This report describes a potent and selective inhibitor of multidrug and toxin extrusion (MATE) protein, pyrimethamine (PYR), and examines its effect on the urinary and biliary excretion of typical Mate1 substrates in mice. In vitro inhibition studies demonstrated that PYR is a potent inhibitor of mouse (m)Mate1 (Ki = 145 nM) among renal organic cation transporters mOctn1 and mOctn2 (Ki > 30 μM), mOct1 (Ki = 3.6 μM), and mOct2 (Ki = 6.0 μM). PYR inhibited the uptake of metformin by kidney brush-border membrane vesicles (BBMVs) (Ki = 41 nM) and canalicular membrane vesicles in the presence of outward gradient of H+. PYR treatment significantly increased the kidney-to-plasma ratio of tetraethylammonium, and both the liver- and kidney-to-plasma ratios of metformin in mice, whereas it did not affect their plasma concentrations and urinary excretion rates. Furthermore, the plasma lactate concentration, a biomarker for inhibition of gluconeogenesis by metformin, was significantly higher in the PYR-treated group than in the control group. These results not only suggest the importance of mMate1 in the efflux of organic cations into the urine and bile in mice but also the importance of canalicular efflux mediated by MATE proteins for the therapeutic efficacy of metformin. PYR is a potent inhibitor of human (h)MATE1 and hMATE2-K (Ki = 77 and 46 nM, respectively) and H+ and organic cation exchanger in human kidney BBMVs (Ki = 31 nM) in the presence of outward gradient of H+. Taken together, PYR can be used as a potent probe inhibitor of human MATE transporters.
The pharmacokinetics of drugs, which involves drug absorption, distribution, metabolism, and excretion, are indispensable for the understanding of pharmacological action and adverse reactions of drugs. Drug transporters mediate the tissue distribution of drugs and facilitate their elimination from the systemic circulation, and thereby play a significant role in the drug response (Giacomini and Sugiyama, 2005). The role of the organic cation transporter (OCT/SLC22) family has already been accumulated for the tissue uptake from the systemic circulation in the liver and kidney for type I organic cations, such as tetraethylammonium (TEA) and metformin (Wang et al., 2002; Wright, 2005; Koepsell et al., 2007). Cumulative studies have demonstrated the indispensable role of organic cation transporter 1 (Oct1/Slc22a1) and Oct2/Slc22a2 in the basolateral uptake of hydrophilic organic cations in rodent kidney (Jonker et al., 2003) and hOCT2 in human kidney (Motohashi et al., 2002). OCT1 is the predominant isoform in the liver, and its importance in the pharmacological action and the adverse reactions of metformin have been reported in animal and clinical studies (Wang et al., 2003; Shu et al., 2007, 2008).
Unlike the basolateral side that is mediated by facilitated diffusion, in vitro transport studies using brush-border membrane vesicles (BBMVs) from the kidney suggest involvement of active transport (Inui et al., 1985). Because the outward gradient of H+ stimulates the uptake of typical organic cations such as TEA by BBMV, an organic cation/H+ exchanger has been considered as the efflux transporter. Organic cation/carnitine transporter 1 (OCTN1/SLC22A4) and 2 (OCTN2/SLC22A5) were reported to mediate an exchange of H+ and TEA (Tamai et al., 1997; Wu et al., 1998). Animal and clinical studies suggest that these transporters mediate luminal efflux of drugs in the kidney. Juvenile visceral steatosis mice that exhibit systemic carnitine deficiency caused by a defect of Octn2 show a significant reduction in the renal clearance of TEA and cephaloridine accompanied by a significant increase in their kidney accumulation (Ohashi et al., 2001; Kano et al., 2009). In addition, a recent clinical study reported that healthy subjects homozygous for the L503F variant of OCTN1, which is associated with decreased OCTN1 function, exhibited smaller tubular secretion of gabapentin in the kidney compared with those homozygous for the reference allele (Urban et al., 2008). However, their importance in the luminal efflux of type I organic cations remains unclear.
