Renal tubular transporter-mediated interactions between mirogabalin and cimetidine in rats

Abstract Cimetidine at a clinical dosage decreased the renal clearance (CLr) of mirogabalin in humans by inhibition of renal secretion. Mirogabalin is a substrate of human OAT1/3, OCT2, MATE1 and/or MATE2-K. To clarify the mechanism behind the above interaction, it was investigated whether cimetidine inhibits the process of mirogabalin uptake at the basolateral side or the process of its efflux at the apical side in rat kidney in vivo. Cimetidine was administered to rats by a constant infusion to achieve an unbound plasma concentration of 7.0 μM and examine its effect on the renal disposition of [14C]metformin, [3H]p-aminohippuric acid (PAH), and [14C]mirogabalin. Cimetidine significantly induced the intrarenal accumulation of radioactivity (Kp, kidney) and decreased the renal clearance (CLr) of [14C]mirogabalin. These effects resulted in significantly decreased total clearance (CLt). Kp, kidney, and CLr of [14C]metformin, except CLt, were also affected, but no parameters of [3H]PAH were affected by cimetidine. These findings clarified that an unbound plasma concentration of cimetidine of 7.0 μM inhibited the apical efflux not the basolateral uptake of [14C]mirogabalin in rat kidney, suggesting that mirogabalin/cimetidine interaction was caused by inhibiting the apical efflux transporter, human MATE1 and/or MATE2-K, not the basolateral uptake transporter, human OCT2, in the kidney.


Introduction
Mirogabalin is a potent and selective ligand of the a 2 d subunit of voltage-gated Ca 2þ channels and was developed for the management of peripheral neuropathic pain (Domon et al. 2018;Kitano et al. 2019). It has been approved for the treatment of peripheral neuropathic pain including diabetic peripheral neuropathic pain and postherpetic neuralgia in Japan (Baba et al. 2019;Kato et al. 2019). Mirogabalin shows potent and selective binding affinities for the human and rat a 2 d subunits and slower dissociation rates for the a 2 d-1 subunit than for the a 2 d-2 subunit (Domon et al. 2018;Kitano et al. 2019). Clinical pharmacokinetic (PK) parameters of mirogabalin after oral administration were determined in phase 1 studies as a single-ascending-dose (SAD) study and a multiple-ascending-dose study of healthy subjects of various ethnicities Jansen et al. 2018). The SAD study showed that plasma exposure increased in a dose-proportional manner from 3 to 75 mg and renal excretion of mirogabalin accounted for 63%-71% of the investigated doses. Absorption, metabolism, and excretion of [ 14 C]mirogabalin at a dose of 30 mg in humans were also investigated. The results clarified that oral absorption was almost complete, the main clearance pathway was renal excretion with active secretion and the other clearance pathway accounting for around 20% of the dose was glucuronidation at the carboxyl and amino moiety of mirogabalin (Yamamura et al. 2021). The PK behaviour of mirogabalin was found to be similar among humans, rats, and monkeys. Absolute bioavailability after the oral administration of mirogabalin at 3 mg/kg was 97.6% in rats and 85.2% in monkeys. Urinary recovery of radioactivity after oral administration of [ 14 C]mirogabalin was 94.5% in rats and 77.7% in monkeys and mirogabalin was only detected in the urine. A small part of the orally administered dose of mirogabalin was metabolised via glucuronidation at the amine and carboxylic acid moiety and oxidation as the primary metabolic pathway (Yamamura et al. 2022a). The high renal excretion of mirogabalin indicated filtration and secretion in rats, monkeys, and humans (Yamamura et al. 2022b).
