Abstract
Excretion of [14C]HR011303-derived radioactivity showed significant species difference. Urine (81.50% of dose) was the main excretion route in healthy male subjects, whereas feces (87.16% of dose) was the main excretion route in rats. To further elucidate the underlying cause for excretion species differences of HR011303, studies were conducted to uncover its metabolism and excretion mechanism. M5, a glucuronide metabolite of HR011303, is the main metabolite in humans and rats. Results of a rat microsome incubation study suggested that HR011303 was metabolized to M5 in the rat liver. According to previous studies, M5 is produced in both human liver and kidney microsomes. We found that M5 in the human liver can be transported to the blood by multidrug resistance-associated protein (MRP) 3, and then the majority of M5 can be hydrolyzed to HR011303. HR011303 enters the human kidney or liver through passive diffusion, whereas M5 is taken up through organic anion transporter (OAT) 3, organic anion-transporting polypeptide (OATP) 1B1, and OATP1B3. When HR011303 alone is present, it can be metabolized to M5 in both sandwich-cultured rat hepatocytes (SCRH) and sandwich-cultured human hepatocytes (SCHH) and excreted into bile as M5 in SCRH. Using transporter inhibitors in sandwich-cultured model and membrane vesicles expressing MRP2 or Mrp2, we found that M5 was a substance of MRP2/Mrp2, and the bile efflux of M5 was mainly mediated by MRP2/Mrp2. Considering the significant role of MRP3/Mrp3 and MRP2/Mrp2 in the excretion of glucuronides, the competition between them for M5 was possibly the determinant for the different excretion routes in humans and rats.
SIGNIFICANCE STATEMENT Animal experiments are necessary to predict dosage and safety of candidate drugs prior to clinical trials. However, extrapolation results often differ from the actual situation. For HR011303, excretory pathways exhibited a complete reversal, through urine in humans and feces in rats. Such phenomena have been observed in several drugs, but no in-depth studies have been conducted to date. In the present study, the excretion species differences of HR011303 can be explained by the competition for M5 between MRP2/Mrp2 and MRP3/Mrp3.
Introduction
Animal experiments are needed prior to the clinical trials of candidate drugs to predict dosage, safety, and toxicity. Rats are the most commonly used animal models (Takano et al., 2010). However, due to obvious species differences in the types and expression levels of metabolizing enzymes (Shiratani et al., 2008; Stringer et al., 2009; Mazur et al., 2010) and transporters (Hillgren et al., 2013; Wang et al., 2015), the extrapolated results often differ from the actual results (Lin et al., 1996). Direct extrapolation of biliary clearance from preclinical small animals sometimes leads to an overestimation of human biliary clearance. For example, an overestimation by 400- and 150-fold was observed for cefamandole and vincristine, respectively (Ratzan et al., 1978; Ansede et al., 2010; Grime and Paine, 2013).
Transporters play an important role in the distribution and elimination of drugs (Tu et al., 2013; Ma et al., 2014; Nigam and Sanjay, 2015). MRP3 is located at the basolateral membrane and mediates the efflux of substrates, particularly glucuronides and glutathione conjugates, into sinusoidal blood. By contrast, organic anion-transporting polypeptide (OATPs) are the major drug uptake transporters in hepatocytes (Hillgren et al., 2013). Breast cancer resistance protein (BCRP), P-glycoprotein, and multidrug resistance-associated protein (MRP) 2 are localized to the canalicular membrane of hepatocytes and mediate the export of xenobiotics into bile (Robertson and Rankin, 2006; König et al., 2013). MRP2/Mrp2 and MRP3/Mrp3 possibly play a decisive role in glucuronide excretion. Mrp2 and Mrp3 provide alternative routes for the hepatic excretion of morphine-glucuronides (Van De Wetering et al., 2007). Mrp2 plays an important role in doxorubicin elimination. The levels of Mrp3 (Abcc3) in the livers of Mrp2−/− mice increase by approximately 2-fold (Nezasa et al., 2006; Vlaming et al., 2006). The upregulation of basolateral MRP3/Mrp3 compensates for the reduced ability of MRP2/Mrp2 (Chandra et al., 2005).