On the other hand, there is a growing interest in multidrug and toxin extrusion (MATE) proteins as alternative candidate transporter for the efflux of type I organic cations in the kidney. MATEs were isolated from the kidney of various species (Otsuka et al., 2005; Masuda et al., 2006). MATE1 is expressed both in the liver and the kidney, whereas another isoform, MATE2/MATE2-K, is specific to the kidney (Moriyama et al., 2008; Terada and Inui, 2008). Both MATE1 and MATE2/MATE2-K exhibit polyspecific substrate specificities for a variety of organic cations, including TEA and metformin, with an extensive overlap (Tanihara et al., 2007). Because MATE1 and MATE2/MATE2-K are expressed in the apical membrane of the epithelial cells, they have been considered as further candidate proteins accounting for the luminal efflux of cationic drugs with an exchange of H+ in the kidney (Moriyama et al., 2008; Terada and Inui, 2008). A single-nucleotide polymorphism (SNP) in the intron (rs2289669) of hMATE1 has been found to be associated with an enhanced pharmacological effect of metformin (HbA1c reduction) in diabetic patients (Becker et al., 2009), but it does not cause a variation of the systemic exposure of metformin (Tzvetkov et al., 2009). Recently, Mate1(−/−) mice have been generated and in vivo studies using Mate1(−/−) mice showed a significant reduction in the urinary excretion of metformin compared with wild-type mice (Tsuda et al., 2009a).
The present report describes a potent and selective inhibitor of MATE transporters, pyrimethamine (PYR), and examines its effect on the urinary and biliary excretion of typical MATE1 substrates TEA and metformin, as well as the pharmacological response of metformin in mice. In mice, only one isoform, Mate1, corresponds to MATE1 and MATE2-K in humans (Hiasa et al., 2006), whereas another isoform identified in rodents, class III MATE protein, was found to be specifically expressed in the testis (Hiasa et al., 2007). The plasma concentrations and urinary and biliary excretion rates were determined under steady-state conditions. Kidney and liver concentrations were also determined to compare the intrinsic efflux activities across the brush-border membrane and canalicular membrane in mice, where Mate1 is expressed, with and without PYR. Furthermore, the effect of PYR on the plasma lactate concentration was examined to elucidate the importance of Mate1 in the pharmacological action of metformin.
Materials and Methods
[14C]TEA (3.2 mCi/mmol) was purchased from PerkinElmer Life and Analytical Sciences (Boston, MA), [14C]metformin (26 mCi/mmol) was purchased from Moravek Biochemicals (Brea, CA), and [3H]inulin (102 mCi/g) was purchased from American Radiolabeled Chemicals (St. Louis, MO). Unlabeled TEA, metformin, and LabAssay ALP were purchased from Wako Pure Chemicals (Osaka, Japan). An l-lactate assay lit was purchased from Biomedical Research Service Center (Buffalo, NY). All other chemicals used were commercially available and of analytical grade.
Male ddY mice were purchased from Japan SLC (Shizuoka, Japan). All animals were maintained under standard conditions with a reverse dark/light cycle and were kept for at least 7 days before pharmacological experiments. Food and water were available ad libitum. The mice used in the present study were from 8 to 9 weeks old. The studies were carried out in accordance with the guidelines provided by the Institutional Animal Care Committee (Graduate School of Pharmaceutical Sciences, The University of Tokyo, Tokyo, Japan).
Construction of Stable Transfectants of mMate1, mOct1, and mOct2 in HEK293 Cells and Cell Cultures.
mMate1 tagged with myc cDNA (NM_026183) was subcloned into pcDNA3.1(+) (Invitrogen, Carlsbad, CA). cDNA of mOct1 and mOct2 open reading frame were ligated into pENTR/d-TOPO vector (Invitrogen) and then recombined into pEF-DEST51 vector (Invitrogen) by LR recombination reaction of a Gateway system (Invitrogen) following the manufacturer's protocol. The plasmids were transfected into human embryonic kidney (HEK)293 cells by lipofection with FuGENE 6 transfection reagent (Roche Diagnostics, Basel, Switzerland) according to the manufacturer's protocol. The transfectants were selected by culturing in the presence of G418 sulfate (800 μg/ml; Invitrogen). Transporter-expressing or vector-transfected HEK293 cells were grown in low-glucose Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal bovine serum (Sigma-Aldrich, St. Louis, MO) and 1% antibiotic-antimycotic (Invitrogen) and incubated at 37°C with 5% CO2 and 95% humidity.