In vitro studies using transporter-expressing cells and human kidney slices indicated that human (h)OAT1, hOAT3, and hOCT2 as basolateral uptake transporters and hMATE1 and hMATE2-K as apical efflux transporters expressed on proximal tubular cells were involved in the renal secretion (Yamamura et al. 2022b). Based on these findings, a clinical drug interaction study of mirogabalin with cimetidine and probenecid was conducted (Tachibana et al. 2018). In that study, 500 mg of cimetidine was administered every 6 h (QDS: four times a day) from Day 1 through 4, while mirogabalin was administered concomitantly on Day 2. The mean unbound maximum plasma concentration of cimetidine after the 5th administration on Day 2, when mirogabalin was administered, was approximately 9.1 mM and the trough plasma concentration of cimetidine at the 5th administration was almost the same as at the 6th and 7th administrations, indicating that the plasma concentration of cimetidine had reached a steady-state condition. Cimetidine decreased the renal clearance (CLr) of mirogabalin and extended its plasma half-life. CLr is determined using the plasma concentrationtime profile (area under plasma concentration: AUC) and amount of urinary excretion (Ae) (CLr ¼ Ae/AUC). The inhibition of either basolateral uptake transporters or apical efflux transporters expressed on proximal tubular cells decreases tubular secretion, resulting in reductions of Ae and CLr. In other words, decreases of Ae and CLr do not indicate which transporters at the basolateral or apical side are inhibited. We compared the inhibition constant (Ki) values of cimetidine for hOCT2 (95-146 lM), hMATE1 (1.1-3.8 lM), and hMATE2-K (2.1-6.9 lM) (Tsuda et al. 2009;Ito et al. 2012) and the observed maximum unbound plasma concentration (9.1 lM) in the discussion of previous articles to suggest inhibition of apical efflux transporter, hMATE1 and/or hMATE2-K, but not basolateral uptake transporter, hOCT2 (Tachibana et al. 2018;Yamamura et al. 2022b). In vitro inhibition data (IC 50 and Ki) of cimetidine on human OCT2, MATE1, MATE2-K, OAT1, and OAT3, which are involved in renal active secretion of mirogabalin, reported in the literature from the University of Washington DDI data base (https://didb.druginteractionsolutions.org) are listed in Supplementary Appendix Table 1. The observed maximum unbound plasma concentration of cimetidine (9.1 lM) at 500 mg QDS is compared to the in vitro inhibition data (9.1 lM/IC 50 or Ki) in the table. The cimetidine plasma concentration is higher than all reported inhibition data of MATE1 (metformin as probe substrate) and lower than those of OCT2 (metformin as probe substrate), OAT1 (p-aminohippuric acid as probe substrate), and OAT3 (estrone 3-sulfate as probe substrate) except one reported IC 50 value on OCT2 (2.9 lM: No actual experimental data of this value generated at Merck & Co. was explained in the literature).
The kidney is an organ to produce urine from blood and to excrete urine from the body. The nephron is a functional unit to do it. The nephron is consisted of the renal corpuscle and renal tubule. The transporters involved in the renal secretion of mirogabalin are expressed in renal proximal tubular cells. Plasma-free fraction of cimetidine could inhibit OAT1, OAT3, OCT2 expressed on the basolateral membrane and intracellular free fraction of cimetidine could inhibit MATE1, MATE2K expressed on the apical membrane of renal proximal tubular cells. The plasma-free fraction of cimetidine can be determined, however, the intracellular free fraction of cimetidine in the renal proximal tubular cells could hard to do actually.
If it would be possible to determine the intracellular mirogabalin concentration in renal proximal tubular cells in a clinical drug interaction study, this knowledge could be used to clarify whether cimetidine inhibits transporters on the basolateral or apical side.
Specifically, a decrease of intracellular mirogabalin concentration in renal proximal tubular cells would indicate inhibition of the basolateral uptake transporter hOCT2, while an increase of this concentration in the cells would indicate inhibition of the apical efflux transporter hMATE1/2-K. The rationale for this is supported by the nephrotoxicity of cisplatin, which is known to be a substrate of hOCT2 and hMATE1/2-K. The inhibitors of hMATE1/2-K enhanced the renal accumulation of cisplatin and induced cisplatin nephrotoxicity (Li et al. 2013), and the impaired function of hOCT2 reduced this nephrotoxicity (Filipski et al. 2009).
In the present study, the intrarenal (kidney and cortex) concentration of [ 14 C]mirogabalin reflected by radioactivity was measured in rats treated with [ 14 C]mirogabalin and cimetidine infused intravenously at a constant rate. Based on the hypothesis described above, the intrarenal concentration of [ 14 C]mirogabalin was used as instead of the intracellular concentration of [ 14 C]mirogabalin in renal proximal tubular cells. Plasma and urinary concentrations of [ 14 C]mirogabalin were also measured, and renal excretion rate (ER), CLr, renal clearance with respect to the intrarenal concentration (CLr/C kidney ), and systemic plasma clearance (CLt) were calculated. We also investigated the effects of cimetidine on the renal disposition of [ 14 C]metformin as rat (r)Oct2 and rMate1 probe (Kimura et al. 2005;Terada et al. 2006) and [ 3 H]p-aminohippuric acid (PAH) as rOat1, rOat3, and rMrp2 probe (Liu et al. 2012).

