INTRODUCTION

Nitric oxide has been recently recognized as an important second messenger in a variety of cell types. The generation of NO from L-arginine is catalyzed by NO-synthase in response to a number of stimuli including bacterial lipopolysaccharide, tumor necrosis factor-, and -interferon (see Kiechle and Malinski, 1993; Anggard, 1994; Weinberg et al., 1995 for review). NO is an unstable intermediate compound which, in aerobic aqueous solutions such as the cytosol, is rapidly metabolized primarily to nitrite (NO₂) and to a lesser extent to nitrate (NO₃). In the presence of oxidizing species such as oxyhemoproteins, NO₂ is rapidly converted to the more stable NO₃ (Ignarro et al., 1993; Veszelovszky et al., 1995). Lipopolysaccharide and cytokine-induced formation of  NO, NO₂, and NO₃ in vitro has been extensively reported in neutrophils, primary macrophages, and monocyte/macrophage cell lines (Iyengar et al., 1987; Miwa et al., 1987; Schmidt et al., 1989; Wright et al., 1989). In animal models, increased production of NO₃ has been reported in sepsis (Ohshima et al., 1994; Oudenhoven et al., 1994; Wright et al., 1992), glomerulonephritis (Sever et al., 1992), and graft rejection (Tanaka et al., 1995; Winlaw et al., 1994; Drobyski et al., 1994). In humans, acute graft vs. host disease, bacterial and viral meningitis, acute gastroenteritis, and sepsis have been associated with increased NO₃ production (Jungersten et al., 1993; Shi et al., 1993; Milstien et al., 1994; Weiss et al., 1995).

The amounts of NO₂ and NO₃ generated from NO are substantial. The cytosolic concentrations of NO₂ and NO₃ in stimulated macrophages were found to be 119.3 nmol/ml and 281.2 nmol/ml, respectively (Marletta et al., 1988). These figures greatly underestimate the amount produced, since NO₂ and NO₃ are not retained within the cell and can be readily recovered in the extracellular milieu. In in vitro experiments, 100- 200 nmol of NO₂ and NO₃ were recovered from the medium bathing 10⁶ stimulated macrophages (Stuehr and Marletta, 1987a, b, c ; Miwa et al., 1991). Plasma nitrate levels as high as 200 µM have been reported in disease states such as acute gastroenteritis (Jungersten et al., 1993). These findings imply that NO₂ and NO₃ must be transported effectively across the plasma membrane for subsequent disposal. However, the pathway(s) for transport of NO₂ and NO₃ in mammalian cells are poorly understood.

This report describes a NO₃ transport system present in a variety of mammalian cells. In the course of measurements of the anion dependence of the intracellular pH of Chinese hamster ovary (CHO)¹ cells, we made the serendipitous observations that addition of extracellular NO₃ induced a reproducible cytosolic acidification. The purpose of this study was to characterize this NO₃-induced cytosolic acidification, to compare the underlying mechanism with known pH_i regulatory systems, and to describe its pharmacological profile.

METHODS

Materials and Media

Nigericin, 27 bis-(2-carboxyethyl)-5(and 6) carboxyfluorescein (BCECF) free acid and acetoxymethyl ester were purchased from Molecular Probes, Inc. (Eugene, OR). Antimycin A, 2-deoxy-D-glucose, 2-(N-morpholino)ethanesulfonic acid (MES), -nicotinamide adenine dinucleotide phosphate (NADPH), glucose-6-phosphate, bafilomycin, 4,4-diisothiocyanostilbene-2,2 disulfonate (DIDS), -cyano-4-hydroxycinnamate (CHC), and p-chloromercuribenzene sulfonate (pCMBS) were obtained from Sigma Chemical Co. (St. Louis, MO). Ethacrynic acid was purchased from Serva (Heidelberg, Austria) and phloretin from K+K (Hollywood, CA). Glucose-6-phosphate dehydrogenase and nitrate reductase were from Boehringer Mannheim (Indianapolis, IN). All other chemicals and salts were purchased from Sigma Chemical Co.

PBS contained (in mM): 140 NaCl, 10 KCl, 8 sodium phosphate, 2 potassium phosphate, pH 7.4. The sodium chloride solution contained (in mM): 117 NaCl, 1.66 MgSO₄, 1.36 CaCl₂, 5.36 KCl, 25 HEPES, 5.55 glucose. Sodium nitrate solution contained (in mM): 117 NaNO₃, 6.2 MgSO₄, 1.36 calcium gluconate, 5.36 potassium gluconate, 25 HEPES, 5.55 glucose. Sodium gluconate solution contained (in mM): 117 sodium gluconate, 6.2 MgSO₄, 1.36 calcium gluconate, 5.36 potassium gluconate, 25 HEPES, 5.55 glucose. Potassium and N-methyl-D-glucammonium (NMG) solutions were made with equimolar substitution of the appropriate salts. To study the nitrate concentration dependence we used media with NaNO₃ concentrations varying from 3.65 to 117 mM that were osmotically balanced with sodium gluconate. Sodium nitrite solutions containing 10-117 mM NaNO₂ were prepared similarly. Unless otherwise indicated, all solutions were titrated to pH 7.5.