In Vitro Transport Study Using cDNA Transfectants.
hMATE1-HEK293, hMATE2-K-HEK293, hOCT1-HEK293, hOCT2-HEK293, mOctn1-HEK293, and mOctn2-HEK293 were already constructed previously (Busch et al., 1998; Tamai et al., 2000; Müller et al., 2005; Matsushima et al., 2009). Cells were seeded 72 h before the transport assay in poly-l-lysine- and poly-l-ornithine-coated 12-well plates at a density of 1.5 × 105 cells per well. For the transport study, the cell culture medium was replaced with culture medium supplemented with 5 mM sodium butyrate 24 h before the transport assay to induce the expression of transporters. The transport study was carried out as described previously (Hirano et al., 2004). Uptake was initiated by the addition of radiolabeled substrates, after the cells had been washed twice and preincubated with Krebs-Henseleit buffer at 37°C for 15 min. The Krebs-Henseleit buffer consisted of 118 mM NaCl, 23.8 mM NaHCO3, 4.8 mM KCl, 1.0 mM KH2PO4, 1.2 mM MgSO4, 12.5 mM HEPES, 5.0 mM glucose, and 1.5 mM CaCl2 and was adjusted to pH 7.4. Uptake was terminated at the designated times by the addition of ice-cold Krebs-Henseleit buffer after the removal of the incubation buffer. The cells were solubilized with NaOH overnight at 4°C and then neutralized with HCl. The radioactivity in aliquots was measured by liquid scintillation counting. For the examination of the inhibitor effect on the uptake by PYR, PYR was added to the incubation buffer. The protein concentration was determined using the Lowry method with bovine serum albumin as the protein standard as described previously (Lowry et al., 1951).
Transport Study Using Kidney Slices.
Uptake studies were carried out as described in a previous report (Hasegawa et al., 2002). Slices (300 μm in thickness) of whole kidneys from male ddY mice were kept in ice-cold oxygenated buffer before use. The buffer for the present study consisted of 120 mM NaCl, 16.2 mM KCl, 1 mM CaCl2, 1.2 mM MgSO4, and 10 mM NaH2PO4/Na2HPO4 adjusted to pH 7.5. Two slices were selected and then incubated at 37°C on a 12-well plate with 1 ml of oxygenated buffer containing 3.8 μM [14C]metformin in each well after preincubation of slices for 5 min at 37°C. After incubation for 15 min, slices were rapidly removed from the incubation buffer, washed twice with ice-cold buffer, blotted on filter paper, weighed, and dissolved in 1 ml of Soluene-350 (Packard Instruments) at 55°C for 12 h. The radioactivity in the specimens was determined in a scintillation counter after adding scintillation cocktail (Hionic Fluor; Packard Instruments). In every experiment, the uptake activity of Oct1 and Oct2 in the mouse kidney slices was checked with TEA as the positive control.
Transport Study Using Kidney BBMVs.
The BBMVs were isolated from the kidney of human and male ddY mice by the Ca2+ precipitation method as described previously (Shah et al., 1979), with several modifications. Human kidney samples were obtained from surgically nephrectomized patients with renal cell carcinoma at Tokyo Women's Medical University (Tokyo, Japan). The protocol was approved by the Ethics Review Boards of both the Graduate School of Pharmaceutical Sciences, The University of Tokyo (Tokyo, Japan), and Tokyo Women's Medical University (Tokyo, Japan). All participants provided written informed consent. Samples were stored in Dulbecco's modified Eagle's medium and kept on ice, immediately after kidney excision.
The kidney samples were homogenized in a 30-fold (w/v) volume of solution containing 30 mM mannitol and 10 mM CaCl2, buffered with 10 mM Tris-HEPES, pH 7.4; and after standing for 15 min, the homogenate was centrifuged at 3000 rpm (TS-7 rotor; Tomy Seiko Co, Ltd.. Tokyo, Japan) for 15 min. Then, the supernatant was centrifuged at 40,000 rpm (50.2 Ti rotor; Beckman Coulter, Fullerton, CA) for 60 min. The vesicle pellet was resuspended in an experimental buffer to give a final protein concentration of 7 mg/ml using a 25-gauge needle. The experimental buffer consisted of 100 mM mannitol, 100 mM potassium chloride, and 10 mM MES, pH 6.0, or HEPES, pH 8.0, and the pH was adjusted with potassium hydroxide. The enrichment was determined by alkaline phosphatase activity using the LabAssay ALP (Wako Pure Chemicals), which was 6- to 8-fold higher compared with the homogenate. All steps were performed either on ice or at 4°C.