Animals and animal welfare
Male F344 (F344/DuCrlCrlj) rats were purchased from Charles River Japan Co., Ltd., and were fed laboratory chow (CE-2; CLEA Japan, Inc.) and given free access to water. The rats were housed in a temperature-and humidity-controlled rooms (19-25 C, 30%-70%), with a 12-h light/dark cycle, for more than 1 week before use. The body weights applied for determining the PK parameters of cimetidine were from 167 to 173 g (n ¼ 3 and n ¼ 1) and those for investigating the effects of cimetidine on the renal and systemic disposition of inulin, PAH, metformin, and mirogabalin were from 235 to 335 g (n ¼ 3 or 4). The sample size (n ¼ 3 or 4 in each group) was decided as the minimum number to show the difference statistically between the control and cimetidine treatment. Animals used for this study was 29. All experimental procedures were performed in accordance with the in-house guidance of the Institutional Animal Care and Use Committee of Daiichi Sankyo Co., Ltd.

Determination of plasma unbound fraction of mirogabalin
Aliquots (230 mL/tube) of the [ 14 C]mirogabalin-added plasma collected from male F344 rats (final concentrations of mirogabalin of 0.1, 1, and 10 mg/mL adjusted by mirogabalin besylate) were placed in polyallomer tubes (0.23 PA; Hitachi, Tokyo, Japan) at room temperature. After 5 min, a 30-mL aliquot of the plasma sample was collected. The remaining plasma sample (200 mL/tube) was centrifuged (200,000 Â g, 4 C, 16 h) and a 50-mL aliquot of the supernatant was collected. Experiments were conducted in triplicate at each concentration. The plasma sample before centrifugation (total) and the supernatant (free fraction) were dissolved in tissue solubiliser (Soluene-350; PerkinElmer Inc.) and mixed with a liquid scintillator (Hionic-Fluor; PerkinElmer Inc.). The radioactivity of each sample was measured using a liquid scintillation counter (Tri-Carb 2900TR; PerkinElmer, Inc.). The plasma unbound fraction of mirogabalin was calculated as the free fraction (dpm/mL) divided by the total (dpm/mL).
Measurement of cimetidine concentration in plasma and measurements of radioactivity in plasma, urine, liver, renal cortex, and kidney The plasma sample (20 lL) was diluted 2-fold using rat control plasma and was added to 4-fold methanol including cimetidine-d3. Acetonitrile (280 lL) was further added to the sample and mixed well. The sample was then filtered using a 0.2 lm filter (Captiva ND; Agilent Technologies, Inc.). An aliquot of the filtrate was subjected to LC-MS/MS. The LC system was a Nanospace SI-2 (Shiseido Co., Ltd.). The column used was a TSKgel ODS-100V (3 lm, 2 mm ID Â 100 mm; Tosoh Bioscience Japan) and the column oven was set at 50 C. For the mobile phase, a gradient method was used. Solvent A was 95% acetonitrile including 5 mM ammonium acetate, and solvent B was 5% acetonitrile including 5 mM ammonium acetate. Its flow rate was 0.5 mL/min. The ratios of solvents A and B were 5:95 at 0 min, 50:50 at 0.5 min, and 95:5 at 2 min, which was then maintained for 1 min. MS analysis was performed using an API 4000 system (AB SCIEX) and the ionisation mode was ESI-positive. The curtain gas and its set value were nitrogen and 40 psi, respectively. The collision gas and its set value were nitrogen and 6 psi, respectively. The ion spray voltage and ion source temperature were 5500 V and 600 C, respectively. The ion source gases 1 and 2 and each of their set values were air and 50 psi and air and 80 psi, respectively. The interface heater was on. Tandem mass spectrometric analysis was performed using nitrogen as a collision gas with a collision energy of 20 eV for cimetidine and cimetidine-d3. Based on the fullscan MS/MS spectrum of both, the most abundant ions were selected and the mass spectrometer was set to monitor the transitions of the precursors to the product ions as follows: m/z 253.5 ! 159.2 for cimetidine and m/z 256.5 ! 162.2 for cimetidine-d3. Data acquisition and analysis were performed using Analyst 1.6.1 (AB SCIEX). The cimetidine concentration range of the standard curve was from 5 to 5000 ng/mL and the quality control samples were set at 20, 200, and 2000 ng/mL. The standard curve was plotted with the y-axis representing the theoretical value and the x-axis representing the peak area ratio (cimetidine/internal standard). Cimetidine in the plasma samples was measured by regression of the standard curve.
To measure radioactivity, the tissue solubiliser and the liquid scintillator were added to plasma and urine. The liver, renal cortex, and kidney were solubilised with the tissue solubiliser with heating at 40 C for overnight. Hydrogen peroxide and isopropanol were added for decolorisation. The liquid scintillator was added to tissue samples. The radioactivity of each sample was measured using the liquid scintillation counter. The external standard source method was used to correct for counting efficacy.

Calculation of pharmacokinetic parameters of cimetidine and determination of its constant infusion rate
The plasma unbound concentration of cimetidine to inhibit rMate1 was supposed to be 7.0 lM and the total concentration was 10 lM (unbound fraction of cimetidine in rat plasma: 0.71 ± 0.04; Ikemura et al. 2013). The supposed plasma unbound concentration of cimetidine was lower than the inhibition constant (Ki) values for rat Oct2 (9.4 lM, Urakami et al. 1998;632 lM, Umehara et al. 2007) and higher than the Ki for mouse Mate1 (1.4-3.6 lM, Ito et al. 2012) and Km for the rat Mate1-mediated uptake of cimetidine (3.01 lM, Ohta et al. 2006). The CLt and plasma concentration at time 0 (C0) of cimetidine were calculated using non- compartmental model analysis with the Winnonlin Noncompartmental analysis program (Pharsight) based on the plasma concentration of cimetidine in rats after intravenous bolus administration at 8 mg/kg (n ¼ 3). V0 was calculated using C0 and dose (V0 ¼ dose/C0). The calculated CLt and V0 of cimetidine in rats were 47.8 mL/min/kg and 1.37 L/kg, respectively (plasma concentration-time profile is not shown). Based on these parameters, cimetidine was administered intravenously by constant infusion at 120 lg/min/kg after bolus intravenous administration at 3.4 mg/kg to achieve a total plasma concentration of 10 lM (2520 ng/mL, cimetidine mw: 252.34) (n ¼ 1). Cimetidine was dissolved in saline containing 4% mannitol. Blood was collected using a heparinised syringe at 5, 10, 20, 30, and 40 min after the beginning of constant infusion and plasma was obtained from the blood by centrifugation. C]mirogabalin in rats treated with and without cimetidine was conducted. A flowchart of the time points of treatment with cimetidine and test compounds and of the collection of urine and blood is shown in Figure 2. The femoral vein and the urinary bladder were catheterised in rats under anaesthesia using pentobarbital sodium salt (35 mg/kg, i.p.) and buprenorphine hydrochloride (0.1 mg/kg, s.c.). Cimetidine dissolved in saline containing 4% mannitol or in saline containing 4% mannitol was infused via the femoral vein at a constant rate of 1.2 mL/h. All four test compounds diluted with saline containing 4% mannitol were infused via the femoral vein after catheterisation of the urinary bladder to achieve a target plasma concentration of radioactivity, which was 500 dpm/0.02 mL for [ 14 C]inulin (3.8 mg-equivalent/mL), 1000 dpm/0.05 mL for [ 3 H]PAH (1.6 mg-equivalent/mL), 500 dpm/0.05 mL for [ 14 C]metformin (7.4 ng-equivalent/mL) and 500 dpm/0.05 mL for [ 14 C]mirogabalin (19 ng-equivalent/mL). The dosing solutions were infused at a constant rate of 1.2 mL/h. To determine the constant infusion rate of the radioactivity of each test compound, CLt and V0 of inulin were calculated using values of the glomerular filtration rate (GFR) and the plasma volume of rats (Lin et al. 1987;Davies and Morris 1993). Those of PAH were calculated using values of the renal plasma flow (RPF) and plasma volume of rats (Davies and Morris 1993;Suzuki and Hierlihy 1993). Those of metformin and mirogabalin were calculated using values in the literature (Choi et al. 2006;Yamamura et al. 2022a). After initiating constant infusion of the radiolabelled compound, urine was collected at 30-40 min, 40-50 min, and 50-60 min, and blood was also collected at 35, 45, and 55 min. At the end of 60 min, portions of the liver, renal cortex of the right kidney, upon catheterisation of the urinary bladder, and left whole kidney were collected immediately after cessation of the constant infusion and the animals were killed by abdominal aortic exsanguination.

Pharmacokinetic analysis
Inulin and PAH are known not to be metabolised, as is the case for metformin as described in its package insert, while mirogabalin has been confirmed to undergo minimal metabolism in rats based on the results of quantitative evaluation of metabolite profiles in plasma, urine, and faeces. After oral administration of [ 14 C]mirogabalin to rats, urinary recovery of radioactivity was 94.5% and no metabolite in urine was found. Only mirogabalin was found in the urine. (Yamamura et al. 2022a). Therefore, pharmacokinetic analysis of [ 14 C]inulin, [ 3 H]PAH, [ 14 C]metformin and [ 14 C]mirogabalin was conducted using radioactivity in plasma, urine, and tissues. CLt was determined by dividing the infusion constant rate of radioactivity (dpm/min) by the plasma concentration of radioactivity (dpm/mL). Meanwhile, CLr was determined by dividing the renal excretion rate of radioactivity (dpm/min) by the plasma concentration of radioactivity. In order to clarify the inhibition site of cimetidine, CLr based on the kidney concentration (CLr/C kidney ), instead of intracellular concentration in renal proximal tubular cells, was calculated. If cimetidine inhibited apical efflux transporter (rMate1) of the cells, kidney concentration of [ 14 C]metformin and [ 14 C]mirogabalin would increase and renal secretion of them would decrease. CLr/C kidney was determined by dividing the renal excretion rate of radioactivity (dpm/min) by the kidney concentration of radioactivity (C kidney : dpm/g). The apparent liver-to-plasma concentration ratio (Kp, liver), renal cortex-to-plasma concentration ratio (Kp, cortex), and kidney-to-plasma concentration ratio (Kp, kidney) were determined by dividing liver, renal cortex, and kidney concentrations of radioactivity (dpm/g) by the plasma concentration of radioactivity (dpm/mL), respectively. Finally, the plasma concentration of radioactivity and the renal excretion rate of radioactivity were determined using plasma and urine collected in the last collection period.

Statistical analysis
Data are presented as the mean ± standard deviation. Student's two-tailed unpaired t-test was used to identify significant differences between the control group and the cimetidine-treated group. p < 0.05 was considered to be statistically significant.

Determination of plasma unbound fraction of mirogabalin
Unbound fractions of mirogabalin at 0.1, 1, and 10 mg/mL in plasma prepared from the blood of male F344 rats were 0.758 ± 0.004, 0.766 ± 0.010, and 0.786 ± 0.004, respectively. Since the observed plasma concentration of [ 14 C]mirogabalin was around 26 ng-equivalent/mL in control rats and 52 ngequivalent/mL in cimetidine-treated rats, 0.758 was used as the plasma unbound fraction (fp) of mirogabalin.

Plasma concentration-time course of cimetidine
Plasma concentration-time profile of cimetidine in a rat is shown in Figure 3. Plasma cimetidine reached around 2500 ng/mL (9.9 lM), which was almost the target concentration of 2520 ng/mL (10 lM), 20 min after initiating constant infusion, and was maintained at this concentration. Since the unbound fraction of cimetidine in rat plasma was reported to be 0.71 ± 0.04 (Ikemura et al. 2013), the plasma-free concentration of cimetidine in this condition was thought to be 7.0 lM (9.9 lM Â 0.71).

Renal clearance studies of [ 14 C]inulin, [ 3 H]PAH, [ 14 C]metformin and [ 14 C]mirogabalin in rats treated with or without cimetidine
To clarify whether the constant infusion of a radiolabelled test compound makes similar plasma concentration and urinary excretion rate at three collection periods, the plasma concentrations and the urinary excretion rates of [ 14 C]mirogablin in rats are shown in Figure 4. The plasma concentrations of [ 14 C]mirogablin at 5, 15, and 25 min ( Figure  4(A)) and the urinary excretion rates from 0 to 10 min, 10 to 20 min, and 20 to 30 min (Figure 4(B)) after initiating urine collection were almost the same. The mean plasma concentrations at 15 and 25 min and the mean urinary excretion rates from 10 to 20 min and 20 to 30 min for [ 3 H]PAH and [ 14 C]metformin were also the same, as shown in Figure 5. Figures 4 and 5 indicate that the constant infusion of each radiolabelled test compound resulted in a steady-state condition in the last period of collection of blood and urine.
[ 3 H]PAH and [ 14 C]inulin were also used for determining RPF and GFR in rats (Hirschberg and Kopple 1989). Renal clearance of inulin and PAH is believed to express GFR and RPF, respectively (Melzer 2013). Inulin consists of chain-terminating glucosyl moieties and a repetitive fructosyl moiety, which are linked by b(2,1) bonds. The degree of polymerisation of standard inulin ranges from 2 to 60 (average 35). Its average molecular weight is 5.5 KDa. Inulin is thought to be free completely in plasma and to disappear from plasma via renal filtration only, indicating renal clearance of inulin equals plasma clearance of inulin and they means GFR. On the other hand, PAH is bound by plasma proteins and the protein binding ratio is known to be variety depending on species. It was reported that the unbound fraction of PAH in rat plasma was 0.92 and it in monkey plasma was 0.27 (Prueksaritanont et al. 2004). Thus, in the case of rats, renal clearance of PAH expresses almost RPF. Table  1 shows the PK parameters of each radiolabelled test compound. The PK parameters of [ 14 C]inulin and [ 3 H]PAH in control rats did not differ from those in cimetidine-treated rats, indicating that cimetidine at 7.0 lM as the unbound plasma concentration did not affect the pharmacokinetics of inulin and PAH in rats. The GFR, CLr of [ 14 C]inulin, was 10.2 ± 2.2 mL/min/kg in control rats and 9.19 ± 1.17 mL/min/kg in cimetidine-treated rats. The Kp, liver of inulin, which is known not to be distributed in cells, was 0.382 ± 0.097 in control rats and 0.605 ± 0.362 in cimetidine-treated rats. The Kp, kidney of inulin, which is known to be only filtrated into the urine, was 3.63 ± 0.67 in control rats and 3.51 ± 0.51 in cimetidine-treated rats. The RPF, CLr of [ 3 H]PAH, was calculated to be 13.2 ± 1.7 mL/min/kg in control rats and 13.2 ± 1.1 mL/min/kg in cimetidine-treated rats. The Kp, kidney values of [ 3 H]PAH in control rats (4.56 ± 0.09) and cimetidine-treated rats (4.77 ± 0.30) were greater than those of inulin in control and cimetidine-treated rats, suggesting renal secretion. Cimetidine significantly decreased CLr and CLr/C kidney and increased Kp, kidney, and Kp, cortex of [ 14 C]metformin. However, CLt and Kp, liver of [ 14 C]metformin were not changed  by cimetidine. The CLr of [ 14 C]metformin in control rats, 18.7 ± 1.5 mL/min/kg, was decreased to 12.8 ± 1.5 mL/min/kg by cimetidine treatment, a value that was similar to the renal filtration clearance of metformin calculated as 9.3 mL/min/kg (fp Â GFR ¼ 0.916 Â 10.2 mL/min/kg, 0.916 was calculated from the reported protein binding ratio of metformin in control rat plasma: 8.4%; Jin et al. 2009). This suggests that the renal secretion of metformin in rats was almost inhibited by cimetidine treatment. CLr (8.32 ± 1.06 mL/min/kg) of [ 14 C]mirogabalin in control rats were similar to renal filtration clearance (fp Â GFR: 0.758 Â 10.2 mL/min/kg ¼ 7.73 mL/min/kg), indicating little renal secretion of mirogabalin in rats. However, surprisingly, cimetidine treatment significantly decreased CL, CLr, and CLr/C kidney and increased Kp, cortex, and Kp, kidney of [ 14 C]mirogabalin. The Kp, liver of [ 14 C]mirogabalin was not affected by cimetidine. With cimetidine treatment the CLr of [ 14 C]mirogabalin (8.32 ± 1.06 mL/min/kg) was decreased to 4.12 ± 0.44 mL/min/kg, which was lower than the renal filtration clearance of [ 14 C]mirogabalin, 7.73 mL/min/kg, indicating reabsorption of [ 14 C]mirogabalin in the kidney of rats.

Discussion
In the present study, the intrarenal (kidney and cortex) concentration of [ 14 C]mirogabalin reflected by radioactivity was measured in rats in a steady-state condition of [ 14 C]mirogabalin and cimetidine by constant infusion, in order to clarify whether cimetidine inhibits the apical efflux transporter rMate1 or the basolateral uptake transporter rOct2. This steady-state infusion study was conducted in rats under anaesthesia. To assess the effects of anaesthesia on renal function, GFR, and RPF were measured using [ 14 C]inulin and [ 3 H]PAH. CLr of [ 14 C]inulin in this study was 10.2 ± 2.2 mL/min/kg in control rats and 9.19 ± 1.17 mL/min/kg in cimetidine-treated rats. These values were consistent with reported values of GFR (8.4-12.9 mL/min/kg, Lin et al. 1987;8.10 ± 2.06 mL/min/kg, Suzuki and Hierlihy 1993). CLr of [ 3 H]PAH in this study was 13.2 ± 1.7 mL/min/kg in control rats and 13.2 ± 1.1 mL/min/kg in cimetidine-treated rats, which are smaller than reported renal clearance of [ 3 H]PAH (2.0 mL/min/100g BW (¼ 20 mL/min/kg), Wabner and Chen 1987; 2.1 mL/min/100g BW (¼ 21 mL/min/kg), Winston and Safirstein 1985; 7.9 ± 2.0 mL/min/290-320 g rats (¼ 25.9 mL/min/kg), Prueksaritanont et al. 2004;35.71 ± 5.45 mL/min/kg, Suzuki and Hierlihy 1993). All of the renal filtration function was maintained while part of the renal secretion function was maintained in rats under these experimental conditions. PAH is a typical substrate of the rat Oat family, which consists of three members in rats (Oat1-3), and of rat Mrp2 (Van Aubel et al. 2000). Cimetidine did not affect the renal behaviour of [ 3 H]PAH in this experiment, confirming that cimetidine at the unbound plasma concentration of 7.0 lM did not inhibit related uptake transporters nor efflux transporters of organic anions in the kidney of rats.
Metformin is a known substrate of rOct2 rather than rOct1 (Kimura et al. 2005). The mRNA expression level of rOct2 in the kidney was higher than that in the liver and the mRNA expression of rOct1 was not detected in the kidney (Kimura et al. 2005). The Ki values of cimetidine were reported to be 632 lM and 9.4 lM for the uptake of [ 14 C]tetraethylammonium by rOct2-overexpressing HEK293 cells (Umehara et al. 2007) and by rOct2-overexpressing MDCK cells (Urakami et al. 1998), respectively. As the Ki value of cimetidine of 9.4 lM for rOct2 was comparable to or somewhat higher than the observed plasma unbound concentration (7.0 lM), it cannot be ruled out that cimetidine inhibited rOct2 to some extent in rats in vivo. However, this effect is considered to be negligible or limited based on our in vivo observation in rats. Specifically, cimetidine coadministration significantly increased the cellular accumulation of metformin in the renal cortex and kidney, suggesting that the effect of inhibition of the apical efflux transporter rMate1, rather than the basolateral uptake transporter rOct2, is more dominant in vivo. In the same manner, cimetidine coadministration resulted in the intracellular accumulation of mirogabalin in kidney cortex, accompanied by a decrease in CLr/C kidney .
In a clinical setting, we observed significant decreases of CL/F and CLr of mirogabalin by the coadministration of cimetidine, where the maximum unbound plasma concentration was lower than the Ki value of cimetidine for hOCT2 in a clinical drug interaction study (Tachibana et al. 2018).   Given that mirogabalin is a substrate of hOAT1, OAT3, OCT2, MATE1 and/or MATE2-K, and PEPT1 and/or PEPT2 (Yamamura et al. 2022b), it is suggested that the clinical drug interaction between mirogabalin and cimetidine is due to the inhibition of apical efflux transporters, hMATE1 and/or hMATE2-K, by cimetidine rather than hOCT2, as supported by in vivo observation in rats in this report. We were surprised by the finding that mirogabalin was reabsorbed from the urinary side into apical tubular cells in the kidney of rats. Since cimetidine did not decrease the CLr of mirogabalin below GFR in the clinical drug interaction study, there is no direct evidence of the reabsorption of mirogabalin in the kidney of humans. Arakawa et al. (2019) reported the activity of peptide transporters 1 (Pept1) and 2 (Pept2) evaluated by measuring the uptake of [ 3 H]glycylsarcosine into rat kidney slices. The reabsorption of mirogabalin in the kidney of rats may involve rat Pept1 and Pept2.

Conclusion
In this study, we found that the renal clearance of mirogabalin in rats was inhibited via the apical efflux transporters rMate1, not the basolateral uptake transporter rOct2, by coadministration of cimetidine. Reabsorption of mirogabalin in the kidney was also observed in rats.