INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

OXALATE IS A METABOLIC END product that is excreted almost exclusively in the urine, principally through glomerular filtration and net secretion in the proximal tubule (4, 16). At least four separate oxalate transporters have been described in the proximal tubule: a sulfate/oxalate exchanger, chloride/oxalate exchanger and OH/oxalate exchanger on the apical membrane (7, 10), and a sulfate/oxalate exchanger on the basolateral membrane (9).

The addition of physiological concentrations of oxalate to microperfused tubules leads to a stimulation of volume absorption by 57% in the proximal tubule and 115% in the distal tubule (18, 20). This effect is abolished by the anion exchange inhibitor 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) and, in the proximal tubule, is dependent on the presence of sulfate (19) while resistant to the effects of Na/H exchange inhibitors. These results support the hypothesis that oxalate participates in NaCl reabsorption along the nephron (7). In the proximal tubule, oxalate probably stimulates salt reabsorption through parallel oxalate/chloride exchange and sodium-sulfate cotransport on the apical membrane (19). To maintain levels of intracellular oxalate sufficient to serve as the driving force for chloride, several methods of recycling oxalate across the apical membrane have been proposed, including OH/oxalate exchange, formate/oxalate exchange as a mode of transport on the chloride/oxalate exchanger, and or /oxalate exchange as separate modes of transport on the sulfate/oxalate exchanger (1, 7, 10, 19). The sulfate dependence of oxalate-stimulated salt reabsorption in the proximal tubule suggests that the sulfate/oxalate exchanger is involved in the recycling process; however, the contribution of each oxalate exchanger in maintaining salt reabsorption along the nephron remains to be determined. The presence of multiple oxalate exchangers with overlapping substrate specificity has complicated attempts to define the substrates and clarify whether the different modes of transport represent anion exchange on one or more than one transporter.

The purpose of this study was to functionally isolate the different oxalate exchangers from the proximal tubule so that each may be examined in the absence of alternate transport pathways. In this study, two distinct oxalate transporters have been solubilized and reconstituted from rabbit renal microvillus membranes. A third oxalate transporter has been solubilized and reconstituted from rabbit renal cortex basolateral membranes. Because tubular chloride reabsorption is dependent on both sulfate and oxalate, we examined the interaction of sulfate and chloride on each of the three reconstituted oxalate transporters.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Membrane isolation. Microvillus membrane vesicles were isolated from renal cortices of male New Zealand White rabbits (2-3 kg), using the Mg²⁺ aggregation method described previously (7). Average enrichment in the apical membrane marker -glutamyltranspeptidase was 11.1-fold (n = 13). In some experiments, standard microvillus membranes were purified further by fractionation on a sucrose gradient (7). Enrichment of fractionated membranes for apical and basolateral markers are described in RESULTS. Basolateral membranes were purified from New Zealand White rabbit renal cortices on a Percoll gradient as previously described (3). Average enrichment of the basolateral membrane marker Na⁺-K⁺-adenosinetriphosphatase (Na⁺-K⁺-ATPase) (2) was 15.4-fold (n = 5). Protein concentrations were determined by the method of Lowry et al. (11) as modified by Peterson (13), using bovine serum albumin as the standard. Purified apical and basolateral membranes were resuspended in 80 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), 40 mM potassium hydroxide (KOH), and 150 mM mannitol, pH 7.5, and stored at 70°C.

Solubilization of apical and basolateral membranes. Either apical or basolateral membranes, at a protein concentration of 9 mg/ml, were solubilized in a buffer consisting of 10 mM potassium phosphate, pH 7.2, 5 mM potassium oxalate, 1 mM dithiothreitol, 2% octylglucoside (wt/vol), and 20% glycerol (vol/vol). The mixture was incubated on ice for 15 min then centrifuged at 100,000 g for 30 min. The supernatant was removed and either used immediately or stored at 70°C. No difference has been identified between freshly prepared and frozen solubilized membranes. In some experiments, a mixture of protease inhibitors was included during membrane solubilization (1.0 µg/ml aprotinin, 50 µg/ml antipain, 0.5 µg/ml leupeptin, 0.5 mg/ml Pefabloc SC, and 1.0 mM disodium EDTA). No difference in activity was observed in the presence or absence of the protease inhibitors.