The uptake of [14C]metformin by BBMVs was measured using a rapid filtration technique. The uptake was initiated by the addition of 80 μl of a transport buffer containing 50 μM [14C]metformin (final concentration, 40 μM) to 20 μl of membrane suspension at 20°C. After 30 s, the incubation was terminated by diluting the reaction mixture with 1 ml of ice-cold buffer. The mixture was then poured immediately onto HA filters (0.45 μm; Millipore, Billerica, MA), and the filters were washed twice with 5 ml of ice-cold buffer. In separate experiments, the nonspecific association of the ligand with filters was estimated by the absence of membrane vesicles. This value was subtracted from the uptake data to evaluate the specific uptake of the ligand. The radioactivity trapped in membrane vesicles was determined by liquid scintillation counting.
Transport Study Using Bile Canalicular Membrane Vesicles.
The CMVs were isolated from the liver of male ddY mice as described previously (Niinuma et al., 1999; Hayashi and Sugiyama, 2007). The vesicle pellet was resuspended in an experimental buffer using a 25-gauge needle. The experimental buffer consisted of 100 mM mannitol, 100 mM potassium chloride, and 10 mM MES, pH 6.0, or HEPES, pH 8.0, and the pH was adjusted with potassium hydroxide. The uptake of [14C]metformin (100 μM) for 2 min by CMVs was measured at 37°C using a rapid filtration technique as described above.
Steady-State Infusion Study of TEA and Metformin in Mice Treated with and without PYR.
After anesthesia with intraperitoneal sodium pentobarbital (51.8 mg/kg for TEA) or isoflurane (for metformin), the urinary bladder or bile duct was catheterized. In the PYR-treated group, a bolus dose of PYR (0.2, 0.5, or 2 μmol/kg in 0.9% NaCl and 10% ethanol) was administered via the jugular vein 30 min before the start of infusion. [14C]TEA (128 nmol/min/kg) and [3H]inulin (49 pmol/min/kg) or metformin (200 nmol/min/kg) were then also infused via the jugular vein. Blood samples were collected via the jugular vein at 30, 50, 70, and 90 min (for [14C]TEA) or at 60, 80, 100, and 120 min (for metformin) after administration and centrifuged to obtain plasma. Urine specimens were collected at 30 to 50, 50 to 70, and 70 to 90 min (for [14C]TEA), and urine or bile specimens were collected at 60 to 80, 80 to 100, and 100 to 120 min (for metformin). At the end of the experiment, the kidneys and liver were removed. To determine the glomerular filtration rate (GFR), [3H]inulin (for TEA) or creatinine (for metformin) renal clearance was measured. Drug concentration was determined by liquid scintillation counting (for TEA and inulin) and LC/MS analysis (for creatinine and metformin).
Quantification of Metformin by LC/MS in Biological Samples.
The kidney and liver were homogenized in a 4-fold volume of phosphate-buffered saline. The urine specimens were diluted with a 100-fold volume of water. All specimens (10 μl) were mixed with 10 μl of methanol and 80 μl of acetonitrile. Mixed solutions were centrifuged at 20,000g for 10 min. The supernatants were mixed with a 4-fold volume of water, centrifuged at 20,000g for 10 min, and subjected to LC/MS analysis. For creatinine, the plasma specimens (10 μl) were mixed with 20 μl of acetonitrile and centrifuged at 20,000g for 10 min. The supernatants (10 μl) were evaporated, and the pellets were reconstituted with 80 μl of water. The reconstituted samples were then centrifuged at 20,000g for 10 min to remove particles, and an aliquot of the supernatants was used for LC/MS analysis. The urine specimens were diluted with a 2000-fold volume of water and centrifuged twice at 20,000g for 10 min. The supernatants were then used for LC/MS analysis.
An LCMS-2010EV system equipped with a Prominence LC system (Shimadzu, Kyoto, Japan) was used for the analysis. The interface voltage was −3.5 kV, and the nebulizer gas (N2) flow was 1.5 l/min. The heat block and curved desolvation line temperatures were 200 and 250°C, respectively. Detailed LC conditions and mass-to-charge ratios are shown in Supplemental Table 1.
The data obtained from the inhibition study can be fitted to the following equation to calculate the inhibition constant (Ki): where CLuptake (+inhibitor) and CLuptake (control) are the uptake clearance determined in the presence or absence of the inhibitor, Pdif is the nonsaturable uptake clearance, and I is the concentration of the inhibitor. Fitting was performed by the nonlinear least-squares method using a MULTI program (Yamaoka et al., 1981), as well as the Damping Gauss-Newton Method algorithm for fitting.