HR011303 is a highly selective urate transporter 1 inhibitor, which is clinically intended to treat gout. It is currently in phase III clinical trials in China (Peng et al., 2016; Wang et al., 2019). When comparing mass balance results in humans and rats, significantly different excretion routes were observed in these two species. Early animal studies showed that after intragastric administration of [14C]HR011303 in Sprague-Dawley (SD) rats, drug-related substances mainly effluxed into bile and excreted in the feces. The recovery rate of radioactivity was 87.20% in the feces and 10.10% in the urine within 0–168 hours (Fig. 1A). The recovery rate was 70.90% in the bile within 0–48 hours (Fig. 1B). In contrast, in healthy male Chinese subjects, urine was the main excretion route after oral administration of [14C]HR011303. Within 0–216 hours, 81.50% of the recovered drug-related substances were excreted in the urine, and 10.26% were excreted in the feces (Fig. 1C). The main metabolite in human urine and rat bile was the glucuronide metabolite M5 (Fig. 2). The in vitro microsome incubation results showed that HR011303 could be metabolized to M5 in human liver microsomes and human kidney microsomes (companion manuscript, DMD-AR-2021-000581). In this study, we explored the underlying reasons for the species differences in the excretion of HR011303 from the perspective of transporters. We aimed to reveal the mechanisms underlying such excretion differences. Results will serve as evidence for similar situations that may arise in the development of new drugs.
Cumulative radioactivity recovery of rats and humans. (A) cumulative recovery in rat urine and feces within 0–168 hours. (B) cumulative recovery in rat bile within 48 hours. (C) cumulative recovery in human urine and feces. Data were expressed as mean ± S.D. (n = 6).
Structure and abundance of major metabolites of HR011303 in rats (A) and humans (B). Metabolites that were less than 4% of the dose were not presented. *multiple isomers.
The huge differences in HR011303 excretion between humans and rats inspired a series of experiments that aimed to uncover the HR011303's disposition and excretion process in rats and humans. Experiments were performed to achieve the following objectives: 1) to identify the main metabolic sites of HR011303 in rats; 2) to characterize the role of MRP3 in the distribution process of HR011303 and M5; 3) to identify the transporters involved in HR011303 and M5 kidney and hepatic uptake; and 4) to investigate the HR011303 and M5 bile efflux in rat and human hepatocytes.
Materials and Methods
Chemicals and Reagents
HR011303 (98.40% purity), M5 (95.40% purity), and internal standard (IS) SHR144764 (98.20% purity) were kindly supplied by Jiangsu Hengrui Medicine Co., Ltd. (Lianyungang, China). High-performance liquid chromatography (HPLC)-grade acetonitrile and methanol were purchased from Sigma (St. Louis, MO, USA). DMSO was obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Formic acid was purchased from Rhawn Chemical Reagent Co., Ltd. (Shanghai, China). Sodium carboxymethyl cellulose was provided by Aladdin (Shanghai, China). The following were purchased from Meilunbio (Dalian, China): uridine 5′-diphosphoglucuronic acid (UDPGA), protease inhibitor cocktail, alamethicin, 3-(N-morpholino) propanesulfonic acid, adenosine 5′-triphosphate disodium salt, adenosine 5′-monophosphate disodium salt, magnesium chloride, Hanks’ buffered salt solution (HBSS), and D-Hanks. Other commercially available reagents used in the experiments were of analytical grade.
Pooled male SD rat liver microsomes (RLMs) were supplied by Corning Gentest (Woburn, MA, USA). Pooled male SD rat kidney microsomes (RKMs) were purchased from XenoTech, Inc. (Lenexa, KS, USA). Inside-out Sf9 insect cell membrane vesicles expressing MRP2, Mrp2, and MRP3 were provided by GenoMembrane Co., Ltd. (Yokohama, Japan). Blank human plasma was kindly supplied by the People's Hospital of Weifang High-tech Industrial Development Zone (Weifang, China). Cryopreserved human hepatocytes were purchased from IVT (Baltimore, MD, USA). OATP1B1-, OATP1B3-, OAT1-, OAT3-, or organic cation transporter 2-expressing human embryonic kidney 293 (HEK293) cell lines and empty-vector-transfected control cells were constructed at HD Biosciences Co., Ltd. (Shanghai, China).
