Abstract
Inhibition of [14C]-urate uptake by uricosuric and antiuricosuric agents was investigated in human brush-border membrane vesicles, urate being transported either by anion exchange mechanisms or by voltage sensitive pathway. The IC50 for drugs on [14C]-urate uptake in vesicles loaded with 1 mM cold urate or with 5 mM lactate was, respectively: 0.7 and 0.3 μM for benzbromarone; 6 and 4 μM for salicylate; 133 and 13 μM for losartan; 520 and 190 μM for sulfinpyrazone and 807 and 150 μM, for probenecid. The IC50 ratio for [14C]-urate uptake in exchange for cold urate or for lactate varied from about 1 for salicylate to 10 for losartan, supporting the hypothesis that two distinct anion exchangers are involved in urate transport. Application of Hill equation revealed that urate/anion exchangers have more than one binding site, possibly two binding sites with high cooperativity, for benzbromarone and sulfinpyrazone, but only one for probenecid, salicylate and losartan. The uricosuric diuretic, tienilic acid was 10 to 50 times more potent than hydrochlorothiazide, chlorothiazide and furosemide, for inhibiting [14C]-urate uptake in exchange for cold urate. This higher potency is the reason of its uricosuric properties. All uricosuric agents, as well as the antiuricosuric agents, pyrazinoate and ethambutol, had a much lower potency for inhibiting [14C]-urate uptake through the voltage sensitive pathway (apical secretory step) than through the urate/anion exchangers. This suggests that antiuricosuria, induced by pyrazinoate and ethambutol, as well as by low concentrations of uricosuric agents, does not result from an inhibition of the apical voltage sensitive pathway.
Under physiological conditions FEurate in human is about 10% (Gutman, 1966). When uricosuric drugs, such as benzbromarone, sulfinpyrazone or probenecid are administred to reduce hyperuricemia, FEurate increases up to values of about 20 to 40% (Gutman, 1966; Sinclair and Fox, 1975). It is generally accepted that these drugs act from the lumenal side of proximal tubules and inhibit urate reabsorption (Diamond, 1978). Other drugs, such as the tuberculostatics pyrazinamide and ethambutol, and most diuretics, reduce FEurate (Emmerson, 1978). Thus, administration of a single dose of 2 to 3 g of pyrazinamide to human can lead to FEurate values of less than 1% (Steele and Rieselbach, 1975). It is generally considered that antiuricosuric drugs act by inhibiting the tubular secretion of urate. However, there is some evidence that they might stimulate urate reabsorption (Guggino and Aronson, 1985).
The membrane mechanisms involved in urate reabsorption by the human kidney have been partly elucidated (Roch-Ramel and Diezi, 1997). To be reabsorbed, urate crosses the apical membrane of proximal tubules through anion exchangers that exchange lumenal urate for intracellular organic anions. Two urate/anion exchangers have been described in human brush-border membranes, one for which urate has more affinity than lactate (that we will call the “high urate affinity exchanger”), and the other one for which lactate has more affinity than urate (the “low urate affinity exchanger”) (Roch-Ramel et al., 1996, a and b). These anion exchangers appear essential for urate reabsorption because only urate reabsorbing species possess anion exchangers with affinity for urate (Guggino et al., 1983;Roch-Ramel and Diezi, 1997). The transport mechanisms allowing transfer of urate from cell to peritubular interstitium, the second step in reabsorption, have still not been fully characterized. Preliminary data suggest that in humans as in rats (Polkowski and Grassl, 1993), the efflux of urate from proximal cell to peritubular space occurs through a voltage sensitive pathway. The membrane mechanisms involved in urate secretion have been only partly elucidated. The mechanism allowing urate uptake from peritubular interstitium to cell remains unknown, whereas the efflux of urate from cell to lumen occurs through a voltage-sensitive pathway (Roch-Ramel and Diezi, 1997; Roch-Ramelet al., 1994).
In our study we investigated the effects of uricosuric and antiuricosuric drugs on the apical transport mechanisms. As the urate/anion exchangers play a major role in urate reabsorption, uricosuric drugs should interact with these transport mechanisms. However, anion exchangers allowing bidirectionel transport, urate could as well use the exchangers to leave tubular lumen and enter the cell (reabsorptive direction), or to leave the cell and enter into the tubular lumen (secretory direction). Consequently, antiuricosuric as well as uricosuric drugs might exert their effect by interacting with the urate/anion exchangers. Drugs acting on the voltage-sensitive pathway, in contrast, are expected to be antiuricosuric, because the cell electronegativity favors urate transfer from cell to lumen. An inhibition of this pathway would result in a decrease of urate secretion.
Our data demonstrate that uricosuric and antiuricosuric compounds inhibited urate transport through the urate/anion exchangers. All compounds, including pyrazinoate, had at least 10 times more affinity for the exchangers than for the voltage sensitive pathway. Thus, the effect of drugs on urate transport at the apical membrane appear to be principally on the urate/anion exchangers, the effect on the voltage sensitive pathway being of secondary importance.
Methods
Membrane vesicle preparation.
BBMV were isolated as already described (Roch-Ramel et al., 1994) from renal tissue of tumor patients (50–72 yr old), immediately after nephrectomy. Briefly, after removal of the capsid, macroscopically tumor free renal cortex was isolated by dissection. The cortex was maintained 12 to 72 hr in ice-cold sterile culture medium until the membrane purification procedure. BBMV were prepared according to the EGTA/Mg2+precipitation method (Biber et al., 1981). Briefly, portions of total cortex (0.8 g) were homogenized in 16 ml of isotonic buffer containing 300 mM mannitol, 5 mM EGTA and 12 mM Tris-HCl, pH 7.4. BBM purification was achieved by two precipitations with MgCl2and a series of differential centrifugations. The purified membranes were suspended in a buffer containing 300 mM mannitol and 20 mM HEPES-Tris, pH 7.4. The final volume of the BBMV suspension was adjusted to yield a protein content of 25 to 30 mg/ml. The vesicles were frozen and stored in liquid nitrogen until use.
As demonstrated earlier (Roch-Ramel et al., 1994), BBM were enriched by a factor of 17, compared to basolateral membranes, leading to only a 6% contamination of BBMV by basolateral membranes vesicles.
Protein determination.
Protein content was determined by the Bio-Rad protein assay kit (Bio-Rad Laboratories GmbH, Münich, Germany) using bovine plasma γ-globulin as the standard.
Transport experiments.
BBMV were thawed on ice and diluted in the appropriate volume of intravesicular buffer (to have 50–65 μg protein/filter) (Roch-Ramel et al., 1994). Intravesicular buffer contained 300 mM mannitol, 20 mM HEPES-Tris, pH 7.4. In trans-stimulation experiments with 1 mM urate or 5 mM lactate loaded BBMV, nine volumes of BBMV were incubated for 90 min in one volume of media containing either 10 mM urate or 50 mM lactate. In chloride trans-stimulation experiments, BBMV were prepared and loaded as described earlier (Roch-Ramel et al., 1994). In experiments in which [14C]-urate uptake was stimulated through the voltage-sensitive pathway, BBMV were preincubated with 4 to 7 μmol/mg protein of valinomycin dissolved in ethanol or for control conditions with ethanol alone (Roch-Ramel et al., 1994).
The uptake of [14C]-urate was studied at 25°C by the rapid-filtration technique, the vesicle suspension being warmed for 15 min at 25°C before initiation of transport. In trans-stimulation experiments, transport was initiated by mixing either 3 μl of unloaded BBMV, or 3 μl of BBMV preloaded with either 1 mM urate or 5 mM lactate, with 120 μl of the incubation medium composed of 300 mM mannitol, 20 mM HEPES-Tris, pH 7.4, 40 to 50 μM [14C]-urate, and the inhibitors of transport at different concentrations. In experiments in which the voltage sensitive pathway for urate was investigated, transport was initiated by mixing 3 μl of BBMV (preincubated with valinomycin or with ethanol alone) with 20 μl of incubation medium composed of 100 mM mannitol, 100 mM potassium gluconate, 50 μM [14C]-urate and the inhibitors under investigation. Uptakes were measured for 15 sec, the uptake reaction being terminated by diluting the reaction medium with 1 ml of ice-cold stop solution. The stop solution contained 230 mM mannitol, 60 mM Na2SO4, 5 mM Tris-H2SO4, pH 7.4, 1 mM probenecid and 5.10−5 M methylmercuric chloride. The quenched solutions were then immediately poured onto prewetted Sartorius nitrocellulose filters (0.65-μm pore size) kept under suction. The filters were washed twice with 3 ml of ice-cold stop solution and dissolved in 3 ml of scintillation fluid (Scintillator 299; Packard Instrument, Downers Grove, IL). The radioactivity associated with the filters was counted in a liquid scintillation spectrophotometer (Tri-Carb 4640, Packard Instrument). All uptake measurements were corrected for nonspecific binding of radiolabeled solutes to the filters.
Expression of data and statistics.
Fifteen second uptakes were calculated in pmol/mg protein. In trans-stimulation experiments, the component due to anion-exchange (“stimulated uptake”) was obtained from the difference between uptakes in anion loaded BBMV and in unloaded BBMV. The transport of urate resulting from the voltage sensitive pathway was obtained from the difference between uptakes in valinomycin treated BBMV (inward positive potential) and in BBMV treated only with ethanol (potential equilibrium). “Stimulated uptake” was measured in control conditions (A), in absence of any compounds under investigation, and in experimental conditions (B), in which different concentrations of uricosuric or antiuricosuric compounds were added to the uptake medium. Inhibitory effect was expressed as % inhibition, and was calculated as follows: inhibition (%) = {1- (A-B)/A}*100. 100% inhibition of transport was obtained when anion or potential stimulated uptakes were equal to nonstimulated transport.
In experiments in which uptake inhibition was related to log-concentration of inhibitor in the uptake medium (figs. 1, 2, 3), data for each inhibitory agent came from at least three different membrane preparations, with each data point measured in triplicate. Data were fitted to the equation: I/Imax = [S]n/K′ + [S]n, in which K′ is the Hill coefficient and n the apparent number (napp) of carrier binding sites for the inhibitor (Segel, 1968). When n = 1, K′ is the inhibitor concentration that yields half-maximum inhibition (IC50), for other n values, IC50 = K′ n. All curve fittings and estimation of IC50± S.E.M. were performed by using Kaleidagraph v.3.05, Abelbeck Software Inc. Significance of differences between IC50measured in urate, lactate or chloride loaded BBMV was determined by Wilcoxon sign rank test.
Inhibitory effect of uricosuric agents on urate uptake through urate/anion exchange mechanisms. The log-concentration inhibitory effect of benzbromarone (□), salicylate (▴), sulfinpyrazone (Δ) and probenecid (•) on [14C]-urate uptake stimulated in exchange for cold urate (A) or for lactate (B), and in the case of probenecid, also for chloride (B, (○)) are represented. Sigmoids were calculated by applying Hill equation (see “Methods”). Data for each sigmoid originate from at least three different membrane preparations. IC50 and nappare given in table 1.
Inhibitory effect of benzbromarone metabolites on [14C]urate uptake in exchange for cold urate. The log-concentration inhibitory effect of benzbromarone shown in Fig 1A was reproduced (□) for direct comparison with data obtained for benzarone (•), M1 (1′-hydroxybenzbromarone) (▴) and M2 (6-hydroxybenzbromarone) (○). Sigmoids were calcultated by applying Hill equation (see “Methods”). Data for each sigmoid came from at least three different membrane preparations. IC50 and napp are given in table 1.
Inhibitory effect of losartan and EXP 3174 on urate uptake through urate/anion exchange mechanisms. Inhibitory effect of losartan was investigated on [14C]-urate uptake stimulated either in exchange for cold urate (○), lactate (Δ) or chloride (▴). EXP 3174 was tested only on [14C]-urate uptake stimulated in exchange for chloride (♦). Sigmoids were calculated by applying Hill equation (see “Methods”). Data for each sigmoid originate from at least three different membrane preparations. IC50 and napp are given in table 1.
Data in figures 4, 5, 6 were expressed as means ± S.E. of experiments performed in triplicates, on BBMV isolated from at least four different kidneys.
Inhibitory effect of uricosuric agents on urate uptake through the voltage sensitive pathway. Data are mean ± S.E. of five experiments performed on different membrane preparations.
Effect of antiuricosuric compounds on urate-stimulated uptake through anion exchange mechanisms and voltage sensitive pathway. A, Ethambutol inhibitory effect on [14C]-urate uptake. B, Pyrazinoate effect on [14C]-urate uptake. Positive values represent inhibition of uptake, whereas negative values represent stimulation of [14C]-urate uptake. Data are mean ± S.E. of four to five experiments performed on different membrane preparations.
Effect of diuretics on urate uptake through anion exchange mechanisms, and voltage sensitive pathway. A, Inhibitory effect on [14C]-urate stimulated uptake in exchange for cold urate. B, Inhibitory effect on [14C]-urate stimulated uptake through the voltage sensitive pathway. Data are mean ± S.E. of four experiments performed on different membrane preparations.
Chemicals.
[14C]-Urate (52.5 and 50 mCi/mmol) was obtained from American Radiolabeled Chemical Inc. (St. Louis, MO). All chemicals were purchased either from Sigma Chemical (St. Louis, MO) or Fluka (Buchs, Switzerland) and were at least of analytical grade, Losartan and EXP 3174 were kindly supplied by Merck Research Laboratories (West-Point, PA) and benzbromarone metabolites by Drs. de Vries and Walter-Sack, Abteilung für Klinische Pharmakologie, Universität Heidelberg, Germany.
Results
The inhibitory potency of uricosuric and antiuricosuric agents was investigated by measuring the uptake of 50 μM [14C]-urate stimulated in exchange for 1 mM cold urate or 5 mM lactate, and in a few experiments in exchange for 40 mM chloride. Similar experimental conditions have been used in a former study (Roch-Ramel et al., 1994). Because of the differences of urate affinity for the urate/anion exchangers, BBMV had to be preloaded with a higher concentration of lactate (5 mM) than of urate (1 mM), to observe a stimulation of [14C]-urate uptake. In control conditions, i.e., in absence of uricosuric or antiuricosuric agents, 15 sec [14C]-urate stimulated uptake in BBMV loaded with 1 mM cold urate was 158 ± 7 pmol/mg protein (n = 15). In 5 mM lactate loaded BBMV and in 40 mM chloride BBMV, this uptake was lower, 26 ± 4 (n = 9) and 32 ± 3 (n = 5) pmol/mg protein, respectively. The inhibitory potency of uricosuric and antiuricosuric agents was also investigated when 15 sec [14C]-urate uptake was stimulated by an inside positive potential, created by an inwardly directed 100 mM potassium gluconate gradient and valinomycin. In these experimental conditions, 15 sec [14C]-urate stimulated uptake in control conditions was 57 ± 12 pmol/mg protein (n = 11).
Effects of probenecid, sulfinpyrazone, benzbromarone and salicylate on [14C]-urate uptake through urate/anion exchangers.
Probenecid, sulfinpyrazone and benzbromarone are uricosuric drugs currently used to reduce hyperuricemia (Diamond, 1978), whereas salicylate, despite its uricosuric properties, is generally not used clinically to that end (Gutman, 1966). The inhibitory potency of these drugs was investigated by measuring the uptake of 50 μM [14C]-urate stimulated in exchange for 1 mM cold urate or 5 mM lactate. The inhibitory effect of probenecid was also investigated in BBMV preloaded with 40 mM chloride.
As shown in figures 1, A and B, the four drugs inhibited [14C]-urate stimulated uptake, benzbromarone being the most potent inhibitor. The sigmoids fitting the log inhibitor concentration-effect data for benzbromarone and sulfinpyrazone were steeper than those for salicylate and probenecid, suggesting that the urate/anion exchangers could have more binding sites for benzbromarone and sulfinpyrazone than for salicylate and probenecid. The napp calculated by Hill equation (see “Methods”) were about 2 for benzbromarone and sulfinpyrazone, and about 1 for salicylate and probenecid, whether urate uptake was stimulated in exchange for cold urate or lactate (table 1). This suggests that benzbromarone and sulfinpyrazone were bound to the exchangers at more than one binding site.
IC50 and Hill napp coefficients for uricosuric agents
Data of figure 1A and table 1 show that IC50 of [14C]-urate uptake in exchange for cold urate was obtained with 0.7 ± 0.2 μM benzbromarone, 5.5 ± 1.4 μM salicylate, 520 ± 140 μM sulfinpyrazone and 807 ± 420 μM probenecid, respectively. Because sulfinpyrazone sigmoid was steeper than that of probenecid, a 20% inhibition of [14C]-urate uptake was observed with a lower concentration of probenecid than of sulfinpyrazone, whereas IC50 was smaller for sulfinpyrazone than for probenecid. Data of figure 1B and table 1 show that IC50 of [14C]-urate uptake in exchange for lactate was 0.3 ± 0.1 μM for benzbromarone, 3.6 ± 0.8 μM for salicylate, 190 ± 20 μM for sulfinpyrazone and 150 ± 60 μM for probenecid. Most uricosuric agents had a higher inhibitory potency for urate uptake through the “low urate affinity exchanger” than through the “high urate affinity exchanger” (table 1). Only salicylate had a similar potency for inhibiting urate uptake through both urate/anion exchangers. The probenecid IC50 was 110 ± 52 μM when [14C]-urate uptake was stimulated in exchange for chloride.
Effects of benzarone and benzbromarone metabolites on [14C]-urate uptake through the “high urate affinity exchanger”.
Benzbromarone is rapidly oxidized in the liver, although its uricosuric effect is long lasting (Walter-Sacket al., 1988). De Vries et al. (1993)demonstrated the presence in plasma and urine of two main metabolites, M1 and M2, which they suggested might be responsible for benzbromarone long lasting uricosuria. Other investigators suggested that benzarone, the debrominated compound, could be the metabolite responsible for this long lasting uricosuria (Broekhuysen et al., 1972). We investigated the inhibitory effects of benzarone and the two metabolites M1 and M2 on [14C]-urate uptake, when stimulated through exchange for urate. The data of figure2 show that benzarone inhibitory potency was about 100 times lower than that of M1 and M2. The sigmoids fitting M1 and M2 data were not as steep as the sigmoid fitting benzbromarone data, napp calculated for M1 and M2 was close to one (table 1), whereas napp benzbromarone was about 2.
Effects of losartan and its metabolite (exp 3174) on [14C]-urate uptake through urate/anion exchangers.
Losartan, an antagonist of angiotensin II with uricosuric properties (Burnier et al., 1995; Nakashimaet al., 1992), is a member of a new class of antihypertensive drugs (Carr and Prisant, 1996). We investigated the ability of losartan and Exp 3174, a losartan metabolite, to interfere with [14C]-urate transport. The data of figure3 and table 1 show that losartan IC50 were 13.2 ± 2, 133 ± 22, and 20 ± 5 μM, respectively, for [14C]-urate uptake stimulated in exchange for lactate, cold urate and chloride respectively. Exp 3174, the metabolite of losartan, was about 20 times less potent than losartan to inhibit [14C]-urate uptake in exchange for chloride, IC50 for Exp 3174 was 593 ± 87 μM compared to 20 ± 5 μM for losartan. Losartan and Exp 3174 nappwere close to one, suggesting that the urate/anion-exchangers had only one binding site for these uricosuric agents (table 1).
Effects of uricosuric drugs on [14C]-urate uptake through the voltage sensitive pathway.
The data of figure 4 show that all investigated uricosuric agents inhibited to some extent [14C]-urate uptake through the voltage sensitive pathway. The inhibitory potencies of all agents were lower for the voltage sensitive pathway than for the exchangers. At the concentration of 1 mM, salicylate, losartan and probenecid inhibited [14C]-urate uptake by less than 50%, whereas 1 mM sulfinpyrazone inhibited uptake by 72 ± 6%. Benzbromarone was more potent, 10 and 100 μM inhibiting [14C]-urate uptake by 52 ± 9% and 90 ± 3%, respectively.
Effects of ethambutol and pyrazinoate, two antiuricosuric compounds, on [14C]-urate transport mechanisms.
Ethambutol and pyrazinamide are antiuricosuric drugs (Emmerson, 1978). The decrease in urate excretion observed during pyrazinamide administration is due to pyrazinoate, a metabolite of pyrazinamide (Weiner and Tinker, 1972). Data shown in figure5 compare the effects of ethambutol (fig. 5A) and pyrazinoate (fig. 5B) on urate transport through the voltage sensitive pathway and the urate/anion exchangers. Ethambutol had a low inhibitory potency on [14C]-urate uptake through the “high urate affinity exchanger” and the voltage sensitive pathway, 5 mM ethambutol inhibiting [14C]-urate uptake by 43 ± 4 and 16 ± 6%, respectively (fig. 5A). It was slightly more potent for inhibiting [14C]-urate uptake through the “low urate affinity exchanger,” as 1 mM inhibited [14C]-urate uptake by 51 ± 4%. The potency of pyrazinoate to inhibit [14C]-urate uptake through the voltage dependent pathway was also low, 1 mM inhibiting uptake by only 36 ± 4%. At this concentration, pyrazinoate inhibited totally [14C]-urate uptake through both urate/anion exchangers (fig. 5B). At the concentration of 0.1 mM, pyrazinoate inhibited [14C]-urate uptake through the “high urate affinity exchanger” by 63 ± 2%, and in contrast, cis-stimulated [14C]-urate uptake through the “low urate affinity exchanger.”
Effects of diuretics on [14C]-urate uptake.
Furosemide and thiazide diuretics reduce the FEurate and consequently induce hyperuricemia. Part of the effect is secondary to the contraction of extracellular volume, but a direct effect on urate tubular transport has also been postulated (Kahn, 1988; Steele and Oppenheimer, 1969). To avoid diuretic-induced hyperuricemia, diuretics with uricosuric properties, such as tienilic acid, were developed (Diamond, 1978). The data of figure6 compare the effects of furosemide, hydrochlorothiazide, chlorothiazide and tienilic acid on [14C]-urate uptake in exchange for urate, and on uptake through the voltage sensitive pathway. Hydrochlorothiazide at 1 mM did not inhibit [14C]-urate uptake by either transport mechanisms, whereas 0.1 and 1 mM chlorothiazide inhibited [14C]-urate uptake in exchange for urate by 38 ± 7 and 85 ± 3%, respectively, and 1 mM chlorothiazide inhibited [14C]-urate uptake through the voltage sensitive pathway by 38 ± 5%. Furosemide at 1 mM inhibited [14C]-urate uptake in exchange for urate by 54 ± 7%, and [14C]-urate uptake through the voltage sensitive pathway by 30 ± 6%. Tienilic acid was more potent than other diuretics to inhibit [14C]-urate uptake through the “urate/anion exchanger,” 10 μM tienilic acid inhibiting [14C]-urate uptake by 41 ± 4%. Tienilic acid, as other diuretics, had a low inhibitory potency on [14C]-urate uptake through the voltage dependent pathway. Thus, 1 mM tienilic acid inhibited [14C]-urate uptake through this pathway by only 23 ± 5%.
Discussion
Our data demonstrate that antiuricosuric as well as uricosuric drugs have a higher affinity for the urate/anion exchangers than for the voltage sensitive pathway. Thus, at the apical membrane, the effect of these drugs is most probably the result of an interaction with urate transport through the urate/anion exchangers. The effect on the vectorial transport of urate will depend on the relative concentrations of the drug in the lumen and in proximal cells. Drugs with affinity for the urate/anion exchangers will be uricosuric when acting from the lumen, whereas they will be antiuricosuric when acting from the intracellular space. The exchange of urate for chloride appears to proceed through the same exchanger as the exchange of urate for lactate, the “low urate affinity exchanger,” because the IC50 of probenecid and losartan, drugs investigated in both experimental conditions, were not statistically different when [14C]-urate uptake was stimulated either through exchange for lactate or through exchange for chloride. In contrast, probenecid and losartan IC50 measured when [14C]-urate uptake was stimulated in exchange for cold urate, differed statistically from those measured when urate uptake was stimulated in exchange for chloride or for lactate.
Mechanism responsible for uricosuria.
There are a few observations suggesting that uricosuric agents act from the lumen. One study reported that probenecid-induced uricosuria is markedly reduced when probenecid tubular secretion is inhibited by p-aminohippurate (Meisel and Diamond, 1977). Another study observed that the uricosuric effects of probenecid and salicylate were strongly reduced by an acidification of urine. In acidic urine the proportion of nonionized molecules of probenecid and salicylic acid is increased, and because these molecules are hydrophobic they diffuse out of the tubular lumen. This decrease in luminal drug concentration resulted in a concomitant decrease in uricosuria (Gutman, 1966). Uricosuric drugs may either bind to the urate/anion exchangers without being transported, limiting urate access to the transporter or they may compete with urate for transport. In both cases, more urate remains in the tubular lumen. Our data do not allow to distinguish between these two possibilities. All uricosuric compounds investigated here were more potent for inhibiting urate uptake through the “low urate affinity exchanger” than through the “high urate affinity exchanger.” The ratio of IC50 for urate transport through the “high urate affinity exchanger” and through the “low urate affinity exchanger” were about 1 for salicylate, 2 to 3 for benzbromarone and sulfinpyrazone, 5 for probenecid and 10 for losartan. These differences in IC50ratios give weight to the hypothesis that two anion exchangers might play a role in urate reabsorption (Roch-Ramel et al., 1996, a and b).
Benzbromarone is metabolized by the liver, and two metabolites have been identified, M1 and M2, both of which had affinity for the “high urate affinity exchanger.” The therapeutic blood concentrations of benzbromarone and its active metabolites are low, about 2 to 4 μM for benzbromarone and about 1 μM for M1 and M2 (De Vries et al., 1993; Walter-Sack et al., 1995). These compounds are strongly protein bound, and consequently only small amounts reach the lumen by glomerular filtration. Nevertheless, benzbromarone administration induces a long lasting uricosuria. The high affinity of benzbromarone and its metabolites for the urate/anion exchangers explain such important uricosuria. Little is known on the renal excretion of benzbromarone, and in particular, it is unknown whether benzbromarone is secreted by the kidney. Benzarone, the debromized benzbromarone, had a much lower potency than benzbromarone and its metabolites. This demonstrates that, as proposed by Walter-Sacket al. (1988), M1 and M2, but not benzarone, are benzbromarone metabolites responsible for its long lasting uricosuric effect.
The log concentration inhibition relationship for benzbromarone and sulfinpyrazone was much steeper than that for salicylate, probenecid and losartan. The application of Hill equation gives a napplarger than one for benzbromarone and sulfinpyrazone, which indicates that the carrier has more than one binding site for benzbromarone and sulfinpyrazone, possibly two binding sites with high cooperativity (Segel, 1968). The benzbromarone metabolites, M1 and M2, had a napp of about 1, which suggests that the hydroxylation of benzbromarone hindered the binding to a second site.
Losartan is a representative compound of a new class of antihypertensive drugs, the antagonists of angiotensin II at the receptor level (Carr and Prisant, 1996). Losartan was more potent than probenecid and sulfinpyrazone for cis-inhibiting [14C]-urate uptake through the urate/anion exchangers. EXP 3174, the main losartan metabolite, was less potent than the parent compound by 30 times. The lower potency of EXP 3174 fits with the observation that uricosuria is better correlated with losartan plasma concentration than with the metabolite concentration (Burnier et al., 1996; Burnier et al., 1995). Our data in human BBMV are close to those published recently by Edwards et al.in rat BBMV, data shown in the last column of table 1 (Edwards et al., 1996). Thus, in rat as in human BBMV, losartan was about six to seven times more potent than probenecid for inhibiting [14C]-urate uptake through the urate/anion exchanger, and in BBMV of both species, EXP 3174 was less potent than losartan. However, EXP 3174 was only six to eight times times less potent than losartan in rat BBMV. Such differences in drug potencies in rat and human BBMV are not surprising, if one considers that the substrate affinities for rat and human urate/anion exchangers differ. In human, hydroxyl ion is not a substrate of the urate/anion exchangers, whereas it is a good substrate in rats (Roch-Ramel et al., 1994). Another substrate affinity difference is illustrated by the lack of p-aminohippurate affinity for human urate/exchangers, while it shows a high affinity for the rat urate/anion exchanger (Kahn et al., 1983; Roch-Ramel et al., 1994). Nevertheless, it appears that rat BBMV can be used as a tool to screen uricosuric compounds, keeping in mind that affinitites of drugs for rat and human urate/anion exchangers might differ. Thus, Dan et al.observed in rat BBMV (data shown in last column of table 1) that the rank order of potency of uricosuric drugs for inhibiting urate uptake, was benzbromarone > tienilic acid > sulfinpyrazone > probenecid (Dan and Koga, 1990). We observed a similar rank order of potency in human BBMV.
Mechanisms responsible for antiuricosuria.
Most uricosuric drugs (probenecid, sulfinpyrazone, salicylate, but not benzbromarone) have been reported to have a biphasic effect on urate transport, antiuricosuria being observed at lower concentrations than uricosuria (Diamond, 1978; Gutman, 1966). Such biphasic effect is particularly evident for salicylic acid (Gutman, 1966). It was suggested that, at low dosage, uricosuric agents could inhibit urate secretion. Our data demonstrate that antiuricosuria does not result from an inhibition of the apical step of secretion, because of the low affinities of uricosuric drugs for the apical voltage-sensitive pathway. It remains that the antiuricosuric effect could be the result of an inhibition of urate uptake at the basolateral membrane, first step of secretion. At present the basolateral transport mechansims of urate have not been identified, thus this possibility remains open. However, as discussed earlier by Diamond (Diamond, 1978), an inhibition of urate secretion at low doses, at least in the case of sulfinpyrazone, is not compatible with the net secretion of urate observed when sulfinpyrazone is administered to patients undergoing osmotic diuresis and urate loading (Gutman et al., 1959). An alternative is that antiuricosuria results from a stimulation of urate reabsorption. Sulfinpyrazone, probenecid and salicylate are bound to plasma proteins and reach tubular lumen by secretion. It might be that at low doses, the basolateral uptake allows a cellular concentration relatively high compared to the lumenal concentration, and thus the drug could stimulate urate uptake from lumen through the exchanger. Part of the antiuricosuria observed by repetitive p.o. administration of thiazide diuretics or of furosemide (Emmerson, 1978) might result from a similar mechanism, whereas uricosuria observed by i.v. infusion of chlorothiazide or furosemide (Diamond, 1978), might be through the inhibition of the exchanger from the lumenal side.
Antiuricosuria induced by stimulation of urate reabsorption is obviously the mode of action of pyrazinoate, as was suggested earlier (Guggino and Aronson, 1985). At the lumenal side, PZA enters proximal cells by sodium-cotransport as well as through the “low urate affinity exchanger” for which it has a much higher affinity than urate (Roch-Ramel et al., 1996b). Lactate itself enters proximal cells through a sodium-cotransport mechanism (Roch-Ramelet al., 1996b). From cells, PZA exchanges for urate and stimulates its transfer from lumen to cell, first step in urate reabsorption, which results in antiuricosuria. Because PZA affinity for the voltage-sensitive pathway is about 10 times lower than that for the urate/anion exchangers at the apical membrane, the antiuricosuric effect of PZA is primarily by stimulating cellular uptake of lumenal urate. Also, the antiuricosuric effect of ethambutol (Postlethwaiteet al., 1972) cannot be explained by the inhibition of urate transport through the voltage-sensitive pathway, for which it has a very low affinity.
In summary, our data demonstrate that uricosuric as well as antiuricosuric agents have more potency for inhibiting urate uptake through the anion exchange mechanisms, which allow transfer of urate from lumen to cell (first step of reabsorption) as well as from cell to lumen (second step in tubular secretion), than for the voltage-sensitive pathway, which is involved only in urate secretion. Drug-induced antiuricosuria could generally result from a stimulation of urate efflux out of the lumen, as is the case for pyrazinoate.
Acknowledgment
The authors thank Dr. F. Trinkler (Department of Urology, University of Zurich) for his collaboration in providing kidney material.
Footnotes
-
Send reprint requests to: Dr. Françdoise Roch-Ramel, Institut de Pharmacologie, Bugnon 27, CH 1005 Lausanne, Switzerland.
-
↵1 This work was supported by the Swiss National Fund for Scientific Research Grant 32–43451.95.
- Abbreviations:
- BBMV
- brush-border membrane vesicles
- FEurate
- fractional excretion of urate
- IC50
- concentration of inhibitor which yielded 50% inhibition of urate uptake
- napp
- number of apparent binding sites calculated by Hill plot equation
- M1
- 1′-hydroxybenzbromarone
- M2
- 6-hydroxybenzbromarone
- Received July 23, 1996.
- Accepted October 7, 1996.
- The American Society for Pharmacology and Experimental Therapeutics