The fractional urinary excretion ratio (Furine), the fractional biliary excretion ratio (Fbile), the total body clearance (CLtot, plasma), the renal clearance with respect to the plasma concentration (CLrenal, plasma), the secretion clearance with respect to the concentration in the kidney (CLrenal, kidney), the biliary clearance with respect to the plasma concentration (CLbile, plasma), and the biliary clearance with respect to the concentration in the liver (CLbile, liver) were calculated using the following equations: where V and I represent urinary or biliary excretion rate and infusion rate. Dose and X represent the amount of ligand administered in 1 h and excreted into urine or bile from 30 to 90 min (for TEA) or 60 to 120 min (for metformin), respectively. AUCp represents the area under the time-plasma concentration curve for ligands from 30 to 90 min (for TEA) or 60 to 120 min (for metformin). The apparent tissue-to-plasma concentration ratio (Kp, tissue) was calculated using the following equation: where Ctissue and Cp represent tissue and plasma concentrations of ligands at 90 or 120 min after administration for TEA or metformin, respectively. GFR was determined by [3H]inulin (for TEA) or creatinine (for metformin) renal clearance.
Plasma Lactate Levels after Metformin Treatment in Mice.
Because isoflurane causes an increase in the plasma lactate level in mice, pentobarbital was used for anesthesia. Mice were anesthetized with intraperitoneal sodium pentobarbital (51.8 mg/kg). PYR (2 μmol/kg i.v.) was given to mice 30 min before starting the intravenous infusion of metformin at the rate of 15 μmol/min/kg for 4 h through the jugular vein. Blood samples were collected in the same way at 0 (just before administration), 150, 180, 210, and 240 min. The blood was centrifuged and the plasma samples were used for lactate determination with an l-lactate assay kit (Biomedical Research Service Center) according to the manufacturer's protocol. After sampling at 240 min, the mice were sacrificed, and the liver was removed immediately. Plasma and liver concentrations of metformin were determined as described above.
Data are presented as the mean ± S.E. Student's two-tailed unpaired t test and a one-way analysis of variance followed by Dunnett's post hoc test were used to identify significant differences between groups where appropriate.
Inhibition Potency and Selectivity of PYR for the Mouse Renal Organic Cation Transporters.
mMate1-HEK constructed in this study exhibited similar transport characteristics (Supplemental Fig. 1) to results reported previously (Hiasa et al., 2006). Intracellular accumulation of TEA, metformin, cimetidine, creatinine, carnitine, and acyclovir were significantly greater in mMate1-HEK compared with mock cells (Supplemental Fig. 2). The effect of PYR on mMate1, mOct1, mOct2, mOctn1, and mOctn2 was investigated for the uptake of their model substrate (TEA). TEA uptake by mMate1, mOct1, and mOct2 was strongly inhibited by PYR, whereas it had little or negligible effect on TEA transport by mOctn1 and mOctn2 (Fig. 1). The Ki values are summarized in Table 1. The results clearly show that PYR is a 25 and 42 times more potent inhibitor for mMate1 than for mOct1 and mOct2, respectively.
Inhibitory Effect of PYR on Metformin Uptake by Mouse Kidney Slices.
The inhibitory effect of PYR on metformin uptake by mouse kidney slices is shown in Fig. 2A. Saturable uptake of metformin was clearly observed in mouse kidney slices. PYR significantly inhibited the uptake of metformin at 10 and 30 μM.
Inhibitory Effect of PYR on Metformin Uptake by Renal BBMVs and CMVs.
The effect of PYR on metformin uptake in the presence of a H+ gradient by renal BBMVs from mouse kidney was examined. Nonspecific binding to the filters, which accounted for 12% of the total uptake, was identical in the presence and absence of PYR. The uptake of metformin determined at pH 8.0 was markedly reduced in the presence of excess metformin (16 mM), showing saturation, and it was inhibited by 8 μM PYR (Fig. 2B). PYR significantly inhibited the uptake of metformin by BBMVs in a concentration-dependent manner, with a Ki value of 41 ± 6 nM (Fig. 2C).
Similarly, the uptake of metformin by CMVs determined at pH 8.0 was markedly reduced in the presence of excess metformin (10 mM), showing saturation, and it was inhibited by 10 μM PYR (Fig. 3). Nonspecific binding to the filters, which accounted for 19% of the total uptake, was identical in the presence and absence of PYR.
