. (1996) Annu. Rev. Physiol. 58, 649–668.
3. Delpire, E., Rauchman, M. I., Beier, D. R., Hebert, S. C. and Gullans, S. R. (1994) J. Biol. Chem. 269, 25677–25683 4. Payne, J. A., Xu, J. C., Haas, M., Lytle, C. Y., Ward, D., and Forbush, B. (1995) J. Biol. Chem. 270, 17977–17985. 5. Cook, D. I., and Young, J. A. (1989) in Handbook of Physiology, Section 6, The Gastrointestinal System, Vol. 3 (Forte, J. G., Ed.), pp.1–23, Oxford University Press, New York. 6. Evans, R. L., and Turner, R. J. (1997) J. Physiol. 499, 351–359. 7. Boyarsky, G., Hanssen, C., and Clyne, L. A. (1996) Am. J. Physiol. 271, C1131–C1145. 8. Roos, A., and Boron, W. F. (1981) Physiol. Rev. 61, 296–434. 9. Manganel, M., and Turner, R. J. (1991) J. Biol. Chem. 266, 10182–10188. 10. Robertson, M. A., Woodside, M., Foskett, J. K., Orlowski, J., and Grinstein, S. (1997) J. Biol. Chem. 272, 287–294. 11. Soltoff, S. P., McMillian, M. K., Cantley, L. C., Cragoe, E. J., and Talamo, B. R. (1989) J. Gen. Physiol. 93, 285–319. 12. Dissing, S., and Nauntofte, B. (1990) Am. J. Physiol. 259, G1044–G1055. 13. Robertson, M. A., and Foskett, J. K. (1994) Am. J. Physiol. 267, C146–C156. 14. Melvin, J. E., Kawaguchi, M., Baum, B. J., and Turner, R. J. (1987) Biochem. Biophys. Res. Commun. 145, 754–759. 15. Melvin, J. E., and Turner, R. J. (1992) Am. J. Physiol. 262, G393–G398. 16. Melvin, J. E., Moran, A., and Turner, R. J. (1988) J. Biol. Chem. 263, 19564–19569. 17. Paulais, M., and Turner, R. J. (1992) J. Clin. Invest. 89, 1142– 1147. 18. Foskett, J. K. (1990) Am. J. Physiol. 259, C998–C1004. 19. Meister, A. (1956) Physiol. Rev. 36, 103-127 20. Good, D. W., Caflisch, C. R., and DuBose, T. D. (1987) Am. J. Physiol. 252, F491–F500. 21. Schneyer, L. H., and Schneyer, C. A. (1960) Am. J. Physiol. 199, 649–652. 22. Lau, K. R., Elliott, A. C., and Brown, P. D. (1989) Am. J. Physiol. 256, C288–C295. 23. Turner, R. J. (1993) Ann. N Y Acad. Sci. 694, 24–35. 24. Good, D. W. (1994) Annu. Rev. Physiol. 56, 623–647. 25. Wall, S. M., Trinh, H. N., and Woodward, K. E. (1995) Am. J. Physiol. 269, F536–F544. 26. Ginns, S. M., Knepper, M. A., Ecelbarger, C. A., Terris, J., He, X., Coleman, R. A., and Wade, J. B. (1996) J. Am. Soc. Nephrol. 7, 2533–2542. 27. Soybel, D. I., Gullans, S. R., Maxwell, F., and Delpire, E. (1995) Am. J. Physiol. 269, C242–C249. 28. Alan, A., Corcia, A., Klemperer, G., and Caplan, S. R. (1980) Am. J. Physiol. 239, G532–G535. 29. Hersey, S. J., and Sachs, G. (1995) Physiol. Rev. 75, 155–189. 30. Prosser, C. L. (1973) in Comparative Animal Physiology, W. B. Saunders Co., Philadephia, PA.
306
AID
BBRC 8428
/
694f$$$141
03-31-98 10:10:13
bbrcg
AP: BBRC