Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Renal organic anion transport plays an important role in the secretion and clearance of a wide range of endogenous metabolites and xenobiotics (Moller and Sheikh, 1983). Transepithelial secretion of organic anions such as PAH occurs exclusively in the proximal tubule and involves the transport of the anion across the BLM into the tubule cell against a steep electrochemical gradient. Movement of PAH out of the cell across the BBM into the tubule lumen occurs down its electrochemical gradient but likely involves a mediated process (Pritchard and Miller, 1993). The events involved in the concentrative uptake of PAH and other molecules that share this transport system have been the focus of much research, but only recently have studies shown that PAH uptake across the BLM is a tertiary active transport process. According to this model (Shimada et al., 1987; Pritchard, 1988), uptake of PAH into the cell across the BLM occurs in exchange for a dicarboxylate moving out of the cell down its concentration gradient. The outwardly directed gradient for the dicarboxylate is maintained by intracellular synthesis as well as by Na⁺-coupled entry of the dicarboxylate into the cell which depends on the inwardly directed Na⁺ gradient produced and maintained by Na⁺-K⁺-ATPase. This model was originally proposed based on studies on BLM vesicles (Shimada et al., 1987; Pritchard, 1988) but has subsequently been shown to function in isolated proximal tubules (Chatsudthipong and Dantzler, 1992; Sullivan and Grantham, 1992; Shpun et al., 1995) and renal cortical slices (Pritchard, 1990).

KG has been identified as the most likely physiological anion involved in dicarboxylate-PAH exchange. This is based primarily on its intracellular abundance in the proximal tubule (Pritchard, 1995) and its potency compared with other dicarboxylates to stimulate PAH uptake in the proximal tubule (Pritchard, 1988; Sullivan and Grantham, 1992). In view of its importance in the transport of organic anions, the purpose of this study was to characterize the uptake of KG in BLM and BBM vesicles isolated from rat kidney.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Membrane vesicle preparation. BBM vesicles were isolated from the renal cortex of male Sprague-Dawley rats (250-300 g, Taconic Farms, Germantown, NY) by the Mg⁺⁺ precipitation method as described previously (Edwards et al., 1996). Renal cortices were homogenized at 4°C in 50 mM mannitol and 10 mM HEPES-Tris (pH 7.5) with a Polytron (3 × 30 sec at speed 6). MgCl₂ was added to a final concentration of 10 mM, and the homogenate was stirred on ice for 20 min. The homogenate was centrifuged for 10 min at 2,000 × g, and the resulting supernatant was centrifuged for 20 min at 35,000 × g. The pellet was resuspended in intravesicular buffer (see below) with use of a glass/Teflon homogenizer and centrifuged for 15 min at 35,000 × g. The loosely packed BBM layer was gently washed off the pellet and recentrifuged an additional three times as described above. The final BBM pellet was resuspended in intravesicular buffer at a protein concentration of 10 to 15 mg/ml and stored at room temperature until uptake studies were performed on the same day.

Uptake studies. The composition of the intra- and extravesicular buffers used for a specific experiment, described under "Results," were prefiltered through 0.22-µm filters. Uptake of ¹⁴C-KG or ³H-PAH was measured at 20°C by a rapid filtration method. Ten microliters of BLM or BBM vesicles (75-100 µg protein) equilibrated with the appropriate intravesicular buffer were mixed rapidly with 120 µl of uptake buffer containing the tracer and other constituents as described under "Results." At various time intervals, 3 ml of uptake buffer without substrate was added to the test tube and the solution was filtered through 0.45-µm Millipore filters. The vesicles retained on the filter were washed with an additional 2 × 3 ml of buffer, and the filters were counted in a scintillation counter. Each experiment was performed by triplicate or quadruplicate determinations, and corrections were made for ¹⁴C-KG or ³H-PAH bound to the filters in the absence of vesicles which was always <0.1% of added counts. Uptakes are expressed as picomoles per milligram of protein. Statistical analysis was performed on the absolute uptake values with Student's t test, or for multiple comparisons, analysis of variance followed by Tukey's or Dunnett's test.

Chemicals. ¹⁴C-KG (281 mCi/mmol) and ³H-PAH (4.9 Ci/mmol) were obtained from New England Nuclear (Boston, MA). Valinomycin, unlabeled KG, PAH and other carboxylates were obtained from Sigma Chemical Co. (St. Louis, MO). All other compounds were of the highest grade commercially available. Valinomycin was dissolved in 100% ethanol and added to the vesicles preparations in a 1:130 dilution. An equivalent volume of ethanol was added to control preparations.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

In both BBM and BLM vesicles, an inwardly directed 100 mM Na⁺ gradient caused a large stimulation of ¹⁴C-KG uptake relative to that observed in the absence of Na⁺ (fig. 1). In both vesicle preparations, a transient overshoot occurred in which the intravesicular KG concentrations exceeded those at equilibrium (2 hr) by which time all solute gradients had dissipated. The initial (5 sec) rate of KG uptake in BLM vesicles (258 ± 8.2 pmol/mg/5 sec) exceeded that in BBM vesicles (126 ± 3.9 pmol/mg/5 sec), and peak uptakes in BLM vesicles reached their maximum in 10 to 30 sec compared with 1 min in BBM vesicles. These experiments were performed in the presence of intra- and extravesicular K⁺ (50 mM inside and outside) and valinomycin to minimize the development of diffusion potentials and clearly indicated Na⁺-coupled KG transport in both BLM and BBM vesicles.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Time course of KG uptake into BBM and BLM vesicles. Membrane vesicles were equilibrated in 100 mM potassium gluconate, 50 mM KCl, 5 mM HEPES, pH 7.5 and 10 µg/ml valinomycin. Uptake of 5 µM 14C-KG was performed in 100 mM sodium gluconate (circles) or 100 mM potassium gluconate (squares) and 50 mM KCl, 5 mM HEPES, pH 7.5 and 10 µg/ml valinomycin. Each point is the mean ± S.E.M. of four different experiments.

In the presence of Na⁺, the initial rate of uptake of KG in BLM and BBM vesicles was saturable (fig. 2). However, the kinetics parameters calculated by Michaelis-Menten equations were markedly different for the two membrane preparations. In three different vesicle preparations, the apparent affinity (K_m) of the BLM transporter for KG was 15.2 ± 3.1 µM with a maximum transport rate (V_max) of 386 ± 39 pmol/mg/5 sec. In contrast, in BBM vesicles, the apparent K_m was 158 ± 8.6 µM and V_max was 1106 ± 152 pmol/mg/5 sec, both of which were significantly different (P < .01) from the corresponding values for BLM.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Concentration-dependent uptake of KG in BLM (top) and BBM (bottom) vesicles. Membrane vesicles were equilibrated in 100 mM potassium gluconate, 50 mM KCl, 5 mM HEPES, pH 7.5 and 10 µg/ml valinomycin. Uptake of 14C-KG supplemented with unlabeled KG was performed in 100 mM sodium gluconate, 50 mM KCl, 5 mM HEPES, pH 7.5 and 10 µg/ml valinomycin. Uptake of KG was measured at 5 sec. Kinetic parameters were estimated from double-reciprocal plots. Results are triplicate determinations from a representative experiment.

Previous studies have shown that the transport of other dicarboxylates such as succinate (Wright and Wunz, 1987) and methylsuccinate (Burckhardt, 1984) is electrogenic, and as such, is influenced by the transmembrane potential. To determine whether KG transport is also influenced by transmembrane potentials, the time course of Na⁺-dependent KG uptake into BBM and BLM vesicles was examined in the presence of valinomycin with or without an inwardly or outwardly directed K⁺ gradient. Compared with control conditions (K⁺_in = K⁺_out), an inside negative potential (K⁺_in > K⁺_out) markedly stimulated KG uptake in both BLM and BBM vesicles (fig. 3). However, an inside positive potential (K⁺_in < K⁺_out) inhibited Na⁺-dependent uptake in BBM vesicles and essentially abolished uptake in BLM vesicles. These results demonstrate that Na⁺-dependent KG transport in BBM and BLM vesicles is an electrogenic process that results in the net movement of positive charges.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effect of potassium diffusion potentials on Na+-dependent 14C-KG uptake (5 µM) into BLM (top) and BBM (bottom) vesicles. For Kin < Kout experiments intravesicular solution was 400 mM mannitol and extravesicular solution was 100 mM sodium gluconate and 100 mM potassium gluconate. For Kin = Kout experiments intravesicular solution was 100 mM potassium gluconate and 200 mM mannitol and extravesicular solution was 100 mM potassium gluconate 100 mM sodium gluconate. For Kin > Kout experiments intravesicular solution was 100 mM potassium gluconate and 200 mM mannitol and extravesicular solution was 100 mM sodium gluconate and 200 mM mannitol. In addition, all solutions contained 10 mM HEPES, pH 7.5 and 10 µg/ml valinomycin. PDin refers to estimated intravesicular potential. Results are from three different membrane preparations.

In the present study, 5 mM Li⁺ markedly inhibited Na⁺-dependent KG uptake across both BBM and BLM vesicles, even in the presence of a 20-fold excess of Na⁺ (fig. 4). This agrees with previous in vitro (Shimada et al., 1987; Pritchard, 1988) and in vivo (Ullrich et al., 1984) studies which showed that Na⁺-dependent dicarboxylate transport is highly sensitive to Li⁺.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effect of 5 mM Li+ on Na+-dependent 14C-KG (5 µM) uptake in BLM (top) and BBM (bottom) vesicles. Vesicles were equilibrated in 100 mM potassium gluconate, 50 mM KCl, 5 mM HEPES, pH 7.5 and 10 µg/ml valinomycin. Uptakes were performed in 100 mM sodium gluconate, 50 mM KCl, 5 mM HEPES, pH 7.5, 10 µg/ml valinomycin and either 5 mM NaCl (control) or 5 mM LiCl (Li+). Results are from four experiments.

The substrate specificity of the KG transporter was examined by measuring the ability of selected mono-, di- and tricarboxylates to cis-inhibit Na⁺-dependent KG uptake. Uptake of 5 µM KG was determined in the presence and absence of 1 mM selected anions. In addition to unlabeled KG itself, the dicarboxylates succinate, fumarate, malate and glutarate virtually abolished Na⁺-dependent KG uptake in both BBM and BLM vesicles (fig. 5). The tricarboxylate, citrate, inhibited KG uptake in BBM and BLM vesicles, but to a lesser extent (P < .05) than the dicarboxylates. The monocarboxylates, pyruvate and lactate, as well as the dicarboxylate, oxalate, had no effect on KG transport.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Inhibition of Na+-dependent 14C-KG uptake (5 µM) by various mono-,di- and tricarboxylates. Intravesicular medium was 100 mM potassium gluconate, 100 mM mannitol, 5 mM HEPES, pH 7.5. Uptake buffer was identical except that 100 mM sodium gluconate replaced potassium gluconate and 1 mM of the tested carboxylates was present. Results are expressed as a percent of control uptake which was 187 ± 1.5 pmol/mg/5 sec for BLM and 68 ± 3.1 pmol/mg/5sec for BBM, n = 4.*P < .001 compared with control values for both BLM and BBM. Inhibition by citrate was significantly less (P < .05) than that produced by the dicarboxylates.

Of the tested dicarboxylates that inhibited KG uptake and presumably compete with this anion for the transporter, only KG and glutarate affected PAH uptake into BLM vesicles (fig. 6). In the presence of an inwardly directed Na⁺ gradient, these two anions stimulated PAH uptake when present at concentrations between 1 and 100 µM (P < .05). Maximum stimulation was observed at concentrations of 3 to 10 µM. Citrate and the other dicarboxylates were unable to stimulate PAH uptake.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Effects of various carboxylates on 3H-PAH uptake into BLM vesicles. Intravesicular solution was 100 mM KCl, 100 mM mannitol and 10 mM HEPES, pH 7.5. Extravesicular medium was 100 mM NaCl, 100 mM mannitol, 10 mM HEPES, pH 7.5, 50 µM 3H-PAH and the indicated concentration of carboxylate. Results are expressed as a percent of uptake in the absence of added carboxylate (control) and were 46.5 ± 3.2 pmol/mg/10 sec, n = 3. Significant (P < .05) stimulation of PAH uptake by -KG and glutarate occurred at concentrations of 1 to 100 µM.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

KG has recently been identified as the likely physiological anion involved in PAH/organic anion counter exchange across the BLM of proximal tubule cells (Shimada et al., 1987; Pritchard, 1988). In the present study, we examined the characteristics of KG uptake into BBM and BLM vesicles. We found that KG transport across these two membranes was Na⁺-dependent, saturable, electrogenic and inhibited by Li⁺. These properties are identical with those described for the uptake of other dicarboxylates such as succinate (Wright and Wunz, 1987), malate (Kahn et al., 1984) and methylsuccinate (Burckhardt, 1984), across the BBM and/or BLM.

Although the qualitative aspects of KG transport were similar in the two membrane preparations, there were marked differences in the kinetics of uptake. First, under Na⁺ gradient conditions, the uptake of KG was much more rapid in BLM vesicles than in BBM vesicles. Second, and perhaps most important, the apparent affinity of the transporter for KG in the BLM was 10-fold greater than in the BBM, whereas the maximum transport rate was about 3-fold greater in the BBM than in the BLM. The presence of a high-affinity, low-capacity transporter in the BLM and a low-affinity, high-capacity transporter in the BBM was also observed for the transport of succinate across rabbit renal vesicles (Wright and Wunz, 1987). These results, together with the enrichment profile of the marker enzymes, also suggest that there was little cross-contamination between the two membrane preparations.

The substrate specificity of the transporter was similar in the two vesicle preparations. The dicarboxylates, succinate, malate, fumarate and glutarate, at a concentration of 1 mM inhibited Na⁺-dependent KG uptake to the same extent (~>90%) in both BBM and BLM vesicles. The tricarboxylate, citrate, also inhibited KG uptake, although to a lesser extent (~70-80%) than the above-mentioned dicarboxylates. The monocarboxylates, pyruvate and lactate, and the dicarboxylate, oxalate, had no effect on KG uptake. These results are generally similar to those seen in previous studies on the substrate specificity of succinate (Wright and Wunz, 1987), malate (Kahn et al., 1984) and methylsuccinate (Burckhardt, 1984) transport, with a few exceptions. For example, Wright and Wunz (1987) found that various dicarboxylates consistently produced a greater inhibition of succinate uptake in BLM vesicles than in BBM vesicles of the rabbit. In addition, both Burckhardt (1984) and Kahn et al. (1984) found that 1 mM pyruvate inhibited methylsuccinate and malate transport in rat BLM vesicles by 33% and 53%, respectively. The reasons for these different results are unknown but may be caused by species differences (rat versus rabbit) and/or different experimental conditions (methylsuccinate and malate uptake versus -KG uptake). However, our results do agree with the substrate specificity of a recently cloned low-affinity Na⁺-dependent dicarboxylate transporter which likely corresponds to the BBM transporter (Pajor, 1995). The transport of succinate by this protein, which has been cloned from both rabbit (Pajor, 1995) and human kidney (Pajor, 1996), is inhibited by malate, citrate and fumarate but not by lactate or pyruvate.

Although malate, fumarate, succinate, glutarate and citrate competed with and are apparently transported by the same system as KG, only glutarate and KG were able to affect PAH transport in BLM vesicles. This high degree of substrate specificity of the PAH/organic anion exchanger as well as the biphasic effect of these two anions on BLM PAH uptake, e.g., stimulation at low concentrations and inhibition at high concentrations, are in direct agreement with previous studies (Pritchard, 1988) and provide further support for the existence of a Na⁺-dependent KG-driven PAH uptake mechanism in the BLM.

The source of KG used for PAH/KG exchange is unknown but could derive from intracellular production and/or uptake from the basolateral and/or luminal compartments. Extramitochondrial cytoplasmic KG levels in rat renal slices have been estimated to range from 90 µM to 400 µM, depending on the metabolic state of the tissue (Pritchard, 1995). Because the apparent K_m of the BLM PAH/KG exchanger for KG is about 150 µM (Pritchard, 1995), the estimated levels of KG, if accessible to the transporter, should be sufficient to support PAH/KG exchange. However, studies in intact perfused proximal tubules have demonstrated that basolateral (Chatsudthipong and Dantzler, 1992; Shpun et al., 1995) and, to a lesser extent, luminal KG (Dantzler and Evans, 1996) can stimulate PAH uptake. The ability of KG to affect PAH transport seems to depend in part on the metabolic state of the tissue. For example, in rabbit proximal tubules preloading with KG from the basolateral side stimulated PAH uptake only in the presence of a HEPES buffer (Shpun et al., 1995), which is known to reduce the cellular levels of ATP and tricarboxylic intermediates (Dickman and Mandel, 1992). In the presence of a bicarbonate buffer, KG had no effect on PAH uptake (Shpun et al., 1995). Similarly, luminal KG stimulated PAH uptake in HEPES-buffered medium but not in bicarbonate-buffered medium (Dantzler and Evans, 1996).

Although it is clear that more research is required to dissect the complex interactions between cellular metabolism and PAH/KG exchange, our results suggest that if extracellular KG plays a role in PAH transport it is likely to be via the BLM transporter. Plasma levels, and presumably the initial concentration of KG in the glomerular filtrate, is approximately 5 to 10 µM (Martin et al., 1989). This concentration is well below the K_m for the BBM transporter described in this study but is poised right at the K_m for the high-affinity KG transporter in the BLM. Coupled with the close structural proximity of the Na⁺-dependent KG transporter and the PAH/KG exchanger, this would permit the rapid recycling of extracellular KG to support PAH transport.