Effect of PYR on the Urinary and Biliary Excretion of TEA and Metformin in Mice.
According to our preliminary study, PYR shows a long half-life in the systemic circulation (>4 h); therefore, PYR was given to mice by a bolus injection 30 min before starting the infusion of substrate drugs, resulting in a near-constant plasma concentration during the experiments. The plasma, kidney, and liver concentrations of PYR were 0.230 ± 0.037, 4.99 ± 0.49, and 2.79 ± 0.37 μM, respectively, at a dose of 2 μmol/kg (n = 4). Considering that unbound fractions of PYR in the plasma, kidney, and liver were 8.1, 2.6, and 1.1%, respectively, the unbound concentrations of PYR in the plasma, kidney, and liver are estimated at approximately 10 to 30, 50 to 150, and 20 to 50 nM.
TEA was administered to mice by constant intravenous infusion, and the concentrations of TEA in the plasma and kidney, as well as the urinary excretion rate, were measured under steady-state conditions with and without pretreatment of PYR. PYR treatment caused a significant increase in the kidney-to-plasma ratio (Kp, kidney), even though both plasma concentrations and urinary excretion rates of TEA were similar (Fig. 4). Kinetic parameters of TEA in control and PYR-treated mice are summarized in Table 2. The secretion clearance with regard to the kidney concentration was significantly decreased by PYR in a dose-dependent manner from 0.5 to 2 μmol/kg (Table 2).
Metformin was administered to mice with and without PYR pretreatment (2 μmol/kg) through constant intravenous infusion. The effect of PYR on the urinary and biliary excretion of metformin was examined in separate experiments. PYR significantly increased the kidney concentration of metformin (3.7-fold), whereas it did not affect the plasma concentrations and urinary excretion rates of metformin (Fig. 5). The secretion clearance of metformin with respect to the kidney concentration was significantly decreased by PYR (Table 3). Metformin was found to undergo biliary excretion, the rate of which was 621-fold smaller than its urinary excretion rate (Table 3). PYR also significantly increased the liver concentration of metformin (2.6-fold; Fig. 5). Because the biliary excretion rate of metformin was unchanged, the biliary clearance with regard to the liver concentration of metformin was significantly decreased in mice treated with PYR (Table 3). In mice whose bile duct was cannulated, the total body clearance was 60% of that in mice whose bladder was cannulated by an unknown reason. The operation of bile duct ligation may affect the renal blood flow rate because the renal clearance of metformin is blood flow limited.
Effect of PYR on Plasma Lactate Concentration in Mice Treated with Metformin.
The time profiles of the plasma lactate concentrations on constant infusion of a dose of 15 μmol/min/kg metformin are shown in Fig. 6A, and the simultaneously determined plasma concentrations of metformin are shown in Fig. 6B. There was a marked difference in the response to metformin between the metformin-administered group and the combined metformin- and PYR-administered group after 240 min of intravenous infusion of metformin (Fig. 6A), although the plasma concentrations of metformin were comparable in both sample sets (Fig. 6B). At 240 min, the plasma lactate concentration was 1.8-fold greater in mice receiving both metformin and PYR than that in mice receiving only metformin (54.8 ± 4.4 versus 31.2 ± 2.6 mg/dl). After the mice were sacrificed, the concentration of metformin in the liver showed a significant increase in PYR-treated mice (1.05 ± 0.14 versus 2.98 ± 0.30 μmol/g liver, p < 0.01).
Inhibition Potency and Selectivity of PYR for Human Renal Organic Cation Transporters, and Metformin Uptake by BBMVs from Human Kidney Samples.
Significant uptake of TEA, metformin, cimetidine, carnitine, and acyclovir were observed both in hMATE1- and hMATE2-K-expressed cell lines. Creatinine was revealed to be specific substrate of hMATE1 but not of hMATE2-K (Supplemental Fig. 2). The effect of PYR on hMATE1, hMATE2-K, hOCT1, and hOCT2 was investigated for the uptake of their model substrate (TEA). PYR inhibited the uptake of TEA by hMATE1, hMATE2-K, hOCT1, and hOCT2 in a concentration-dependent manner (Fig. 7). The Ki values are summarized in Table 1. PYR is not an isoform-selective inhibitor of hMATE1 and hMATE2-K. PYR is a more potent inhibitor for hMATE1 and hMATE2-K than for the basolateral organic cation transporters hOCT1 and hOCT2.