The composition of the low-buffer solution used to fill the patch pipettes was (in mM): 1 MES, 120 KCl, 1 MgCl₂, 12 potassium gluconate, 50 glucose, 1 Mg-ATP, pH to 7.5 with NMG -OH. The high-buffer pipette solution contained (in mM): 50 HEPES, 50 Tris, 20 KCl, 1 MgCl₂, 12 potassium gluconate, 50 glucose, 1 Mg-ATP, pH to 7.3 with HCl. All pipette solutions also contained 300 µM BCECF-free acid. External solutions used in patch clamp experiments were composed of (in mM): 120 KCl, 1 MgCl₂, 12 KOH, 50 HEPES, pH 7.0. The potassium nitrate solution was made with equimolar substitution of KNO₃ for KCl. All solutions were nominally bicarbonate-free and were adjusted to 290 ± 10 mosM with the major salt.

Cells

WT5 is a sub-line of wild-type CHO cells. AP1 is a cell line derived from WT5 that is devoid of endogenous Na⁺/H⁺ activity (Rotin and Grinstein, 1989). HEK 293 is a human embryonic kidney cell line (Lee et al., 1991). J774 is a murine monocyte/macrophage cell line. Cells were grown in -MEM (Ontario Cancer Institute, Toronto, ON) containing 25 mM NaHCO₃ supplemented with 10% FCS, 100 U/ml penicillin, and 100 µg/ml streptomycin (Life Technologies, Inc., Grand Island, NY) and were incubated in a humidified environment containing 95% air and 5% CO₂ at 37°C. Cultures were reestablished from frozen stocks regularly, and cells from passages 3 to 20 were used for the experiments.

Where indicated, intracellular ATP was depleted by incubating the cells for 10 min in glucose-free medium with 5 mM 2-deoxy- D-glucose and 1 µg/ml antimycin A, to inhibit both glycolysis and oxidative phosphorylation. This protocol has previously been shown to deplete >90% of the ATP in CHO cells within 10 min (Goss et al., 1994). Subsequent fluorescence measurements were performed in glucose-free media containing 5.5 mM 2-deoxy- D-glucose.

Measurements of Cytosolic pH

Microfluorimetry The cytosolic pH (pH_i) of small groups of cells was determined by microphotometry of the fluorescence emission of BCECF using dual wavelength excitation. Cells grown to confluence on 25-mm glass coverslips (Thomas Scientific, Swedesboro, NJ) were loaded with BCECF by incubation with 2 µg/ml of the precursor acetoxymethyl ester form for 10 min at 37°C. The coverslips were then mounted in a Leiden Coverslip Dish (Medical System Corp., Greenvale, NY) and placed into a thermostatted holding chamber heated to 37°C (Open Perfusion Microincubator; Medical Systems Corp., Greenvale, NY) attached to the stage of a Nikon Diaphot TMD inverted microscope (Nikon Canada, Toronto, ON). Cells were visualized using a Nikon Fluor 40×/1.3 N.A. oil-immersion objective and a Hoffman modulation contrast video system with an angled condenser (Modulation Optics) through a CCD-72 video camera and control unit (Dage-MTI, Michigan City, IN) connected to a Panasonic monitor. Clusters of 6-12 cells from the confluent culture were selected for analysis with an adjustable diaphragm. The chamber was continuously perfused at 0.5 ml/min to allow for complete exchange of the bath solution once every minute using a gravity-driven system and a Leiden aspirator. Solutions could be switched by opening solenoid valves (General Valve, Fairfield, NJ). When rapid solution changes were required, 3 aliquots of 1 ml of the new medium were quickly pipetted (<15 s) into the chamber and perfusion was continued using the new medium (Kapus et al., 1994).

Fluorescence Imaging For imaging of single cell pH_i, cells were grown on round glass coverslips to 60-70% confluence and loaded with BCECF as described above. Ratio fluorescence imaging was performed on a Zeiss Axiovert 100 TV inverted microscope (Zeiss, Oberkochen, Germany) equipped with a 75 W Xenon lamp (XBO 75, Zeiss), a shutter/filter wheel assembly (Lambda 10, Sutter Instrument Co., Novato, CA), a NeoFluar 63×/1.25 N. A. objective and a high-resolution (1,317 × 1,025 pixels, KAF 1400, Kodak) cooled, digital CCD camera (TEA/ CCD 1317; Princeton Instruments, Trenton, NJ) interfaced to a Pentium 90 computer (Dell Inc., Toronto, ON, Canada) via a 12 bit, 1 MHz camera-controller (Princeton controller: ST-38; Princeton Instruments, Trenton, NJ). Image acquisition and excitation filter selection were controlled by the Metafluor software (Universal Imaging Corp., West Chester, PA). All measurements of pH_i were performed at 37°C.