Sprague-Dawley rats were obtained from Jiesijie Laboratory Animal Co., Ltd (Shanghai, China). The study was approved by the Animal Ethics Committee of Shanghai Institute of Materia Medica, Chinese Academy of Sciences. Rats were housed and handled according to the Guidelines for Care and Use of Laboratory Animals at Shanghai Institute of Materia Medica, Chinese Academy of Sciences.
Rat Mass Balance Study
[14C]HR011303 and HR011303 sodium salts were dissolved in 0.5% Sodium carboxymethyl cellulose for mass balance study in rats, and the concentration for oral administration was 8.95 mg/kg (81.9 µCi/kg). The following samples were collected: pre-dose urine and fecal samples; urine samples at 0–8, 8–24, 24–48, 48–72, 72–96, 96–120, 120–144, and 144–168 hour post-dose; and feces samples at 0–24, 24–48, 48–72, 72–96, 96–120, 120–144, and 144–168 hour post-dose (3 male and 3 female rats, respectively). Bile samples were collected pre-dose and at 0–4, 4–8, 8–24, and 24–48 hours after administration (3 male and 3 female rats).
For urine and bile samples, a liquid scintillation counter (LSC) (Tri-Carb 3110TR, PerkinElmer, Waltham, MA, USA) was used to measure the radioactivity after adding an appropriate amount of urine or bile to 5 ml of scintillation fluid, while feces were mixed with 3 times weight of acetonitrile–water (1:1, v/v) and then homogenized. About 300 mg of homogenate was combusted using an OX-501 Biologic Oxidizer (Harvey, Tappan, NY, USA), and the generated 14CO2 was captured by alkaline RDC liquid scintillation cocktail. Finally, the radioactivity was measured with a Tri-Carb 3110TR liquid scintillation counter.
In Vitro Incubations of HR011303 with Pooled RLMs and RKMs
HR011303 was dissolved in methanol-water (50:50, v/v) to prepare a 2 mM stock solution. Alamethicin was dissolved in DMSO to prepare a stock solution with a concentration of 40 mg/mL, which was then diluted to obtain a working solution with 50 mM of Tris-HCl buffer (8 mM MgCl2, pH 7.5). After incubation at 37°C for 3 minutes, UDPGA was added to initiate the reaction. The final incubation system was 200 μL and included 50 mM of Tris-HCl buffer, 8 mM of magnesium chloride, 25 μg/mL of alamethicin, 5 μM of HR011303, RLMs or RKMs (1 mg protein/mL), and 2 mM UDPGA. After incubation at 37°C for 1 hour, the reaction was terminated by adding 200 μLof cold acetonitrile. An incubation system without microsomes was used as the negative control.
Efflux Transporter MRP3 Study
The efflux of HR011303 and M5 was evaluated in human MRP3-expressing membrane vesicles. In transport assays, a modified rapid filtration technique was used based on the manufacturer’s protocol (GenoMembrane Co). Membrane vesicles (50 μg of protein) and test compounds were incubated with or without ATP (5 mM) in the transport medium (50 μL, pH 7.4), which contained 50 mM of 3-(N-morpholino) propanesulfonic acid-Tris and 10 mM of MgCl2. The process was terminated after 10 minutes of incubation by adding 200 μL of ice-cold quenching solution (40 mM of 3-(N-morpholino) propanesulfonic acid-Tris and 70 mM of KCl). The incubation mixture was quickly transferred to a Millipore 96-well glass fiber filter plate (Millipore, MA, USA) and washed twice with ice-cold quenching solution. The compound trapped in the membrane vesicles was retained on the filters and released by adding 200 μL of ethanol-water (50:50, v/v). The compounds inside the vesicles were analyzed by liquid chromatography/tandem mass spectrometry (LC-MS/MS). ATP-dependent transport was evaluated according to the ratio of transport with ATP to that without ATP (Gao et al., 2012).