The effect of PYR on the metformin uptake by BBMVs from human kidney was examined. Nonspecific binding to the filters, which accounted for 12% of the total uptake, was identical in the presence and absence of PYR. The uptake of metformin by BBMVs was determined in the presence of a H+ gradient, and it was shown that the uptake determined at pH 8.0 was markedly reduced in the presence of excess metformin (24 mM; Fig. 8A). PYR significantly inhibited the uptake of metformin by BBMVs in a concentration-dependent manner, with a Ki value of 31 ± 4 nM in humans (Fig. 8B).
We examined the role of mMate1 in the disposition of the typical type I organic cations TEA and metformin and in the pharmacological action of metformin by using a potent MATE inhibitor, PYR. MATE transporters are organic cation transporters driven by an exchange of H+ and have been a candidate transporter mediating the efflux of type I organic cations in the liver and kidney. Inhibition of MATE proteins could prolong the systemic and tissue exposure of its substrate drugs, such as metformin, enhancing the pharmacological action and the adverse reaction.
First, we examined the selectivity and inhibition potency of PYR in cDNA transfectants, mouse kidney slices, and BBMVs. Among the organic cation transporters, PYR could specifically block mMate1 (Fig. 1). Even though PYR inhibited both mOct1 and mOct2, the potency was quite low compared with mMate1 (Table 1). Thus, PYR can specifically inhibit the luminal efflux mediated by mMate1 without inhibiting the uptake process mediated by mOct1 and mOct2. In fact, PYR could not inhibit the uptake of metformin by kidney slices at the concentration (1 μM) adequately greater than the Ki value for mMate1 (Fig. 2A). The uptake of metformin by BBMVs determined in the presence of a H+ gradient was completely inhibited by PYR (Fig. 2B), with a Ki value similar to that observed in mMate1-expressed HEK293 cells (Fig. 2C). Therefore, mMate1 accounts for the major part of the H+-coupled organic cation transport in the kidney. Furthermore, CMVs shows saturable uptake of metformin in the presence of outward gradient of H+, which was inhibited by PYR. This is in a good agreement with the membrane localization of mMate1 in the liver (Hiasa et al., 2006).
Second, the effect of PYR was examined in vivo in mice. PYR could significantly increase the kidney-to-plasma ratio of TEA and metformin, which is ascribed to an inhibition of the efflux into the urine across the brush-border membrane because the secretion clearance with respect to the kidney concentrations was significantly reduced in PYR-treated mice (Tables 2 and 3). Absence of an effect of PYR on the renal clearance with respect to the plasma concentrations in spite of significant inhibition of the luminal efflux seems to contradict the fact that both TEA and metformin are exclusively eliminated into the urine by tubular secretion as well as by glomerular filtration (Tables 2 and 3). The following points can explain this discrepancy: 1) the renal elimination of TEA and metformin is blood flow limited, and 2) the fraction inhibited by PYR is not large enough to disturb the directional transport across the proximal tubules involving the transporters both in the uptake and in the subsequent efflux processes. In fact, the renal clearance of TEA and metformin was close to the previously reported renal blood flow rate (65 ml/min/kg; Davies and Morris, 1993). The unbound concentration of PYR in the kidney was 150 nM at the highest dose examined, which was similar to the Ki value of PYR for mMate1, and thus, PYR could reduce mMate1-mediated efflux to 40% of the control.