Reconstitution of apical and basolateral membranes. Lipid for reconstitution was prepared by dissolving 48 mg soy extract and 16 mg cholesterol in 600 µl chloroform. The lipid/cholesterol mixture was dried under a stream of argon, followed by desiccation under vacuum for at least 1 h. The dried lipid was resuspended in 800 µl buffer consisting of 60 mM HEPES/tetramethyammonium hydroxide (HEPES/TMA-OH), pH 7.4, 30 mM potassium oxalate, and 1 mM dithiothreitol. The resuspended lipid was sonicated in a bath-type sonicator at room temperature until it was semitranslucent (~5 min). Six hundred microliters of the sonicated lipid was added to 200 µl 8% octylglucoside and centrifuged at 3,600 g for 5 min. The supernatant was removed and used for reconstitution. Solubilized membrane protein was added to the lipid/cholesterol mixture so that the lipid/cholesterol was always 35% of the final reconstitution volume. A solution consisting of 10 mM potassium phosphate, pH 7.2, 5 mM potassium oxalate, 1 mM dithiothreitol, 1.6% octylglucoside (wt/vol), and 8 mg/ml soy extract (AP-10 buffer) was used to achieve the appropriate dilution of the membrane protein prior to adding to lipid/cholesterol. The protein/lipid/cholesterol mixture was dialyzed overnight at 5°C using a Spectrum Biotech membrane (8,000 M_r cutoff) against a dialysis buffer consisting of 60 mM HEPES/TMA-OH, pH 7.4, and 1 mM dithiothreitol, plus a counter anion to drive isotope uptake. The concentration of the counter ion is listed in the corresponding figure legend.

Measurement of reconstituted transport activity. For anion exchange, the proteoliposomes formed by dialysis were applied to a 1.0-ml anion exchange column of Bio-Beads (AG 1 × 4, 50-100 mesh) that had first been equilibrated with fluoride according to the manufacturer's instructions. The counter ion that remains outside the proteoliposomes following dialysis is exchanged for fluoride, thereby providing an outwardly directed gradient of the unlabeled anion (usually either oxalate or chloride). In this way, the uptake of isotope can be measured without carryover of the counter ion into the uptake medium. The proteoliposomes are eluted from the anion exchange column in a buffer consisting of 84 mM HEPES/KOH, pH 7.4, plus 25 mM KF (HKF buffer). To initiate anion exchange, proteoliposomes eluted from the anion exchange column (44 µl) are added to 22 µl of uptake buffer containing the appropriate isotope. After the specified time interval, the mixture is passed over a second anion exchange column to remove isotope not taken up by the proteoliposomes. The proteoliposomes are eluted from the column in 1 ml HKF buffer, added to scintillation cocktail (RPI, 3a7B), and the intraliposomal radioisotope activity is measured by scintillation spectroscopy. To determine that isotope uptake is protein mediated, parallel studies were performed in liposomes made in an identical manner except for the absence of solubilized membrane protein. To determine the contribution of unincorporated isotope eluting from the anion exchange columns, isotope was added to HKF buffer and was run on parallel columns in the absence of proteoliposomes. Counts obtained by nonspecific elution were subtracted from the experimental conditions.

Hydroxyapatite chromatography. A 1-ml hydroxyapatite column (Econo-Pac CHT-II cartridge, Bio-Rad) was equilibrated with 3 ml AP-10 buffer. Approximately 2 mg solubilized microvillus membrane protein was mixed with stock solutions of soy extract and octylglucoside to achieve a final concentration of 8 mg/ml soy extract, 1.6% octylglucoside (wt/vol) in 10 mM potassium phosphate, pH 7.2, 5 mM potassium oxalate, and 1 mM dithiothreitol. The protein solution was applied to the hydroxyapatite column and eluted with a linear 6 ml, 10-200 mM potassium phosphate gradient, pH 6.8, in 8 mg/ml soy extract, 1.6% octylglucoside (wt/vol), 5 mM potassium oxalate, and 1 mM dithiothreitol. All solutions were prefiltered prior to use on the HP column (Millex-HA, 0.45 µm; Millipore). Individual fractions were collected (450 µl) and mixed with lipid/cholesterol. After overnight dialysis, the proteoliposomes formed from each fraction were tested for anion exchange activity.

Material. Octylglucoside ULTROL grade was purchased from Calbiochem. Soy extract was from Avanti Polar Lipids. [¹⁴C]oxalic acid was from Sigma. [³⁵S]-labeled sulfuric acid and glycerol (high-purity grade) were purchased from ICN. Dithiothreitol and Pefabloc SC were from Boehringer Mannheim.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

The first set of experiments were designed to test the feasibility of solubilizing and reconstituting a functional oxalate transporter from rabbit renal microvillus membrane vesicles. An outwardly directed gradient of unlabeled oxalate was imposed across the proteoliposomes to serve as the driving force for the uptake of [¹⁴C]oxalate. In preliminary studies, solubilization and reconstitution was attempted with different concentrations of various detergents, including Triton X-100, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS), octylglucoside, and deoxycholate. Under the conditions tested, octylglucoside was the only detergent where functional reconstitution was observed (see below) and was subsequently used in the remainder of the studies.

When 60 µg of solubilized microvillus membrane protein is reconstituted in the presence of an outwardly directed oxalate gradient, [¹⁴C]oxalate uptake is driven above its equilibrium value at 5 and 10 min time points (Fig. 1). In the presence of the anion exchange inhibitor DIDS (0.5 mM), the uptake of [¹⁴C]oxalate is nearly abolished. Decreasing the concentration of solubilized protein from 60 to 30 µg/mg lipid decreased the rate of oxalate/oxalate exchange activity. When proteoliposomes were formed in the absence of an outwardly directed oxalate gradient by substituting either fluoride or gluconate for oxalate, the rate of [¹⁴C]oxalate uptake was similar to the results shown in Fig. 1 for uptakes performed in the presence of DIDS. In addition, [¹⁴C]oxalate uptake in liposomes made in the presence of an outwardly directed oxalate gradient but in the absence of solubilized protein was negligible (0.1 ± 0.4 pmol · mg lipid¹ · min¹). Taken together, the observations that [¹⁴C]oxalate uptake in proteoliposomes is dependent on the presence of solubilized membrane protein, is stimulated by an outwardly directed gradient of oxalate, and is inhibited by DIDS demonstrate that we have successfully reconstituted an oxalate exchanger from microvillus membrane vesicles. In the remainder of the studies, oxalate transport is defined as isotope uptake in the absence of DIDS minus isotope uptake in the presence of 0.5 mM DIDS.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Effect of an outward oxalate gradient on oxalate uptake. Either 60 or 30 µg of solubilized microvillus membrane protein/mg lipid were reconstituted in the presence of 80 mM HEPES/KOH, 25 mM potassium oxalate, and 1 mM dithiothreitol, pH 7.4. Uptake of 19 µM [14C]oxalate was measured at various time points in uptake buffer consisting of 72 mM HEPES/KOH, 45 mM potassium fluoride, pH 7.4, in the presence or absence of 0.5 mM 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS). Results are means ± SE of 4 experiments.

To identify the nature of the reconstituted oxalate exchanger, oxalate-stimulated [¹⁴C]oxalate uptake was measured in the presence of various anions added to the uptake medium. Anions that are substrates for the exchanger would be expected to cis-inhibit the uptake of [¹⁴C]oxalate. To avoid the possibility that reconstituted oxalate transport is the result of basolateral membrane contamination, microvillus membranes were prepared in the standard manner and then fractionated on a sucrose gradient prior to solubilization (7). Individual fractions from the sucrose gradient were tested for -glutamyl transpeptidase and ouabain-inhibitable Na⁺-K⁺-ATPase activity, markers of apical and basolateral membranes, respectively. Sucrose-gradient fractions with minimal basolateral membrane marker activity were pooled for solubilization and reconstitution. In 18 consecutive microvillus membrane preparations that were further purified by sucrose-gradient fractionation, ouabain-inhibitable Na⁺-K⁺-ATPase activity was barely detectable when measured as inorganic phosphate release at 705 nm absorbance (0.04 OD/mg protein). This represents an 80 ± 4% reduction of basolateral membrane Na⁺-K⁺-ATPase activity compared with the standard microvillus membrane preparation and demonstrates the nearly complete removal of any basolateral membrane contamination from the sucrose-purified microvillus membranes. -Glutamyl transpeptidase activity was not statistically different between the sucrose-purified membranes and the standard membranes, confirming that basolateral membranes constitute only a minor component of the standard microvillus membrane preparation. Unless indicated otherwise, sucrose-purified membranes were used in the remainder of the studies to examine apical oxalate transport.

Figure 2 shows the ability of various anions to inhibit oxalate/[¹⁴C]oxalate exchange in proteoliposomes reconstituted from sucrose-purified apical membranes. To minimize the generation of voltage gradients across the membrane that might influence [¹⁴C]oxalate uptake, these experiments were performed under voltage-clamped conditions (K_i = K_o in the presence of valinomycin). In separate experiments, we have not found any change in oxalate-stimulated [¹⁴C]oxalate uptake in the presence or absence of 50 mM fluoride (0.7 ± 0.19 vs. 0.7 ± 0.25 nmol · mg protein¹ · min¹). Therefore, the data are presented as the percent of inhibition compared with fluoride as the control.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Cis-inhibition of oxalate/oxalate exchange. Sucrose gradient-purified apical membranes were solubilized and reconstituted at 76 µg/mg lipid in 60 mM HEPES/tetramethylammonium hydroxide (HEPES/TMA-OH), 30 mM potassium oxalate, 1 mM dithiothreitol, pH 7.4. Three-minute uptake of 317 µM [14C]oxalate was determined in uptake buffer consisting of 84 mM HEPES/KOH, pH 7.4, 25 mM potassium fluoride (control), either 8 mM of KCl, potassium formate, or potassium acetate, or 4 mM of potassium oxalate or potassium sulfate substituted for 8 mM potassium fluoride. DIDS was used at a concentration at 0.5 mM. Valinomycin (0.34 mg/ml) was added to the proteoliposomes prior to initiating uptakes. Results are means ± SE of 4 separate experiments. Different membrane preparations were used for each experiment, with uptakes performed in duplicate.

The order of potency for cis-inhibition of oxalate/oxalate exchange is DIDS > oxalate > chloride > formate  acetate. This is consistent with the reconstitution of the apical membrane chloride/oxalate exchanger, which accepts chloride, oxalate, and formate, but not acetate, as substrates. The interaction of sulfate with the reconstituted oxalate transporter is more complex. As evident in Fig. 2, sulfate is able to inhibit oxalate-stimulated [¹⁴C]oxalate uptake by 27%, suggesting that, in addition to chloride and formate, sulfate is a substrate for the reconstituted oxalate exchanger. However, previous studies have been unable to identify appreciable amounts of either chloride/³⁵ exchange or sulfate/³⁶Cl exchange in rabbit renal microvillus membrane vesicles (7, 10). One possible explanation for the results in Fig. 2 is that the reconstituted oxalate/oxalate exchange activity represents oxalate transport on two separate exchangers. A sulfate/bicarbonate exchanger has been observed in bovine, rat, and rabbit microvillus membrane vesicles (10, 14, 17). The ability of oxalate to serve as a substrate on the sulfate/bicarbonate exchanger has been described in the rabbit (10).

To confirm that chloride and sulfate can serve as substrates for reconstituted oxalate exchange, we measured both chloride-stimulated [¹⁴C]oxalate uptake and oxalate-stimulated ³⁵ uptake under voltage-clamped conditions. As shown in the time courses presented in Fig. 3, [¹⁴C]oxalate uptake in the presence of an outwardly directed chloride gradient and ³⁵ uptake in the presence of an outwardly directed oxalate gradient are nearly abolished by DIDS. When experiments were performed with liposomes made in the absence of solubilized protein, chloride-stimulated [¹⁴C]oxalate uptake and oxalate-stimulated ³⁵ uptake are not observed (data not shown), confirming that the reconstitution of solubilized apical membrane protein is required for anion exchange.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effect of outward chloride and oxalate gradients on [14C]oxalate and [35S]sulfate. For chloride-stimulated [14C]oxalate uptake, microvillus membranes were fractionated on sucrose gradients, solubilized, and reconstituted at 76 µg/mg lipid in 60 mM HEPES/TMA-OH, 60 mM KCl, and 1 mM dithiothreitol, pH 7.4. Uptake of [14C]oxalate (317 µM) was measured in uptake buffer consisting of 84 mM HEPES/KOH, pH 7.4, and 25 mM potassium fluoride, plus or minus 0.5 mM DIDS. For oxalate-stimulated [35S]sulfate uptake, solubilized membranes were reconstituted at 60 µg/mg lipid in 60 mM HEPES/TMA-OH, 30 mM potassium oxalate, and 1 mM dithiothreitol, pH 7.4. Uptake of 50 µM [35S]sulfate was measured in uptake buffer consisting of 84 mM HEPES/KOH, pH 7.4, and 25 mM potassium fluoride, plus or minus 0.5 mM DIDS. Valinomycin (0.4 mg/ml) was added to proteoliposomes prior to uptakes. Results are means of 3 separate experiments from 3 different membrane preparations.

To differentiate between the presence of multiple oxalate transporters or a single transporter that accepts sulfate and chloride as substrates, we compared the sensitivities of reconstituted chloride/[¹⁴C]oxalate exchange and oxalate/³⁵ exchange to inhibition by chloride, sulfate, and the anion exchange inhibitor, phenol red (12). The results of these experiments are shown in Fig. 4 and demonstrate contrasting patterns of inhibition between the two different modes of oxalate exchange. Whereas oxalate/³⁵ exchange is nearly abolished by unlabeled sulfate and is stimulated slightly by chloride, chloride/[¹⁴C]oxalate exchange is minimally affected by external sulfate and inhibited 52% by unlabeled chloride. In addition, oxalate/³⁵ exchange is inhibited 69% by phenol red, whereas chloride/[¹⁴C]oxalate exchange is inhibited only 41%.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Inhibition of oxalate/sulfate exchange and chloride/oxalate exchange in reconstituted, sucrose-purified apical membranes. Microvillus membranes were fractionated on a sucrose gradient, solubilized, and reconstituted in 60 mM HEPES/TMA-OH and 1 mM dithiothreitol, pH 7.4, plus either 30 mM potassium oxalate or 60 mM KCl. Three-minute uptakes of either 50 µM [35S]sulfate in the presence of an outward directed oxalate gradient (n = 5) or 317 µM [14C]oxalate in the presence of an outwardly directed chloride gradient (n = 4) were determined in uptake buffer consisting of 84 mM HEPES/KOH, pH 7.4, and 25 mM potassium fluoride as the control or with the substitution of either 8 mM KCl or 4 mM K2SO4 for 8 mM potassium fluoride. Final concentration of phenol red was 0.5 mM in uptake buffer. Valinomycin (0.4 mg/ml) was added to proteoliposomes prior to uptakes. Data are percentage of remaining activity compared with control. Results are means ± SE. Different membrane preparations were used for each experiment, with uptakes performed in duplicate.

The results in Fig. 4 are most consistent with the reconstitution of two distinct oxalate transporters from apical membranes. One probably represents the apical sulfate/anion exchanger, which is similar to the basolateral sulfate/anion transporter in that it is inhibited by phenol red and DIDS and accepts sulfate, bicarbonate (or carbonate), and oxalate as substrates. The other is the DIDS-inhibitable chloride/oxalate exchanger, which accepts chloride, formate, and oxalate as substrates. To address whether sulfate is a substrate for the chloride/oxalate exchanger or whether chloride has an affinity for the sulfate/oxalate exchanger, we separated the individual transporters by hydroxyapatite chromatography. Individual fractions eluting from the hydroxyapatite column were reconstituted and tested for oxalate/oxalate exchange. The results shown in Fig. 5 represent an elution profile of a typical experiment. Peak oxalate exchange activity was observed in both early and late fractions, with minimal activity detected in the remaining fractions. The higher level of activity in the later peak as illustrated in Fig. 5 was typical for these experiments.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Oxalate/[14C]oxalate exchange activity eluted from hydroxyapatite column. Individual fractions (450 µl) were eluted from an hydroxyapatite column and reconstituted in 60 mM HEPES/TMA-OH, 30 mM potassium oxalate, and 1 mM dithiothreitol, pH 7.4. An aliquot was removed from each fraction prior to dialysis for protein analysis. Five-minute uptakes of [14C]oxalate from individual fractions were measured in uptake buffer consisting of 84 mM HEPES/KOH, pH 7.4, and 25 mM potassium fluoride, plus or minus 0.5 mM DIDS. Uptakes were performed in duplicate. Results are expressed as [14C]oxalate uptake in the absence of DIDS minus [14C]oxalate uptake in the presence of DIDS for each fraction (solid line) and as the total amount of protein measured in each fraction (dotted line). Results are from a representative experiment.

To determine whether the peaks in oxalate exchange activity represent the presence of two different oxalate exchangers, we separated solubilized apical membrane protein on an hydroxyapatite column and pooled the fractions containing the highest levels of activity into early and late peak fractions. After reconstitution, we tested the effects of unlabeled sulfate and chloride on oxalate-stimulate [¹⁴C]oxalate uptake. As illustrated in Fig. 6, sulfate nearly abolished oxalate/oxalate exchange activity in the early fraction but inhibits oxalate exchange only 41% in the late fraction. In contrast, chloride inhibits oxalate/oxalate exchange activity 70% in the late fraction but only 32% in the early fraction. Furthermore, in the early fraction, there was a 40-fold increase of oxalate-stimulated ³⁵ uptake in the absence of DIDS compared with the presence of DIDS, whereas oxalate/³⁵ exchange is undetected in the late hydroxyapatite fraction (data not shown). These results are consistent with the elution of the sulfate/oxalate exchanger in the early fraction and the chloride/oxalate exchanger in the later fraction, confirming the presence of two distinct oxalate transport proteins in renal microvillus membranes.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Effect of chloride and sulfate on oxalate/[14C]oxalate exchange in early and late hydroxyapatite fractions. Fractions eluted from a 1-ml hydroxyapatite column that contained significant oxalate/oxalate exchange activity were pooled into early and late activity fractions following overnight dialysis in 60 mM HEPES/TMA-OH, 30 mM potassium oxalate, and 1 mM dithiothreitol, pH 7.4. Five-minute uptakes of 317 µM [14C]oxalate in the presence of an outwardly directed oxalate gradient were determined in uptake buffer consisting of 84 mM HEPES/KOH, pH 7.4, and 25 mM potassium fluoride, plus or minus 0.5 mM DIDS as the control or with the substitution of either 20 mM KCl or 10 mM K2SO4 for 20 mM potassium fluoride. Valinomycin (0.3 mg/ml) was added to proteoliposomes prior to uptakes. Results are percentage of remaining DIDS-inhibitable activity compared with the control and are means ± SE of 4 separate experiments performed using different membrane preparations.

We next examined the feasibility of solubilizing and reconstituting the sulfate/oxalate exchanger from the basolateral membrane to compare it with the apical sulfate/oxalate exchanger. Basolateral membranes were isolated from rabbit renal cortex, solubilized with octylglucoside, and reconstituted into proteoliposomes. An outwardly directed oxalate gradient was imposed, and the uptake of ³⁵ was measured at various time points in the presence and absence of 0.5 mM DIDS. As shown in Fig. 7, there is a significant stimulation of sulfate uptake in the absence of DIDS compared with uptake in the presence of DIDS. The pattern of cis-inhibition by sulfate, chloride, and phenol red on the reconstituted basolateral oxalate/sulfate exchanger is shown in Fig. 8. Nearly complete inhibition of oxalate-stimulated ³⁵ uptake is observed in the presence of 4 mM sulfate or 0.5 mM phenol red, whereas there is a slight stimulation of sulfate uptake when 8 mM chloride is included in the uptake medium. Taken together, these results are most consistent with the reconstitution of the sulfate/oxalate exchanger from the basolateral membrane. Furthermore, in the presence of an outwardly directed chloride gradient, DIDS-inhibitable [¹⁴C]oxalate uptake was undetectable in proteoliposomes reconstituted with basolateral membrane protein (data not shown), demonstrating that chloride/oxalate exchange is not a mode of transport on the basolateral sulfate/oxalate exchanger.

View larger version (11K): [in this window] [in a new window]   Fig. 7.   Effect of an outwardly directed oxalate gradient on [35S]sulfate uptake in reconstituted basolateral membranes. Basolateral membranes were solubilized and reconstituted in 60 mM HEPES/TMA-OH, 30 mM potassium oxalate, and 1 mM dithiothreitol, pH 7.4. Uptake of 50 µM [35S]sulfate was measured in uptake buffer consisting of 84 mM HEPES/KOH, pH 7.4, and 25 mM potassium fluoride, plus or minus 0.5 mM DIDS. Valinomycin (0.4 mg/ml) was added to proteoliposomes prior to uptakes. Results are means ± SE of 3 separate experiments performed in duplicate.

View larger version (9K): [in this window] [in a new window]   Fig. 8.   Cis-inhibition of oxalate/sulfate exchange in reconstituted basolateral membranes. Solubilized basolateral membranes were reconstituted in 60 mM HEPES/TMA-OH, 30 mM potassium oxalate, and 1 mM dithiothreitol, pH 7.4 (65-75 µg protein/mg lipid). Three-minute uptake of [35S]sulfate was measured in uptake buffer consisting of 84 mM HEPES/KOH, pH 7.4, and 25 mM potassium fluoride as the control or with substitution of either 8 mM KCl or 4 mM K2SO4 for 8 mM potassium fluoride. Final concentration of phenol red was 0.5 mM in uptake buffer. Valinomycin (0.4 mg/ml) was added to proteoliposomes prior to uptakes. Data are percentage of activity compared with control. Results are expressed as means ± SE of 4 separate experiments. Different membrane preparations were used for each experiment, with uptakes performed in duplicate.

In preliminary studies, we have observed that at high detergent-to-protein ratios, oxalate transport activity is inactivated and that the rate of inactivation increases at elevated temperatures (22 vs. 5°C). The addition of protease inhibitors during membrane solubilization had no effect on the rate of inactivation. Detergents have the capacity to inactivate membrane transport proteins (5), and some investigators have noted that inactivation can be prevented by the addition of exogenous lipid (6, 21). Similarly, we found that the inactivation of oxalate transport activity by octylglucoside can be prevented by the addition of exogenous lipid, as long as the lipid-to-detergent ratio (wt/wt) is at least 5:1. Presumably, at this lipid-to-detergent ratio, the interaction of protein with lipid in the mixed micelles preserves the native protein structure and maintains transport function. In the next experiment, we compared the rate of detergent inactivation of oxalate/sulfate exchange on apical and basolateral membranes. For these experiments, basolateral membranes and sucrose-purified apical membranes were solubilized in the standard manner at 5°C. The solubilized membrane proteins were then either added directly to the lipid/cholesterol mixture (zero time point) or diluted to an octylglucoside-to-protein ratio of 11:1 (wt/wt). The octylglucoside/protein mixture was allowed to incubate at 22°C for various lengths of time and then added to lipid/cholesterol to quench the detergent-induced inactivation. After the formation of proteoliposomes by dialysis, oxalate-stimulated uptake of ³⁵ was determined. As illustrated in Fig. 9, the rate of inactivation of oxalate/sulfate exchange was significantly faster in apical membranes compared with basolateral membranes.

View larger version (13K): [in this window] [in a new window]   Fig. 9.   Effect of excess detergent on oxalate/sulfate exchange from apical and basolateral membranes. Basolateral membranes and sucrose gradient-purified apical membranes were solubilized with octylglucoside at a detergent-to-protein ratio of 2:1 as described in METHODS. Octylglucoside was then added to a detergent-to-protein ratio of 11:1, and solubilized protein was incubated at 22°C. At indicated time points, aliquots of 180 µg solubilized protein were mixed with 2.4 mg lipid/cholesterol to quench inactivation by detergent. For controls, 180 µg solubilized protein was first mixed with lipid/cholesterol to protect against detergent inactivation, then octylglucoside was added to a detergent-to-protein ratio of 11:1. After overnight dialysis to form proteoliposomes, the uptake of [35S]sulfate was measured under the conditions described in Fig. 6. Results are means ± SE of 5 separate experiments and are depicted as percentage of remaining DIDS-inhibitable activity compared with controls. Different membrane preparations were used for each experiment, with uptakes performed in duplicate. * P < 0.05, significant difference from basolateral values.

It has been observed that the addition of substrate may preserve enzyme activity during solubilization (8). In the experiment shown in Fig. 9, inactivation proceeded, despite the presence of oxalate during exposure to detergent; however, a very high detergent-to-protein ratio was used in that experiment as evident by the complete loss of activity by 5 min. At these levels of detergent, any beneficial effects of substrate may not be observed. Therefore, in the following experiment, we examined whether the presence of oxalate or chloride during solubilization could preserve transport activity at a lower detergent concentration. Microvillus membranes were incubated for 60 min in octylglucoside (detergent-to-protein ratio, 3:1) at 22°C in the presence or absence of either 10 mM oxalate or 40 mM chloride. Solubilized membranes added directly to lipid to quench detergent inactivation served as zero time points. Under the conditions employed for this experiment, exposure to octylglucoside for 60 min in the absence of both chloride and oxalate reduces oxalate transport activity by 67% (Fig. 10). However, when 10 mM oxalate is present during incubation, transport activity is preserved. In contrast, chloride is unable to protect against inactivation by detergent.

View larger version (16K): [in this window] [in a new window]   Fig. 10.   Effect of oxalate or chloride on detergent inactivation of oxalate/oxalate exchange. Solubilized microvillus membrane protein was mixed with octylglucoside at a detergent-to-protein ratio of 3:1 in solubilization buffer in the presence or absence of either 10 mM potassium oxalate or 40 mM KCl. After incubation at 22°C for 1 h, 180 µg solubilized protein were diluted 4-fold with AP-10 buffer and then added to 3 mg lipid/cholesterol, and dialyzed overnight to form proteoliposomes in 60 mM HEPES/TMA-OH, 30 mM potassium oxalate, and 1 mM dithiothreitol, pH 7.4. For zero time points, solubilized protein was added directly to AP-10 buffer and the lipid/cholesterol mixture, without incubation for 60 min in excess detergent. Five-minute uptake of 317 µM [14C]oxalate was determined in uptake buffer consisting of 84 mM HEPES, 25 mM fluoride, and 60 mM potassium, pH 7.4, plus or minus 0.5 mM DIDS. Results are means ± SE of 4 separate experiments and are expressed as DIDS-inhibitable oxalate uptake. Different membrane preparations were used for each experiment, with uptakes performed in duplicate. * P < 0.05, significant difference from zero time points.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In the mammalian proximal tubule, three separate anion exchangers involved in the transport of oxalate have been identified on the apical membrane. These include the sulfate/bicarbonate exchanger that accepts oxalate as a substrate (10, 14), the electrogenic chloride/oxalate exchanger that accepts formate as a substrate (7), and a recently identified OH/oxalate exchanger (10). On the basolateral membrane, a sulfate/bicarbonate exchanger that is functionally similar to the apical sulfate transporter also accepts oxalate as a substrate (9).

Understanding the interaction of sulfate and chloride with each of these oxalate exchangers has been complicated by species differences and the inherent difficulty of analyzing similar transporters in the same membrane population. For example, significant chloride/sulfate exchange has been observed in rat microvillus membrane vesicles (14) but not in the rabbit (7, 10). Despite the inability to detect chloride/sulfate exchange in rabbit renal microvillus membranes, 1 mM sulfate has been observed to inhibit chloride/oxalate exchange ~60% (10). This suggests either that sulfate serves as a poorly transported inhibitor of the chloride/oxalate exchanger or that a component of chloride/oxalate exchange occurs on the apical membrane sulfate/oxalate exchanger. Current methods have been unable to resolve this issue.

In the present study, we have developed a reconstitution assay that allows for the isolation of oxalate transporters from their native lipid environment while maintaining transport function. We have successfully solubilized and reconstituted oxalate/oxalate exchange activity from rabbit renal microvillus membranes and have physically separated two distinct oxalate exchangers by hydroxyapatite chromatography. The activity found in the early fractions eluted from the hydroxyapatite column probably represents activity on the sulfate/oxalate exchanger. This conclusion is based on the detection of oxalate-stimulated sulfate uptake and the observation that sulfate is a potent inhibitor, whereas chloride is a weak inhibitor, of oxalate transport in this fraction. The activity observed in the later fractions probably represents transport on the chloride/oxalate exchanger, since oxalate-stimulated sulfate uptake is not detected and chloride is a potent inhibitor of oxalate transport activity in this fraction. We also find that 10 mM sulfate inhibits oxalate exchange activity in the later hydroxyapatite fractions by 41%. Because oxalate/sulfate exchange is not detected in this fraction, these results suggest that sulfate is poorly transported on the chloride/oxalate exchanger; however, at millimolar concentrations, it appears that sulfate is capable of blocking the chloride/oxalate exchanger to a modest degree.

Our observation that apical membrane sulfate/oxalate exchange and chloride/oxalate exchange are mediated by separate transporters supports the hypothesis proposed by Kuo and Aronson (10) that an oxalate exchanger, functioning either in the sulfate or carbonate exchange mode, would serve as a mechanism of recycling oxalate across the apical membrane. In turn, oxalate would serve as the driving force for the transport of chloride on the chloride/oxalate exchanger, thereby providing a mechanism of chloride entry across the proximal tubule apical membrane.

We have also functionally reconstituted a sulfate/oxalate exchanger from the basolateral membrane. This probably represents the basolateral sulfate/bicarbonate exchanger described by Pritchard and Renfro (15). This transporter also accepts oxalate as a substrate and probably serves as a mechanism of oxalate entry and sulfate exit in the proximal tubule. Comparison of both the reconstituted apical and basolateral sulfate/oxalate exchangers demonstrate that they have similar substrate specificities. These results are consistent with observations in membrane vesicles and raise the possibility that they are identical or closely related proteins. However, although both apical and basolateral membrane sulfate/oxalate exchange activity are reduced in the presence of excess detergent, sulfate/oxalate exchange activity from apical membranes is inactivated at a faster rate compared with basolateral membranes. In these experiments, the proteins were solubilized initially at a low-detergent concentration to remove the protein from the native lipid environment and then treated with a high concentration of octylglucoside for inactiviation. Exposure to octylglucoside at a detergent-to-protein ratio of >10 reduces the likelihood that lipid from native membranes is present as mixed micelles (5); however, we cannot exclude the possibility that the difference in the rate of inactivation represents either residual lipid or some other factor solubilized from the basolateral membrane exerting a stabilizing effect on the sulfate/oxalate exchanger. Alternatively, the different rates of inactivation by octylglucoside might be explained if apical and basolateral sulfate/oxalate exchange is mediated by two different transport proteins with similar substrate specificity.

It is of interest that, in proteoliposomes reconstituted from microvillus membrane protein, the substrate oxalate stabilizes oxalate/oxalate exchange activity in the presence of octylglucoside. Because the oxalate/oxalate exchange mode represents transport on both the sulfate/oxalate and chloride/oxalate exchangers, oxalate must exert its protective effect on both transporters. A detailed examination of the effects of different substrates on detergent inactivation may provide insight into how different substrates interact with the different oxalate transport proteins.

In summary, we have solubilized and reconstituted three of the known oxalate transport processes from the rabbit renal cortex. Solubilization and reconstitution will allow the individual transporters to be characterized in terms of their substrates and kinetics and may help to identify those modes of oxalate transport involved in the stimulation of proximal tubule NaCl reabsorption.