Electrophysiology and pH_i Measurements in Voltage-clamped Cells

Cells were patch-clamped in the whole-cell configuration of the patch clamp technique using an Axopatch-1D amplifier (Axon Instruments Inc., Foster City, CA), as described (Kapus et al., 1993). Electrodes were made from filament-filled borosilicate glass capillaries (World Precision Instruments Inc., Sarasota, FL) using a horizontal puller (P-87; Sutter Instrument Co., Novato, CA) and a microforge (MF-9; Narishige USA, Greenvale, NY). Pipette resistance ranged from 2 to 10 M; seal resistance ranged from 10 to 50 G. Series resistance varied between 5 and 30 M, and cell capacitance was between 12 and 34 pF.

Cytosolic pH in voltage-clamped cells was measured microfluorimetrically on the Photon Technologies Inc. photometric system described above, using pipette solutions with a low buffering power (1 mM MES) to maximize the NO₃-induced intracellular pH changes. Cells were patch-clamped in the whole-cell configuration and loaded with BCECF-free acid by diffusion of the dye from the pipette solution into the cytosol. Measurements were initiated 5 min after attaining the whole-cell configuration to allow for equilibration of the cytosolic pH with the pipette pH and for adequate BCECF loading. Cells were superfused at 0.5 ml/ min with the indicated solutions. Calibration of fluorescence ratio vs. pH was performed on single nonpatched cells, using the nigericin technique described above (Thomas et al., 1979). One calibration curve was obtained every day by averaging data from 3 to 6 cells sequentially perfused with KCl media buffered at 3 different pH values ranging from 6.0 to 7.5. All patch clamp experiments were carried out at room temperature. Single-cell pH_i measurements were also performed microfluorimetrically in nonpatched cells at room temperature to ensure that the NO₃-induced cytosolic acidification was present under these conditions as well (data not shown).

Measurements of Intracellular Nitrate Content

AP1 cells were grown to confluence on 6-well plastic tissue culture dishes (Costar Corp., Cambridge, MA). Culture medium was aspirated and cells were incubated with isotonic NaNO₃ solution at 0 or 37°C for the indicated times. Where indicated, the solution contained 100 µM DIDS. At the end of the incubation period, each well was rapidly washed 3 times with 10 ml of PBS at 4°C and subsequently the cells were lysed using 1 ml distilled water and repeated freeze-thawing. Whole cell lysates were centrifuged for 10 min in a microcentrifuge (Beckman Microfuge, 13,500 rpm; Beckman Instruments, Fullerton, CA) to remove cellular debris. The NO₃ content of the resultant supernatant was measured using the Greiss reaction after reduction of NO₃ to NO₂, as previously described (Green et al., 1982; Verdon et al., 1995; Gilliam et al., 1993). Reduction of NO₃ to NO₂ was performed by incubating the supernatant with 1 µM NADPH, 0.16 U/ml glucose-6-phosphate dehydrogenase, 500 µM glucose-6-phosphate, and 0.80 U/ml nitrate reductase at room temperature for 45 min. A 500-µl aliquot of the sample was then incubated with 250 µl of 0.1% (N-1-naphthyl)-naphthylethylene diamine dihydrochloride and 250 µl of 1% sulfanilamide in 5% phosphoric acid at 37°C for 10 min. Absorbance was measured spectrophotometrically at 540 nm in a Hitachi U-2000 spectrophotometer. A calibration curve was constructed by adding increasing concentrations of NaNO₃ to lysates of cells not exposed to exogenous NO₃. A second aliquot of the cell lysate was used for protein determination using the BioRad Protein Assay Kit (BioRad Laboratories, Richmond, CA). Total cell number was estimated by normalizing the protein content in the whole cell lysate to that of a cell suspension containing a known number of cells, determined electronically using the Coulter counter. The cell suspension was prepared by addition of 1 ml of trypsin-EDTA (GIBCO-BRL, Life Technologies Inc., Grand Island, NY) to AP1 cells grown to confluence in a 75-ml tissue culture flask. The volume of the suspended cells was determined with the Coulter-Channelyzer combination. The intracellular NO₃ concentration was then calculated by dividing the NO₃ content of the cell lysate by the corresponding cellular volume.

Data Analysis and Statistics

Quantification of cell-associated fluorescence was performed using the Felix software package (Photon Technologies, Inc.) or the Metamorph/Metafluor package (Universal Imaging, Inc., West Chester, PA). Mean H⁺_(equivalent) flux was calculated by multiplying the rate of pH_i change (pH_i/time) by the buffering capacity of CHO cells, measured to be 25 mmol/pH/liter of cells in the pH range of our measurements (Kapus et al., 1994). The rate of pH_i change was derived by linear regression of the pH_i vs. time curve over 4-s intervals using the Origin software (MicroCal Software Inc., Northampton, MA). Data were graphed using the Origin software and are shown as mean ± one standard error (SE) of the number of experiments indicated. Significance was calculated using Student's t test.

RESULTS

Effect of Nitrate on Intracellular pH

The effect of external NO₃ on pH_i was evaluated microfluorimetrically in CHO cells loaded with BCECF. To facilitate the detection of NO₃-induced changes in pH_i, the contribution of other acid/base transporters, which might have a compensatory effect, was minimized. For this purpose, the initial experiments were performed in nominally HCO₃-free and Na⁺-free solutions, to minimize Cl/HCO₃ exchange and Na⁺-dependent acid/base transport. As shown in Fig. 1 A, superfusion of the cells with an NO₃-rich solution induced a sizable cytosolic acidification. The change in pH_i cannot be attributed to removal of external Cl, since substitution of Cl with equimolar gluconate did not significantly alter pH_i. This finding also implies that Cl/ HCO₃ exchange activity is negligible, since the alkalinization predicted to result from uptake of HCO₃, in exchange for exiting Cl, was not detectable. It is therefore unlikely that the NO₃-induced acidification results from exchange with intracellular HCO₃ via the anion exchanger.

It was important to ascertain that the NO₃-induced acidification of the cells is neither artifactual nor the result of a toxic effect. Cell morphology, which was monitored continuously using Hoffman optics, was not altered by addition of NO₃. Moreover, BCECF was retained by the cells throughout the observation period, attesting to the viability of the cells. Finally, the effect of external NO₃ was reversible since reperfusion with the Cl-rich medium induced recovery of pH_i (Fig. 1 A). These observations argue against a deleterious effect of NO₃.

In our experiments, pH_i is estimated from the ratio of BCECF fluorescence recorded at two excitation wavelengths. Differential quenching of fluorescence at one of these wavelengths by NO₃ could mimic the appearance of acidification. Several experimental approaches were used to rule out this potential artifact. We first compared the spectral properties of the free acid of BCECF in vitro in isotonic KCl or KNO₃ solutions, titrated to pH levels ranging from 5.84 to 7.56 with HCl or HNO₃. Excitation and emission spectra acquired in Cl and NO₃ solutions were identical (data not shown). That NO₃ does not alter the behavior of the dye inside cells was shown by clamping pH_i with 5 µM nigericin, a K⁺/H⁺ ionophore, in cells bathed in media containing 140 mM K⁺. Under these conditions the fluorescence ratio was unaffected by substitution of extracellular Cl for NO₃ (Fig. 1 B, right).

The effects of extracellular NO₃ on pH_i were also eliminated when the buffering capacity of the cytosol was greatly increased. This was accomplished by patch-clamping CHO cells in the whole-cell configuration with pipettes filled with a high buffer (50 mM HEPES, 50 mM Tris) solution. Microfluorimetric measurements of the individual patched cells revealed no change in pH_i when extracellular Cl was replaced by NO₃ and vice versa (Fig. 1 B, left). Together, these findings indicate that NO₃ does not artifactually alter the fluorescence of BCECF and indicate that this anion induces a bona fide change in pH_i.

Microfluorimetric measurements like those of Fig. 1 A represent the average pH_i of clusters of 6-12 cells. To assess whether the NO₃-induced cytosolic acidification occurs in all or most of the cells in the population and to further validate the microfluorimetric observations, the pH_i of individual cells was measured by ratio fluorescence imaging, as described in EXPERIMENTAL PROCEDURES. For these experiments, CHO cells were grown to submaximal confluence on glass coverslips, to facilitate the demarcation of individual cells, and were loaded with BCECF as described. Cells were perfused for 5 min in isotonic Cl or NO₃ solution before image acquisition, to allow adequate time for equilibration. As shown in Fig. 1 C, NO₃-induced cytosolic acidification occurred in virtually all the cells studied (n = 153 cells in Cl-rich solution and n = 174 in NO₃-rich solution). The mean pH_i in Cl solution was 7.55 ± 0.02, whereas 5 min after switching to NO₃, pH_i had decreased to 7.19 ± 0.02. These values were statistically different with P = 2.84 × 10³⁴ (Student's t test). It is noteworthy that the recording systems used for the imaging and photometry experiments are entirely different, indicating that the pH changes recorded are independent of the optical path, detector, and analysis software used.

We also tested whether other cell types also display the NO₃-induced changes in pH_i. The murine monocyte-macrophage cell line J774 was tested since, as detailed in the INTRODUCTION, NO₃ production is greatly enhanced in stimulated phagocytes (Miwa et al., 1987; Iyengar et al., 1987; Schmidt et al., 1989; Wright et al., 1989). When bathed in NO₃-rich media, J774 cells underwent a cytosolic acidification at a rate similar to that observed in CHO cells (Table I).

Table I. NO3-induced Acidification in Different Cell Types [View Table]

NO₃-induced Cytosolic Acidification Is Accompanied by NO₃ Uptake

NO₃ could induce the observed pH_i changes by acting on an extracellular receptor, by altering the transmembrane potential or by driving the transport or generation of acid equivalents as it enters the cell. To test whether the NO₃-induced intracellular acidification is accompanied by entry of the anion into the cells, the intracellular NO₃ content was measured using the Greiss method, after reduction of NO₃ to NO₂ (see EXPERIMENTAL PROCEDURES). For these experiments, CHO cells were incubated with isotonic NaNO₃ for 2-10 min at 37°C. Extracellular trapping was estimated by exposing cells momentarily to ice-cold NaNO₃ solution at 0°C (time = 0 in Fig. 2). Subtraction of this value from the individual determinations also accounted for any endogenous NO₃ or NO₂. Uptake of NO₃ by the cells was linear for at least 10 min. In cells incubated with 117 mM NO₃ the initial rate of entry of the anion, derived from linear regression, averaged 6.75 ± 0.15 mmol/liter cells/min (n = 3 at 2 min; n = 6 for all other time points; R = 0.995).

Exchange of NO₃ for Cl (Simchowitz, 1988; Simchowitz and Davis, 1989; Zhang and Solomon, 1992; Kurtz et al., 1994) has been documented to occur via the stilbene-sensitive anion exchanger (AE). To determine whether this process contributes to NO₃ uptake in CHO cells, measurements were also performed in the presence of 100 µM 4,4-diisothiocyanostilbene-2,2-disulfonate (DIDS), a concentration of the inhibitor that is expected to completely block the AE1 (Bruce et al., 1994a, b) and AE3 (Lee et al., 1991) isoforms and largely block the AE2 isoform of the exchanger (Simchowitz and Davis, 1989; Lee et al., 1991). NO₃ flux in DIDS-treated cells was 2.66 ± 0.07 mmol/liter cells/ min (n = 3 at 2 min; n = 6 for all other time points; R = 0.994). Thus 39% of the total NO₃ flux was insensitive to stilbenes and may be, at least in part, coupled to the translocation of H⁺ equivalents (see below).

Since the NO₃-induced pH_i changes were reversible, we tested the reversibility of the NO₃ fluxes. In three experiments, NO₃ content was measured in cells incubated with the anion for 15 min, followed by incubation in a NO₃-free (Cl-rich) medium for 10 min at 37°C. Upwards of 85% of the NO₃ taken up by the cells was lost during the washout period, implying that transport of the anion is bi-directional (data not shown).

Jointly, these experiments indicate that NO₃ is transported across the membrane during the course of the NO₃-induced pH_i changes. The flux of NO₃ may alter pH_i directly, by driving the transport of H⁺ equivalents across the membrane through a formerly unidentified pathway. Alternatively, the anion could conceivably modulate the activity of known acid/base transporters, such as Na⁺/H⁺ exchangers (NHE), anion exchangers or vacuolar-type H⁺ pumps (V-ATPases). These possibilities were considered experimentally below.

NO₃-induced Cytosolic Acidification Is Not Mediated by the Na⁺/H⁺ Antiporter

In erythrocytes, NHE has been reported to be inhibited by NO₃ (Parker, 1983; Parker and Castranova, 1984; Jennings et al., 1986). Because NHE is thought to contribute to the maintenance of the steady-state pH_i, inhibition of this transporter by NO₃ could conceivably result in a cytosolic acidification like that illustrated in Fig. 1 A. To test this hypothesis, we compared the effect of NO₃ on pH_i changes in two clones of CHO cells, the wild-type CHO cell (WT5) which expresses the NHE-1 isoform of the exchanger and the CHO mutant (AP1) which is devoid of NHE. As illustrated in Fig. 3 A, WT5 cells recover readily from an acid load upon addition of extracellular Na⁺. Such recovery, which is inhibited by amiloride and its analogues (not shown), is the hallmark of NHE activity. By contrast, AP1 cells failed to recover when Na⁺ was reintroduced to the perfusate, implying that they are devoid of NHE.

Despite the absence of NHE, AP1 cells exhibited a cytosolic acidification upon addition of external NO₃. The rate and extent of acidification were similar in the presence and absence of extracellular Na⁺ (Fig. 3 B). The mean NO₃- induced H⁺ flux was 2.54 ± 0.40 mmol/liter/min (n = 6) in Na⁺-rich solution and 2.22 ± 0.45 mmol/liter/min in NMG⁺-rich solution (n = 6). These observations imply that NHE is not essential for NO₃ to induce cytosolic pH changes.

Furthermore, NO₃-induced cytosolic acidification was observed in wild-type CHO cells bathed in the absence of external Na⁺ (mean NO₃- induced H⁺ flux in NMG⁺-rich solution was 2.39 ± 0.39 mmol/liter/min; n = 6). Under these conditions, forward NHE activity is abrogated and only backward exchange can occur, which can lead to cytosolic acidification. Inhibition of this process by NO₃ would be expected to have the converse effect, namely, cytosolic alkalinization. However, as shown in Fig. 3 C, NO₃ produced an acidification in WT5 cells that was indistinguishable from that noted in AP1 cells. Together, the data using ion substitution and genetic deletion of the antiporter indicated that the NO₃-induced cellular acidification is not mediated by inhibition of NHE.

NO₃-induced Cytosolic Acidification Is Not Mediated by the Anion Exchanger

The Cl/HCO₃ exchanger has affinity for other anions, including NO₃ (see Cabantchik and Greger, 1992; Reinertsen et al., 1989 for review), and NO₃/ HCO₃ exchange has been reported (Kemp and Boyd, 1993; Humphrey et al., 1994; Zhao et al., 1994). Though our experiments were conducted in the nominal absence of HCO₃, we cannot a priori exclude the possibility that exchange of cellular HCO₃ for external NO₃ accounts for the acidification, particularly considering the finding that DIDS inhibited a sizable fraction of NO₃ uptake. To address this possibility directly, the effect of NO₃ on pH_i was measured in the presence of DIDS. As shown in Fig. 4 A, the NO₃- induced acidification persisted when WT5 cells were treated with 100 µM DIDS. The H⁺_(equivalent) flux in control cells was 2.5 ± 0.5 mmol/liter/min and 2.7 ± 0.5 mmol/liter/min for DIDS treated cells (n = 4). No effect on the rate of cytosolic acidification was found using up to 1 mM of DIDS for 15 min (results not shown).

To further assess the role of the anion exchanger in the NO₃-induced pH changes, we tested a human embryonic kidney (HEK) cell line which has been reported to lack endogenous Cl/HCO₃ exchange activity (Lee et al., 1991). As shown in Fig. 4 B, perfusion of HEK cells with external NO₃ promoted a robust acidification, at a rate that was in fact greater than that observed in CHO cells (Table I). Jointly, these data suggest that the anion exchanger is not responsible for the NO₃-induced cytosolic acidification.

NO₃-induced Cytosolic Acidification Is Not Mediated by the V-ATPase

The V-ATPase, a H⁺-pump whose primary function is acidification of intracellular organelles (see Gluck, 1993 for review), also plays a role in the homeostasis of pH_i in some cell types (Swallow et al., 1991). In osteoclasts as well as in organelles studied in vitro, NO₃ has been reported to inhibit the V-ATPase (Chatterjee et al., 1993; Dschida and Bowman, 1995). Because the V-ATPase continuously removes H⁺ from the cytosol, inhibition of this enzyme by NO₃ could result in acidification. Two approaches were used to evaluate this possibility. In the first, the pump was inhibited by depleting cells of ATP by inhibiting glycolysis and oxidative phosphorylation using 2-deoxy-D-glucose and antimycin A. In CHO cells this results in depletion of >90% of the cellular ATP within 10 min (determined using luciferin-luciferase; Goss et al., 1994). As illustrated in Figs. 5, A and B, such ATP-depleted cells exhibited cytosolic acidification when exposed to extracellular NO₃. The mean H⁺ flux of ATP-depleted cells was comparable to that of control cells (2.38 ± 0.31 mmol/liter/min and 2.03 ± 0.42 mmol/liter/min, respectively; n = 4).

A second approach to evaluate the possible role of the V-ATPase used bafilomycin A₁, a macrolide antibiotic which is a potent and specific inhibitor of the pump (Crider et al., 1994; Zhang et al., 1994). In these experiments, WT5 cells were pre-incubated for 2 min in a Cl-rich, NO₃-poor solution, with 50 nM bafilomycin, a concentration shown previously to fully inhibit V-ATPases in a variety of cell types (Zhang et al., 1994; Crider et al., 1994). Subsequently, pH_i was microfluorimetric measured upon exposure of cells to isotonic NO₃-rich solution. Comparison of four experiments in bafilomycin-treated cells with their respective controls revealed no differences in H⁺_(equivalent) flux (1.46 ± 0.41 and 1.42 ± 0.32 mmol/liter/min in control and bafilomycin-treated cells, respectively; Fig. 5 B). Jointly, the results of the ATP depletion and bafilomycin studies suggest that the NO₃-induced cytosolic acidification is not mediated by the inhibition of the V-ATPase. Furthermore, the NO₃-induced cytosolic acidification appears to be an ATP-independent process.

Possible Role of NO₂ and Other Nitrogen Oxides in NO₃-induced Cytosolic Acidification

Solutions of NO₃ can contain small amounts of NO₂ and/or other nitrogen oxides. Of relevance, addition of NO₂ to pancreatic acinar cells was recently shown to induce a sizable cytosolic acidification (Zhao et al., 1994). This pH change was attributed, to a small extent, to the generation and permeation of HNO₂, a weak acid with pK_a of 3.2. The majority of the acidification, however, was seemingly due to the reaction between poorly defined nitrogen oxides (possibly NO, N₂O₃, and/or N₂O₄, intermediates in the oxidation of NO₂ to NO₃) and intracellular H₂O or OH. In acinar cells, the latter reaction was found to be catalyzed by carbonic anhydrase (Zhao et al., 1994). These observations raised the possibility that a similar mechanism might underlie the NO₂-induced acidification in CHO cells. Experiments designed to explore this possibility are illustrated in Fig. 6. The effect of NO₂ on pH_i was compared with that of acetate, another weak acid with somewhat higher pK (pK_a = 4.7). Acetate produced a rapid acidification, as expected from the permeation of the uncharged protonated species, acetic acid. An equimolar concentration of NO₂ induced a somewhat smaller and slower acidification (0.27 ± 0.05 pH U/min; n = 3), consistent with the >10-fold lower concentration of the protonated species. In both cases, the pH_i changes were rapidly reversed upon removal of the weak acids.

To determine whether nitrogen oxides contributed to the acidification by a carbonic anhydrase-mediated process, the cells were treated with methazolamide, an inhibitor of the anhydrase. A representative experiment is shown in Fig. 6 B. Neither the rate nor the extent of the NO₂ -induced acidification were noticeably altered by methazolamide (pH_i = 0.27 ± 0.05 in control cells and 0.29 ± 0.07 in cells treated with methazolamide; n = 3). Similarly, the more gradual acidification induced by NO₃ was unaffected by inhibition of carbonic anhydrase (pH_i = 0.11 ± 0.03 in control cells and 0.10 ± 0.04 in cells treated with methazolamide; n = 3). Thus, it appears that H⁺ release by reaction of nitrogen oxides with H₂O or OH is not an important component of the NO₂ or NO₃ response in CHO cells.

To ensure that the NO₃-induced acidification observed in CHO cells was not due to reduction of extracellular NO₃ to NO₂, the NO₂ content of solutions containing 29 to 117 mM NO₃ was measured using the Greiss method (as described in EXPERIMENTAL PROCEDURES). NO₂ content in the solutions was found to be negligible (mean [NO₂] = 0.44 ± 0.05 µM, n = 8). Similarly, intracellular NO₂ levels were found to be negligible up to 10 min after exposure to extracellular NO₃ (results not shown). Jointly, these results suggest that neither permeation of HNO₂ nor metabolism of contaminating nitrogen oxides are responsible for the acidification induced by NO₃.

Properties of the NO₃-induced H⁺ Flux

The results summarized above confirmed that the pH changes promoted by NO₃ are not due to modulation of the predominant homeostatic pathways and are accompanied by transport of NO₃ across the membrane. The simplest model to explain the effects of the anion is therefore the cotransport of NO₃ with H⁺, or its equivalent, NO₃/OH counter-transport. For simplicity, and by analogy with the system described in plants and fungi (Downey and Gedeon, 1994; Mehrag and Blatt, 1995), we will tentatively assume hereafter that NO₃-H⁺ cotransport is responsible for the observed pH_i changes.

The kinetic properties of the putative NO₃-H⁺ cotransport were investigated next. The extracellular [H⁺] dependence was determined in AP1 cells suspended in NaNO₃ solutions titrated between pH 6.0 to 7.5. Because reducing the extracellular pH (pH_o) is expected to reduce pH_i by pathways other than the NO₃-H⁺ cotransporter, the NO₃-independent component of the pH change was also measured by perfusing the cells in NaCl solution titrated to the appropriate pH. The NO₃-independent H⁺ flux at pH 6.0, 6.5, and 7.0 was 4.11 ± 0.88, 1.71 ± 0.93, and 1.36 ± 0.77 mmol/ liter/min, respectively. The NO₃-independent component was then subtracted from the total change recorded in the presence of NO₃ at an identical pH_o. The results of these determinations are illustrated in Fig. 7. Increasing external [H⁺] in the pH 7.5-6.0 range increased the rate of NO₃-induced cytosolic acidification, consistent with NO₃-H⁺ cotransport. More extreme levels of pH_o were not studied for fear of inducing cell damage. In the range studied, the half maximal rate of acidification was attained at [H⁺] = 0.09 ± 0.01 µM.

To assess the external NO₃ concentration ([NO₃]_o) dependence of the putative cotransporter, pH_i was measured in AP1 cells perfused with isotonic solutions containing 3.65-117 mM NO₃. These experiments were performed at both pH_o 6.5 and 7.5, and the solutions were osmotically balanced with gluconate since this anion was shown earlier to have no discernible effect on pH_i (Fig. 1 A). The rate of H⁺_(equivalent) flux increased with increasing external NO₃ concentration at both pH_o. The apparent NO₃ affinity and maximal velocity of the putative cotransporter can be inferred by fitting the data with an Eadie-Hofstee plot (Fig. 8). At pH_o 7.5, V_max and K_m were 5.81 ± 0.58 mmol/liter/min and 86.2 ± 12.6 mM, respectively (R = 0.98). At pH_o 6.5, V_max and K_m were 10.7 ± 1.3 mmol/ liter/min and 40.1 ± 10.6 mM, respectively (R = 0.97). Because the apparent affinity of the transporter for NO₃ is modified by [H⁺]_o, we conclude that binding of the ions to the transporter cannot occur independently in random order. The altered affinity is consistent with an allosteric effect of protons on the conformation of the NO₃ binding site(s).

Effect of Monovalent Cations on NO₃-induced Cytosolic Acidification

Some anion transporters are cation dependent (Roos and Boron, 1981; Schron et al., 1985; Shrode et al., 1995). To assess whether cationic species affect the rate of NO₃-induced cytosolic acidification, pH_i was measured in isotonic NO₃-rich solutions containing Na⁺, K⁺, or NMG⁺ as the principal cation. The cells were initially equilibrated with Cl-rich solutions of the same cationic composition and then switched to NO₃ media. AP1 cells were used for these measurements to eliminate possible confounding effects due to NHE. As summarized in Table II, no statistically significant differences in rate of H⁺ flux were observed in K⁺ or NMG⁺ solutions when compared with Na⁺ solutions.

Table II. Cationic Dependence of NO3-induced Acidification [View Table]

Voltage Sensitivity of the NO₃-induced Cytosolic Acidification

The DIDS-insensitive rate of NO₃ influx into AP1 cells was 2.66 ± 0.07 mmol/liter/min (Fig. 2). Under comparable experimental conditions, the rate of H⁺_(equivalent) flux was calculated to be 2.63 ± 0.99 mmol/liter/min (Table III). The apparent stoichiometry of the putative NO₃-H⁺ cotransporter derived from these flux rates is therefore one-to-one. This calculation assumes that all the stilbene-insensitive NO₃ flux is coupled to H⁺ transport, a premise that has not been validated experimentally. An alternative approach to estimate the stoi-chiometry of the cotransport process is to analyze its voltage sensitivity. An electroneutral one-to-one exchanger is likely to be voltage insensitive, whereas changes in voltage are more likely to affect a transporter with unequal stoichiometry.

Table III. Effects of Pharmacological Agents on the Rate of NO3-induced Cytosolic Acidification [View Table]

To gain further insight into the mechanism of NO₃-H⁺ cotransport, we evaluated its electrical properties in cells patch-clamped in the whole cell configuration. We estimated that it would be difficult to measure a current mediated by the transporter, since this was calculated to be as low as 4.2 or 8.4 pA if assuming a stoichiometry of 2:1 or 3:1, respectively. Instead, we assessed whether the NO₃-induced pH_i changes would be affected by drastic changes in the membrane potential. A pipette solution with low buffering capacity (1 mM MES) was used to maximize the NO₃-induced pH_i changes. A representative pH_i measurement in a voltage-clamped cell is shown in Fig. 9. The cell was initially clamped at 60 mV and superfused with a Cl-rich solution. Under these conditions, pH_i was unaffected by a sudden depolarization to 0 mV, confirming that CHO cells have a comparatively low H⁺_(equivalent) conductance at physiological pH_i and at normal resting membrane potential (Demaurex et al., 1995). Cytosolic acidification was observed when NO₃ replaced Cl in the bathing solution. The occurrence of an acidification in voltage clamped cells implies that the effect of NO₃ on pH_i is not mediated by, and does not require changes in membrane potential. Moreover, the rate of pH_i change was not affected by sequentially stepping the membrane potential up from 60 to +40 mV; only the normal exponential decay was noted. A comparable decay was observed when the order of the voltage changes was reversed, from depolarized to repolarized (not illustrated). The data in patch clamped cells indicate that the NO₃-induced cytosolic acidification is voltage insensitive, consistent with an electrically neutral process. This conclusion is compatible with the tentative estimates of stoichiometry of one-to-one.

Effect of Inhibitors on the NO₃-induced Cytosolic Acidification

To establish possible analogies with other mammalian transporters, we tested a variety of compounds known to inhibit other plasmalemmal ion transport systems. Ethacrynic acid (Poole and Halestrap, 1993; Koechel, 1981; Palfrey and Leung, 1993), a dichlorophenoxyacetic acid derivative, is a potent alkylating reagent that has loop diuretic action. It is known to inhibit several ion carriers, including the Na⁺/K⁺/Cl cotransporter (Palfrey and Leung, 1993). NO₃-induced cytosolic acidification was 67% inhibited by 100 µM ethacrynic acid (Table III). The mean H⁺_(equivalent) flux in ethacrynic acid treated-cells was 1.01 ± 0.48 mmol/liter cells/min (n = 4) compared with 3.06 ± 0.43 mmol/ liter cells/min in paired controls (n = 19). Inhibition by ethacrynic acid was almost complete at 200 µM (not shown).

-Cyano-4-hydroxycinnamate (CHC) inhibits both the anion exchanger (Simchowitz, 1988) and the monocarboxylate transporter (Poole and Halestrap, 1993) in other cells. CHC (1 mM) inhibited NO₃- induced cytosolic acidification to 20% of the control rate (Table III). H⁺_(equivalent) flux in CHC-treated cells was 0.62 ± 0.49 mmol/liter cells/min (n = 4).

Phloretin, the aglycone of phlorizin, is a reversible, relatively nonspecific inhibitor of several membrane transport processes, including urea transport, monocarboxylate transport and Cl/HCO₃ exchange (Wang et al., 1993; Melnik et al., 1977; Chou and Knepper, 1989). In our experiments, 100 µM phloretin was found to have no effect on the rate of NO₃-induced cytosolic acidification (Table III).

pCMBS, an organomercurial sulfhydryl reagent, irreversibly inhibits the monocarboxylate transporter (Munzel et al., 1995) but does not affect the Cl/HCO₃ exchanger (Poole and Halestrap, 1993; Zhang and Solomon, 1992). NO₃-induced cytosolic acidification was unaffected by 250 µM pCMBS (Table III). Similarly, NO₃-induced cytosolic acidification was not affected by furosemide (data not shown), a loop diuretic that inhibits several anion transport systems, including the Cl/HCO₃ exchanger (Cabantchik et al., 1978), the monocarboxylate transporter (Poole and Halestrap, 1993) and the SO₄²/OH exchanger (Schron et al., 1985) which is reportedly distinct from the SO₄²/Cl