Third, PYR highlighted the significance of mMate1 in the liver (Fig. 5F). Metformin was found to be eliminated into the bile, although the biliary excretion made a negligible contribution to the systemic elimination (Table 3). PYR significantly inhibited the efflux of metformin across the bile canalicular membrane and thereby increased the liver-to-plasma concentration ratio (Table 3). In fact, Na+/H+ exchanger 3 protein is present on the canalicular membrane of hepatocytes and generates a H+ gradient (Mennone et al., 2001). Considering that the pharmacological target of metformin for inhibition of gluconeogenesis is located inside the hepatocytes, it is reasonable to announce that PYR could enhance the pharmacological action of metformin (Fig. 6). Since a marked reduction in the hepatic concentration of metformin caused by the deletion of the Oct1 gene resulted in a reduced effect of metformin on AMP-activated protein kinase phosphorylation (Shaw et al., 2005), PYR will increase AMP-activated protein kinase phosphorylation after metformin treatment. Recently, it was reported that a SNP in the intron (rs2289669) of hMATE1 is associated with enhanced pharmacological effects of metformin (HbA1c reduction) in diabetic patients (Becker et al., 2009). Because hMATE1 is also expressed in the canalicular membrane of human hepatocytes (Otsuka et al., 2005), it is most probable that MATE1 can be a determining factor in the pharmacological action of metformin through the regulation of liver concentrations in human. In addition, MATE1 can potentially be the site of a drug-drug interaction between metformin and concomitant drugs. For example, cimetidine has been reported to inhibit the renal elimination and thereby enhance the pharmacological action of metformin in healthy volunteers (Somogyi et al., 1987). Considering the fact that cimetidine is a clinically relevant inhibitor of MATE1 (Matsushima et al., 2009; Tsuda et al., 2009b), it is possible that the interaction involves the inhibition of the canalicular efflux as well as inhibition of renal elimination. In vitro inhibition studies using cDNA transfectant of hMATE1 and concomitant drugs will contribute to avoid severe drug-drug interaction by comparing the inhibition constant with the unbound concentration of the concomitant drugs at the clinical dose. It should be noted that, unlike in mice, the human kidney expresses two MATE transporters, MATE1 and MATE2-K. Since MATE1 SNP was not associated with the variation of systemic exposure of metformin, (Tzvetkov et al., 2009), presumably because of compensation of MATE2-K, specific inhibition of MATE1 in the kidney by concomitant drugs will not affect the renal elimination of metformin.
Finally, PYR was found to be a significantly potent inhibitor of MATE1 and MATE2-K (Table 1). The uptake of metformin by BBMVs in the presence of a H+ gradient was completely inhibited by PYR, with a Ki value similar to those for MATE1 and MATE2-K (Fig. 8). Thus, the metformin transport driven by an exchange of H+ is probably accounted for by MATE1 or MATE2-K. PYR is used clinically as an antimalarial and antitoxoplasmic drug (Daraprim, fansidar (fansidar is the sulfadoxine-pyrimethamine combined drug)). The unbound plasma concentration of PYR at the clinical dosage is 200 nM on average, adequately greater than the Ki values for hMATE1 and hMATE2-K but lower than its Ki values for hOCT1 and hOCT2 (Table 1). Therefore, the clinical dose of PYR would be sufficient to inhibit MATE1 and MATE2-K in humans, without affecting the uptake process. To our knowledge, there is no report on any drug-drug interaction involving PYR except creatinine. PYR decreases the renal clearance of endogenous creatinine without affecting the renal clearance of inulin, a true GFR marker (Opravil et al., 1993). Creatinine is known to undergo tubular secretion in the kidney that involves hOCT2 as well as hMATE1 and hMATE2-K (Urakami et al., 2004; Tanihara et al., 2007). Reduction of renal clearance of creatinine may be ascribed to an inhibition of MATE1 or MATE2-K in the kidney by PYR.
In summary, MATE proteins play a significant role in the urinary and biliary excretion of metformin. MATE1 is particularly important for the pharmacological action of metformin through the regulation of liver concentration. PYR is a potent and selective inhibitor of mMate1, MATE1, and MATE2-K and is probably a potent probe inhibitor to investigate the significance of MATE1 and MATE2-K in the disposition of type I organic cations in humans.
We thank Drs. Yukio Kato and Akira Tsuji (Kanazawa University, Ishikawa, Japan) for providing us HEK293 cells expressing mOctn1 and mOctn2 and Larissa Kogleck (University College London, London, UK) for efforts to improve this manuscript.
- Received November 11, 2009.
- Accepted January 6, 2010.
This study was supported by the Translational Research Promotion Project, New Energy and Industrial Technology Development Organization of Japan (in 2008 to Y.S.); the Ministry of Education, Culture, Sports, Science and Technology, Jpan [Grant-in-Aid for Scientific Research on Priority Areas KAKENHI 20056005] (to H.K.); Japan Research Foundation for Clinical Pharmacology (to H.K.); and The Smoking Research Foundation (to Y.M.).
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- organic cation transporter
- brush-border membrane vesicle
- organic cation/carnitine transporter
- multidrug and toxin extrusion
- single-nucleotide polymorphism
- human embryonic kidney
- 2-(N-morpholino)ethanesulfonic acid
- canalicular membrane vesicle
- glomerular filtration rate
- liquid chromatography/mass spectrometry.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics