Abstract
In the present study, we investigated the interactions between antibiotics, especially β-lactam antibiotics, and rat renal organic anion transporter 1 (OAT1). [14C]p-Aminohippurate (PAH) uptake via OAT1 expressed in Xenopus laevis oocytes was inhibited by all of the penicillins and cephalosporins tested. Penicillin G, carbenicillin, cephaloridine, cephalothin, cefazolin, and cephalexin inhibited [14C]PAH uptake via OAT1 in a competitive manner (Ki = 0.29–2.33 mM). Cinoxacin, a quinolone gyrase inhibitor, also inhibited PAH uptake via OAT1. Other antibiotics, such as gentamicin, streptomycin, and vancomycin, which do not contain anionic moieties, did not interact with OAT1. [3H]Penicillin G and [14C]cephaloridine were demonstrated to be transported via OAT1. Using the cells that stably expressed OAT1, we analyzed the cytotoxicity of several β-lactam antibiotics. Cells expressing OAT1 showed higher susceptibility to cephaloridine (a potentially nephrotoxic β-lactam antibiotic) toxicity than did control cells. The present study suggests that OAT1 is the major organic anion transporter in the kidney that is responsible for the renal secretion of antibiotics, especially that of β-lactam antibiotics. Furthermore, the culture cell system expressing OAT1 was revealed to be useful for the prediction of the nephrotoxicity of β-lactam antibiotics.
The excretion of drugs and their metabolites is an important function of the kidney. A variety of clinically important drugs are classified as organic anions, and the proximal tubule cells possess secretory pathways specific for organic anions. There are several anion transport systems in proximal tubule cells, such as the sulfate/oxalate exchanger, Na+-dicarboxylate cotransporter, andp-aminohippurate (PAH) transporter (Ullrich and Rumrich, 1988). Among them, the PAH transporter is known to be a multispecific organic anion transporter and is considered to play a central role in the excretion of xenobiotics from the kidney (Pritchard and Miller, 1993).
Recently, a renal organic anion transporter, organic anion transporter 1 (OAT1), was isolated from rat kidney (Sekine et al., 1997; Sweet et al., 1997). OAT1 was shown to mediate the transport of PAH, cyclic AMP, cGMP, prostaglandin E2, urate, α-ketoglutarate, and methotrexate when expressed in Xenopus laevis oocytes, and [14C]PAH uptake via OAT1 was inhibited by various anionic drugs. OAT1 is considered to be the multispecific organic anion transporter responsible for the excretion of anionic drugs from the kidney.
β-Lactam antibiotics (penicillins, cephalosporins, and carbapenems) are among the most important groups of drugs excreted by the kidney (Nightingale et al., 1975; Kamiya et al., 1983;Tune et al., 1997). Based on previous in vivo pharmacokinetic analyses, β-lactam antibiotics are suggested to be not only filtered through the glomerulus but also actively secreted by the proximal tubules. It has been reported that secretion of penicillin G (PCG) by proximal tubule cells of the kidney is sensitive to probenecid (Barza et al., 1975; Bergeron et al., 1975; Nierenberg, 1987; Tsuji et al., 1990). PCG exerted a moderate inhibitory potency on the contraluminal PAH transport (Ullrich et al., 1989; Makhuli et al., 1995). Cephalosporins inhibited PAH uptake in rat renal slices (Hori et al., 1982) and renal plasma membrane vesicles (Takano et al., 1989). Cephaloridine, a cephalosporin that possesses both anionic and cationic moieties, inhibited PAH transport but not N-methylnicotinamide transport in basolateral membrane vesicles (Kasher et al., 1983). Thus, β-lactam antibiotics have been considered as being secreted by the S2 segment of the proximal tubule via the PAH transporter system (Moller and Sheikh, 1983; Ullrich et al., 1989). The relatively short serum half-life of β-lactam antibiotics has been attributed to this active secretion via the renal organic anion transporter.
On the other hand, several cephalosporins such as cephaloridine and cephalothin have been shown to be toxic to the renal proximal tubule cells (Barza, 1978). Cephaloridine, after being taken up by the proximal tubule cells, causes acute proximal tubular necrosis in humans and laboratory animals (Goldstein et al., 1988). The transport of β-lactam antibiotics via a renal organic anion transporter is important to understand their pharmacokinetics and toxicokinetics.
Because OAT1 is presumed to be the major renal organic anion transporter, we investigated the interaction and transport of β-lactam antibiotics via OAT1. Furthermore, we studied the cytotoxicity of these drugs using kidney-derived culture cells expressing OAT1.
Experimental Procedures
Materials.
[14C]PAH (2.0 GBq/mmol) was purchased from DuPont-NEN (Boston, MA). [phenyl-4(n)-3H]Benzylpenicillin (666 GBq/mmol) was purchased from Amersham International, Ltd. (Buckinghamshire, UK). [14C]Cephaloridine (514 MBq/mmol) was supplied by Sankyo Co. (Tokyo, Japan). Collagenase was purchased from Boehringer Mannheim (Indianapolis, IN). Dulbecco’s modified Eagle’s medium was purchased from Nissui Pharmaceutical Co., Ltd. (Tokyo, Japan). All other compounds used in the present study were purchased from Sigma Chemical Co. (St. Louis, MO).
cRNA Synthesis.
Capped cRNAs for OAT1 were synthesized in vitro using T7 RNA polymerase as described elsewhere (Sekine et al., 1997).
Oocyte Isolation.
X. laevis oocytes were digested in OR2 solution containing 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl2, 5 mM HEPES, pH 7.5, and 1.5 mg/ml collagenase for 30 to 40 min at room temperature. After digestion, the oocytes were defolliculated and kept in Barth’s medium containing 88 mM NaCl, 1 mM KCl, 0.33 mM Ca(NO3)2 · 4H2O, 0.41 mM CaCl2 · 2H2O, 0.82 mM MgSO4 · 7H2O, 2.4 mM NaHCO3, and 10 mM HEPES, pH 7.4, at room temperature until cRNA injection, which was performed on the same day.
Oocyte Injection.
The defolliculated oocytes were injected with OAT1-cRNA and the control oocytes with 50 nl of distilled water. After injection, the oocytes were incubated in Barth’s medium supplemented with 0.05 mg/ml gentamicin at 18°C for 2 to 3 days.
Transport and Inhibition Studies.
At 2 to 3 days after cRNA injection, transport studies were performed. The oocytes were preincubated in ND96 solution containing 1 mM glutarate for 2 to 3 h to generate an outwardly directed glutarate gradient before the uptake or inhibition studies. After the glutarate was washed off, the oocytes were transferred to ND96 solution containing 96 mM NaCl, 2 mM KCl, 1.8 mM CaCl2 · 2H2O, 1 mM MgCl2 · 6H2O, 5 mM HEPES, pH 7.4, and a radiolabeled substrate ([14C]PAH, [3H]PCG, or [14C]cephaloridine), and were incubated with or without nonradiolabeled drugs. For inhibition studies, norfloxacin, erythromycin, and chloramphenicol were first dissolved in dimethyl sulfoxide, and diluted to 2 mM in ND96 solution. All other inhibitors were dissolved directly in ND96 solution. The incubation was terminated by adding ice-cold ND96 solution, and the oocytes were washed five times with icecold ND96 solution. Each oocyte was transferred to a scintillation vial and solubilized in 0.25 ml of 10% SDS. After the addition of 2.5 ml of Aquasol-2, the radioactivity in each oocyte was counted with a liquid scintillation counter (LSC-3100; Aloka, Tokyo, Japan).
Kinetic Analysis.
After preincubation with glutarate, the oocytes expressing OAT1 were incubated for 1 h in ND96 solution with various concentrations of PAH in the presence or absence of β-lactam antibiotics. A Lineweaver-Burk plot was used to evaluate the type of inhibition. Because all the drugs tested revealed competitive inhibition of PAH uptake, the Ki value for each drug was calculated from the following equation:Ki = concentration of inhibitor/[(KmPAH with inhibitor/KmPAH without inhibitor) − 1].
Cytotoxicity Test of β-Lactam Antibiotics in S3 Cells.
We had previously established a cell line derived from the terminal portion of the mouse proximal tubule (S3) that was microdissected from the kidneys of SV40 large T-antigen-harboring transgenic mice (Hosoyamada et al., 1996). OAT1-expressing S3 cells were obtained by transfecting S3 cells with OAT1 cDNA in the mammalian expression vector pcDNA3.1 by the electroporation method. S3 cells expressing OAT1 and the mock S3 cells were used in this experiment. Both were subcultivated in a 96-well plate at a concentration of 9 × 103 cells/well and cultured in Dulbecco’s modified Eagle’s medium in the presence or absence of different concentrations of drugs. Cell viability was determined by the staining of basophilic cellular compounds, mainly nucleic acids, using the methylene blue colorimetric technique 3 days after subcultivation (Pelletier et al., 1988; Monks et al., 1991). This procedure provides rapid data collection compared with cell counting or DNA extraction methods. Briefly, the cells were washed three times with 0.15 M PBS, pH 7.4, and fixed with 100 μl of 10% (v/v) Formalin in PBS at room temperature for 10 min. After being washed three times with 0.01 M borate buffer, pH 8.4, the cells were stained with 1% methylene blue in 100 μl of borate buffer at room temperature for 10 min. They were then washed and solubilized in 100 μl of 0.1 N HCl. The optical densities (O. D.) of the samples were measured at 662 nm using a spectrophotometer (DU640; Beckman Instruments, Berkeley, CA). Data are expressed in terms of percent cell viability [(O. D. of treated cells/O. D. of control cells) × 100].
Statistical Analysis.
In oocyte experiments, 6 to 10 oocytes were used. For cell experiments, three wells were used for each drug concentration. The values obtained in each experiment were expressed as the mean ± S.E. Statistical comparisons between the two groups were performed using the Student’s unpaired t test.
Results
Inhibitory Effects of β-Lactam Antibiotics on OAT1-Mediated [14C]PAH Uptake.
The inhibitory effects of antibiotics on [14C]PAH uptake were determined in oocytes expressing OAT1. Figure 1A shows the result for penicillins. All of the tested penicillins inhibited [14C]PAH uptake. At 2 mM, cloxacillin, nafcillin, and piperacillin almost totally abolished the uptake of 2 μM [14C]PAH, whereas PCG showed only a moderate inhibitory effect. Figure 1B shows the results for cephalosporins. All of the cephalosporins tested at a concentration of 2 mM inhibited [14C]PAH uptake. Cefamandole and cephalothin exhibited the most potent inhibitory activity. Cefotaxime and cefadroxil showed a low affinity for OAT1. The three zwitterions (cephaloridine, cefsulodin, and ceftazidime), as well as other anionic cephalosporins, inhibited PAH uptake by oocytes expressing OAT1. Cefsulodin, with an additional sulfonic group, was the most effective among the three zwitterions.
Inhibition of [14C]PAH uptake into OAT1 cRNA-injected oocytes by antibiotics. A, penicillins. B, cephalosporins. C, quinolone gyrase inhibitors. D, nonanionic antibiotics. OAT1 cRNA-injected oocytes were incubated with 2 μM [14C]PAH for 1 h in the presence or absence of 2 mM antibiotics. Each group consisted of from six to eight oocytes. Values are mean ± S.E. of percent uptake of [14C]PAH with respect to the control. The uptake rate of PAH by water-injected oocytes was always less than 5% of those by oocytes expressing OAT1. Levels of significant difference from control (no drugs) were calculated by Student’s unpaired t test. A and B, **P < .01, ***P < .001. C, ***P < .001. Levels of significant difference from control (dimethyl sulfoxide without drugs) were calculated by the Student’s unpaired t test. D, *P < .05.
Three new quinolone antibiotics were also tested (Fig. 1C). Neither norfloxacin nor ofloxacin showed any significant inhibitory effect on PAH uptake; however, cinoxacin potently inhibited PAH uptake.
We also tested other types of antibiotics not containing anionic moieties at a concentration of 2 mM, including gentamicin, streptomycin, vancomycin, and erythromycin (Fig. 1D). These antibiotics did not inhibit PAH uptake via OAT1. However, chloramphenicol, a neutral compound, significantly inhibited PAH uptake, although with low potency (P < .05).
Kinetic Analysis of Drug Inhibition of [14C]PAH Uptake.
Among the 17 β-lactam antibiotics tested in the above experiments, the inhibitory kinetics of two penicillins and four cephalosporins were analyzed with respect to OAT1-mediated [14C]PAH uptake. The double-reciprocal plot of PCG and cephaloridine is shown in Fig. 2. The results revealed that the maximum velocity of PAH transport via OAT1 was not altered by PCG and cephaloridine. The result with four other drugs (i.e., carbenicillin, cephalothin, cefazolin, and cephalexin) also indicated a competitive type of inhibition (data not shown). The Ki values of all the six drugs are shown in Table 1.
Kinetic analysis of inhibition by PCG and cephaloridine of [14C]PAH uptake in OAT1 cRNA-injected oocytes. Double reciprocal plots of [14C]PAH uptake in the presence or absence of antibiotics 5 mM PCG (A) and 2 mM cephaloridine (B). Oocytes were incubated with various concentrations of [14C]PAH for 1 h. Each point represents mean ± S.E.
Ki values of various β-lactam antibiotics to competitively inhibit [14C] PAH uptake
Uptake of Radiolabeled PCG and Cephaloridine.
The transport of PCG and cephaloridine were directly determined in oocytes expressing OAT1. The results of [3H]PCG and [14C]cephaloridine uptake by control oocytes and OAT1-expressing oocytes are presented in Fig.3. The uptake rate of 100 μM [3H]PCG and [14C]cephaloridine by OAT1-expressing oocytes (7.02 ± 0.68 and 8.77 ± 0.64 pmol/oocyte/3 h) were significantly higher than that by control oocytes (2.62 ± 0.23 and 2.56 ± 0.17 pmol/oocyte/3 h).
Uptake of [3H]PCG and [14C]cephaloridine by OAT1-expressing oocytes. Oocytes were incubated with 100 μM PCG (A) and 100 μM cephaloridine (B) for 3 h. Values are mean ± S.E. Levels of significant difference from water-injected oocytes were calculated by Student’s unpairedt test. ***P < .001.
Cytotoxicity Test.
Figure 4shows relative cell viability after treatment with several β-lactam antibiotics of mouse S3 proximal tubule cells expressing OAT1 and mock S3 cells. When exposed to cephaloridine at concentrations of 0.2, 0.5, 0.7, and 1 mM, viability of S3 cells expressing OAT1 decreased to a greater extent than that of mock S3 cells. At much higher concentrations, cephaloridine showed cytotoxicity against both types of cells. When exposed to cephalothin, viability of S3 cells expressing OAT1 was reduced significantly only at a concentration of 0.5 mM. Other drugs, such as amoxicillin and ampicillin, did not show any differences in cytotoxic activity for OAT1- expressing and mock S3 cells.
Cytotoxicity test of β-lactam antibiotics in S3 cells expressing OAT1 and mock S3 cells. Cells were subcultivated in 96-well plates with drugs at different concentrations, and cell viability of S3 cells expressing OAT1 (filled column) and mock cells (open column) was measured after 3 days. Values are mean ± S.E. Levels of significant difference from mock S3 cells were calculated by Student’s unpaired t test. *P < .05, **P < .01, ***P < .001.
Discussion
OAT1 is a multispecific organic anion transporter exclusively located on the basolateral membrane of the middle proximal tubule, S2 segment (Tojo et al., 1999). PAH is a high-affinity substrate of OAT1 (Km = 14.3 μM), and OAT1-mediated PAH transport was inhibited by a variety of anionic drugs (Sekine et al., 1997). OAT1 is an organic anion/dicarboxylate exchanger; preloaded glutarate trans-stimulates the uptake of PAH via OAT1 (Sekine et al., 1997). From these findings, we inferred that OAT1 is the major renal organic anion transporter at the basolateral membrane of the proximal tubule. However, the contribution of OAT1 in the renal excretion of organic anions remains to be elucidated. In fact, we have already identified several isoforms of OAT1 (Sekine et al., 1998), one of which shows higher affinity to PCG.
This study demonstrated the inhibition of PAH transport via OAT1 by all the penicillins and cephalosporins tested and OAT1-mediated transport of [3H]PCG and [14C]cephaloridine (Fig. 3). The present data are in good agreement with those obtained in membrane vesicles (Kasher et al., 1983; Takano et al., 1989; Tsuji et al., 1990) and renal cortical slices (Hori et al., 1982; Nierenberg, 1987). In these reports, the inhibitory effect of various β-lactam antibiotics on the renal organic anion transporter were reported relatively weak. TheKi values of PCG, cephalexin, and cephaloridine for OAT1 (1.68, 2.31, and 2.33 mM, respectively) are similar to the results of the stop-flow peritubular capillary perfusion method (Ullrich et al., 1989), where theKi values of PCG, cephalexin, and cephaloridine are 0.81, 2.3, and >5 mM, respectively. The IC50 value of PCG on PAH uptake by the rabbit basolateral membrane vesicles was reported to be 500 μM (Makhuli et al., 1995). Furthermore, from the kinetic experiment using the in vivo tissue-sampling single-injection technique, theKm value of PCG transport was determined to be 1.39 mM (Tsuji et al., 1990). This value is identical with the Ki for OAT1 (1.68 mM, Table1). The comparison of inhibitory kinetics of OAT1 with those obtained in the previous study on the renal organic anion transport pathway strongly suggests that OAT1 plays the central role for the renal secretion of β-lactam antibiotics.
The affinity of β-lactam antibiotics for OAT1 is not so high. Although β-lactam antibiotics possess a core chemical structure (β-lactam ring) and an anionic moiety, they have considerably different side chains. Furthermore, the hydrophobicity of the substrate is supposed to be closely related to its affinity for the PAH transporter (Moller and Sheikh, 1983; Ullrich et al., 1997). Thus, it is of no wonder that such hydrophilic compounds like β-lactam possess relatively high Ki values for OAT1. β-Lactam antibiotics are only one of the substrates of the multispecific organic anion transporter OAT1.
The patterns of interactions between quinolone gyrase inhibitors and OAT1 are also similar to those reported previously. In humans, urinary excretion of cinoxacin was reduced in the presence of probenecid (Rodriguez et al., 1979). Ullrich et al. (1993) demonstrated weak interactions of norfloxacin and ofloxacin with the PAH transporter, in which the transport of the latter was not demonstrated. These authors concluded that quinolone gyrase inhibitors not possessing a piperazine group interact moderately with the PAH transporter, whereas compounds with an additional piperazine group interact weakly. Cinoxacin, a weak organic acid (pKa = 4.7) not possessing a piperazine group, interacted with OAT1 as expected, whereas neither norfloxacin nor ofloxacin, both of which possess a piperazine group, interacted significantly with OAT1 (Fig. 1C).
In contrast to these anionic antibiotics, the polycationic aminoglycoside gentamicin and vancomycin did not interact with OAT1. Gentamicin was reported to interact with the renal organic cation/H+ exchanger but showed no inhibitory effect on PAH transport across the renal brush-border membrane vesicles (Sokol et al., 1989). Vancomycin inhibited tetraethylammonium uptake across the basolateral membrane (i.e., via the organic cation transport pathway; Sokol, 1991), whereas this drug did not affect PAH transport across the basolateral membrane of the proximal tubule. Although it has a wide substrate selectivity, substrate recognition by OAT1 is limited to antibiotics with anionic moieties.
The luminal secretion of β-lactams has also been supposed to be performed by the carrier-mediated pathways. The transporter molecule responsible for this luminal secretion, however, has not yet been identified. Recently, a distinct type of multispecific organic anion transporter, multidrug resistance-associated protein 2 (MRR2; or canalicular multispecific organic anion transporter) MRP2 (or cMOAT), was localized to the luminal membrane of all proximal tubule segments (S1–S3; Shaub et al., 1997). MRP2, a member of the ATP binding cassette transporter superfamily, mediates the efflux of anionic lipophilic substances, such as leukotriene C4, leukotriene D4, glucuronides, and glutathione conjugates, from the cells (van Aubel et al., 1998). MRP2 seems to transport more lipophilic compounds than those by OAT1. Although there is no direct evidence that the members of the ATP binding cassette transporter superfamily mediate the efflux of antibiotics, they are possible candidates for the luminal transport of β-lactams and quinolone gyrase inhibitors. Now several isoforms of MRP2, such as MOAT-B (Lee et al., 1998) and MLP-1 and MLP-2 (Hirohashi et al., 1998), have been identified. It is of interest to investigate the possible role of these MRP isoforms in the organic anion transport in the kidney because the luminal secretion of organic anions, along with the basolateral uptake of organic anions, plays the crucial role in the renal handling of organic anions, especially β-lactams.
In general, β-lactam antibiotics are safe for clinical use; however, several compounds are known to be potentially nephrotoxic. Cephaloridine is one of such drugs. There are two factors in the development of renal tubular damage caused by β-lactams: (1) accumulation of β-lactams in the cell via active transport mechanisms, followed by (2) intracellular events causing cytotoxicity (e.g., lipid peroxidation or mitochondrial dysfunction; Tune, 1997). Because the nephrotoxicity of cephaloridine can be suppressed by the concomitant administration of probenecid (Tune et al., 1977), its toxic effects have been partly attributed to its active uptake by proximal tubule cells via the renal organic anion transporter. In the present study, we did not directly demonstrate the transport of β-lactams except PCG and cephaloridine. Therefore, we could not draw solid conclusion as to their transport via OAT1. Nevertheless, this system seems to mimic well the nephrotoxicity associated with β-lactam antibiotics. In the experiment shown in Fig. 4, we demonstrated that culture cells expressing OAT1 revealed higher susceptibility to cephaloridine toxicity than mock cells. In contrast, ampicillin and amoxicillin, which are known to be non-nephrotoxic, did not show any such toxic effects. Thus, culture cells expressing OAT1 may be used as a screening system for nephrotoxicity of drugs. If we could use a system of cells expressing human OAT1, it would provide more important information regarding drug transport and possible toxicity in the human kidney.
In conclusion, this study provides for the first time the molecular basis on the renal excretion of β-lactams and quinolone gyrase inhibitors by renal organic anion transporter. Further studies to identify OAT isoforms and to determine their tissue localization and substrate selectivity will help to clarify the molecular mechanisms underlying the pharmacokinetics of β-lactams. Cells expressing OAT1 showed higher susceptibility to the potent nephrotoxic drug cephaloridine than mock cells. We conclude that OAT1 is responsible for the secretion of β-lactam antibiotics and that a culture cell system expressing OAT1 will be a useful tool for the prediction of drug-induced nephrotoxicity.
Footnotes
-
Send reprint requests to: Dr. Hitoshi Endou, Department of Pharmacology and Toxicology, Kyorin University School of Medicine, 6-20-2 Shinkawa, Mitaka, Tokyo 181-8611, Japan.
-
↵1 This work was supported in part by grants from the Japanese Ministry of Education Science, Sports and Culture; Science Research Promotion Fund of the Japan Private School Promotion Foundation; Uehara Memorial Foundation; and Tokyo Biochemical Research Foundation.
- Abbreviations:
- PAH
- p-aminohippurate
- OAT1
- organic anion transporter 1
- PCG
- penicillin G
- MRP2
- multidrug resistance-associated protein 2
- O.D.
- optical density
- Received February 8, 1999.
- Accepted April 13, 1999.
- The American Society for Pharmacology and Experimental Therapeutics