Transporter-Transfected Cell Culture
Transporter-transfected cells were cultured as previously reported (Takeuchi et al., 2016). HEK293 cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 2 mM of L-glutamine, 1% modified Eagle’s medium nonessential amino acid, 100 μg/mL of penicillin, and 100 μg/mL of streptomycin. Cultures were maintained in a humidified atmosphere containing 5% CO2 at 37°C.
Uptake Transporters Study
Uptake of HR011303 and M5 was evaluated using HEK-OATP1B1, HEK-OATP1B3, HEK-OAT1, HEK-OAT3 and HEK-mock cell lines according to a previously reported method (Han et al., 2010). Transfected HEK293 and mock control cells were seeded on BioCoat poly-D-lysine-coated 24-well plates (BD Biosciences) at a density of 3.0 × 105 cells/well. After 36 hours of culture, the cells were washed twice and equilibrated in prewarmed Hank’s balanced salt solution (HBSS) with or without inhibitor for 10 minutes. The uptake was initiated by adding 0.5 mL of HBSS containing test compounds or a mixed solution of the test drug and transporter inhibitor to the corresponding cell wells. Then, the uptake was terminated at a designated time by discarding the medium and washing the cells twice with ice-cold HBSS. Cells were lysed with methanol-water (70:30, v/v). The test compound concentrations were determined by LC-MS/MS. Total cellular protein levels were measured using a BCA protein assay kit. The main kinetic parameters were calculated using Prism 8 (GraphPad Software, Inc., San Diego, CA).
Stability of M5 in Human Plasma
Stock solution (10 mM) was prepared by dissolving an appropriate amount of M5 in DMSO and was diluted to 50 μM with PBS. The incubation mixtures (100 µL of total volume) consisted of 90 μL of human plasma (5% volume of cocktail inhibitor or DMSO) and 10 μL of the M5 solution. The final incubation concentration of M5 was 5 μM. The reactions were maintained at 37°C for 0, 5, 10, 20, and 40 minutes and terminated with 1000 μL of ice-cold acetonitrile (0.1% FA, IS:100 ng/mL). Each incubation was conducted in duplicate. M5 concentration was quantified by using LC-MS/MS. The results are expressed as the percentage of the remaining M5 (Jiang et al., 2016).
Preparation and Sandwich-Culture of Human and Rat Primary Hepatocytes
Cryopreserved human hepatocytes were thawed according to the manufacturer’s instructions (Baltimore, MD, USA). Rat hepatocytes were isolated from male SD rats using a previously reported two-step recirculating collagenase perfusion (Xue et al., 2020). Cells were suspended at approximately 7 × 105 cells/mL in a plating medium and then added at approximately 0.5 mL per well to 24-well plates pretreated with rat tail collagen. After 4 hours of incubation at 37°C and 5% carbon dioxide, the cells were overlaid with 0.25 mg/mL of Type I Matrigel solution prepared in a feeding medium. The medium was replaced every 24 hours. Studies were conducted on day 5 for sandwich-cultured rat hepatocytes (SCRH) and day 6 for sandwich-cultured human hepatocytes (SCHH).
Hepatobiliary Efflux of HR011303
SCRHs were used to investigate the bile efflux of HR011303 in rats. Different transporter inhibitors were used to investigate the contribution of different transporters to the efflux process (Shen et al., 2012). Hepatocytes were washed thrice with 400μL of either HBSS containing calcium (standard buffer) or HBSS without calcium (calcium-free buffer) and incubated with the third wash solution either in the presence or absence of transporter inhibitors (verapamil for MDR1 (P-glycoprotein)/Mdr1, novobiocin for BCRP/Bcrp, MK571 for MRP2/Mrp2) for 10 minutes at 37°C. The wash solution was removed. Then, the cells were incubated with the test drug solution or a mixed solution of the test drug and transporter inhibitor according to the experimental design at 37°C for 15 minutes. The medium was discarded after the incubation. Cells were rinsed with ice-cold HBSS thrice, and the cell samples were stored at -80°C for analysis. After the repeated freezing and thawing of the cells, the total amount of HR011303 or M5 in hepatocytes plus bile (standard HBSS) and hepatocytes (Ca2+-free HBSS) was determined by LC-MS/MS. Total cellular protein levels were measured using a BCA protein assay kit.
The bile excretion index (BEI) represents the fraction of the analyte excreted into the bile pockets (Chandra and Brouwer, 2004). A BEI value greater than 10% indicated that the compound was considered to be excreted through the bile (Pan et al., 2012). The BEI value was calculated using the following equation:
where Acells+bile and Acells are the accumulation of compounds in sandwich-cultured hepatocytes measured in the presence and absence of Ca2+.
Efflux Transporter MRP2/Mrp2 Study
To further evaluate the involvement of MRP2 or Mrp2 in the efflux of HR011303 and M5, human MRP2 and rat Mrp2-expressing membrane vesicles were used. The experimental operation was the same as the “Efflux transporter MRP3 study.”
Analytical Conditions
The analysis method used for microsome incubation was the same as that used in Zheng’s manuscript (companion manuscript, DMD-AR-2021-000581). For LC-MS/MS quantitative analysis, separation of analytes from the matrix was achieved via an Agilent 1290 ultra performance liquid chromatography (UHPLC) system (Agilent, Santa Clara, CA, USA) on an HSS T3 column with temperature maintained at 40°C. MS detection was conducted by an Agilent 6495 triple-quadrupole mass spectrometer (Agilent). Multiple reaction monitoring fragmentation transitions of m/z 338.1→240.0 (HR011303), m/z 514.1→338.2 (M5), and m/z 328.1→230.0 (IS) were monitored in positive electrospray ionization mode (Wang et al., 2019). The mobile phase consisted of a mixture of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B). The flow rate was set to 0.3 mL/min. Gradient elution was initiated at 35% B, maintained for 0.5 minutes, and increased to 95% B linearly in 0.9 minutes, and maintained for 1.1 minutes. The elution was finally decreased to 35% B to re-equilibrate the column for 0.9 minutes. The mass spectrometer parameters were as follows: capillary voltage, +3.0 kV; nozzle voltage, +1.5 kV; nebulizer gas pressure, 30 psi; carrier gas, 14 mL/min and 250°C; and sheath gas, 11 mL/min and 300°C.
Results
Mass Balance in Rats
Fecal excretion was the predominant route of elimination constituting 87.20% of the administered dose. In contrast, urine excretion accounted for 10.10% of the administered dose, as shown in Fig. 1A. Radioactivity recovery in bile was 70.90% (Fig. 1B).
Metabolism of HR011303 in Pooled RLMs and RKMs
To study the metabolic sites of HR011303 in rats, incubation was performed with RLMs and RKMs in the presence of UDPGA. The target peak was determined by comparing the retention time and the major product ions with the standard. Fig. 3 shows that unlike in human microsomes, M5 was formed only in RLMs. M5 formation was not observed in the RKMs.
Relative amounts of HR011303 and M5 after HR011303 was incubated with human (A) and rat microsomes (B) (n = 2).
Efflux Transporter MRP3 Study
M5 is the predominant metabolite in the human liver. Thus, MRP3-expressing membrane vesicles were used to determine the interaction between transporters MRP3 and HR011303 as well as M5. As shown in Fig. 4, the ratios of ATP-dependent uptake to nonspecific adsorption (+ATP/-ATP ratio) of M5 were 9.96 (1 μM) and 6.33 (10 μM), whereas the ratios of HR011303 were 0.84 (0.5 μM) and 0.91 (2 μM). No significant ATP-dependent transport was observed in HR011303. M5 rather than HR011303 was a substrate of MRP3, which indicated that M5 produced in human hepatocytes was transported into the blood by MRP3.
Direct transport of HR011303 and M5 by multidrug resistance protein 3. HR011303 or M5 was incubated with MRP3-expressing membrane vesicle with or without the presence of ATP for 10 minutes. Data were expressed as mean + S.D. (n = 3).
Uptake Transporter Study
HEK293 cells expressing OATP1B1, OATP1B3, OAT1, OAT3, and organic cation transporter 2 were used to identify transporters that mediated the uptake of HR011303 and M5. In mock cells, the uptake rates of HR011303 were 27.80 and 2.47 pmol/min/mg protein (10 and 2 μM, respectively). The difference in uptake rates between transfected and mock cells of HR011303 was not significant, and transporter inhibitors had a negligible influence on HR011303 uptake. HR011303 had high permeability and entered the liver and kidney through passive diffusion instead of transporter mediation. Significant active uptake of M5 was observed for OAT3-, OATP1B3-, and OATP1B1-transfected cells, and the ratios of their uptake rates to mock cells were 46.46, 18.98, and 14.21, respectively (Fig. 5). After adding the inhibitors of these transporters, the uptake rates of M5 decreased significantly. Unlike these transporters, only slight or negligible active uptake was observed in OAT1- and organic cation transporter 2-transfected cells compared with mock cells.
Uptake of M5 in the organic anion-transporting polypeptide (OATP) 1-, OATP1B3-, OATP2B1-, OAT1-, or OAT3-transfected human embryonic kidney 293 cells. Incubation concentration was 5 μM, and the uptake time was 10 minutes for all cells. Data were expressed as mean + S.D. (n = 3).
The time-dependent uptake of M5 by OAT3, OATP1B3, and OATP1B1 was measured at a substrate concentration of 5 μM. Uptake was linear with time in the first 2 minutes of incubation (Fig. 6). OAT3-, OATP1B3-, and OATP1B1-mediated uptakes of M5 were further determined at various substrate concentrations after 2 minutes of incubation. For OATP1B1, OATP1B3, and OAT3, the uptake of M5 increased with the substrate concentration ranging from 1 μM to 50 μM and reached a plateau at approximately 50 μM for OATP1B3 and 100 μM for OATP1B1 and OAT3 (Fig. 6). The kinetic parameters are presented in Table 1. Only the OATP1B3-mediated uptake of M5 showed apparent substrate inhibition after reaching 50 μM (Ki=252.7 μM).
The time dependence of M5 in human embryonic kidney-mock and human embryonic kidney–organic anion-transporting polypeptide (OATP) 1B1 (A), OATP1B3 (B), and OAT3 (C) at an incubation concentration of 5 μM. The active uptake kinetic curves of M5 by OATP1B1 (D), OATP1B3 (E), and OAT3 (F) at an incubation time of 2 minutes. Data were expressed as mean ± S.D. (n = 3).
Main kinetic parameters for the transportation of M5 in human embryonic kidney–organic anion-transporting polypeptide 1B1, 1B3, and organic anion transporter 3 cell lines
Stability of M5 in Human Plasma
HR011303 accounted for 87.9% of the total exposure of drug-related substances, whereas M5 was less than 1% in human plasma samples that mixed according to the AUC pooling principle (Hop et al., 1998). This finding contradicted our result that M5 in the liver was transported to the blood via MRP3. To explain this phenomenon, stability of M5 in human plasma at physiologic body temperature was tested, and the results are shown in Fig. 7. After incubation at 37°C for 5 minutes, the M5 concentration dropped to less than 40% of the initial incubation concentration. After 20 minutes, less than 10% of M5 remained in human plasma. The protease inhibitor cocktail significantly improved M5 stability.
Relative remaining of M5 (initial concentration: 5 μM) in human plasma after incubation for up to 40 minutes (n = 2).
Hepatobiliary Efflux of HR011303
SCRHs were used to study the bile efflux process of drug-related substances of HR011303 after administration. When SCRHs were treated with HR011303 alone, M5, the glucuronide metabolite of HR011303, was the major compound excreted into bile. The BEI value of M5 was 23.44% (Fig. 8). With the addition of MK571, an inhibitor of MRP2/Mrp2, the BEI value decreased to 12.25% (Fig. 8). However, novobiocin (an inhibitor of BCRP) and verapamil (an inhibitor of P-glycoprotein) had negligible effects on M5 bile efflux. M5 in rat hepatocytes became an effluent in the bile via MRP2. Based on the rat hepatobiliary efflux results, SCHHs were used to explore bile efflux in humans, and MK571 was used to evaluate the role of MRP2 in the efflux process. Similar to the results in SCRH, both HR011303 and M5 were detected in cells treated with HR011303 alone. However, the BEI values of HR011303 and M5 in SCHH were 7.49% and 4.11%, respectively, and these values were both less than 10%. MK571 inhibited the entry of M5 into bile (Fig. 8). The means of HR011303 and M5 excretion showed significant inter-species differences. HR011303 and M5 were not excreted into bile in humans, whereas they were transported to bile via Mrp2 in rats.
Accumulation of HR011303 in sandwich-cultured rat hepatocytes (A) and sandwich-cultured human hepatocytes (C) as well as the accumulation of M5 in sandwich-cultured rat hepatocytes (B) and sandwich-cultured human hepatocytes (D) with or without Ca2+ after treatment with HR011303 alone. The calculated bile excretion index value was marked on the corresponding column. Data were expressed as mean + S.D. (n = 3), **, P < 0.005, ***, P < 0.001.
Efflux Transporter MRP2/Mrp2 Study
As presented in Fig. 9, the ratios of ATP-dependent uptake to the nonspecific adsorption (+ATP/-ATP ratio) of M5 were 33.35 (1 μM) and 15.27 (10 μM) for MRP2, while for Mrp2, the +ATP/-ATP ratios were 18.24 (1 μM) and 6.65 (10 μM). No significant ATP-dependence transport was observed in HR011303 for both MRP2 and Mrp2. Similarly, we conclude that M5 rather than HR011303 was the substrate of MRP2 and Mrp2, which also means that M5 produced in hepatocytes can be transported into bile by MRP2/Mrp2.
Direct transport of HR011303 and M5 by multidrug resistance protein (MRP) 2 (A) and Mrp2 (B). HR011303 or M5 was incubated with MRP2 and Mrp2-expressing membrane vesicle with or without the presence of ATP for 10 minutes. Data were expressed as mean + S.D. (n = 3).
Discussion
After the oral administration of [14C]HR011303, 87.20% of radioactive substances were excreted in the feces of rats versus 81.50% in human urine, which was completely opposite. Similar phenomena have been reported in previous studies. After administration of [14C]fenclofenac, the cumulative radioactivity recovery in feces accounted for 86% in rats. However, in humans, 93% of the radioactive compounds underwent urine excretion. Greeslade proposed that the extreme differences in excretion might be due to the difference in the molecular weight threshold for biliary excretion in diverse species (Greenslade et al., 1980). In addition, it is proposed that the molecular weight cutoffs for anionic compounds were approximately 400 Da in rats and 475 Da in humans. It means that anionic compounds with molecular weight over 400 Da are preferentially excreted in bile in rats (Yang et al., 2009). The results of a study on the excretion of pranoprofen in different species (rats, mice, guinea pigs, and rabbits) indicated that the drug was more likely to be excreted through urine with increasing urine pH level (Arima and Kato, 1990). Bromhexine also showed species differences in terms of excretion in mice, rats, rabbits, dogs, and humans. This phenomena can be explained by the differences in the ability of the species to conjugate basic metabolites (Kopitar et al., 1973). Lesinurad is a first-generation selective urate transporter 1 inhibitor (Hoy, 2016). According to published data, the radioactivity recovery rate shows a discrepancy in excretion between humans and rats. Approximately 75.3% of the radioactivity was excreted in rat feces, whereas 63.4% was excreted in human urine (EMA, 2016; Shah et al., 2019). The reason for the differences in lesinurad excretion among species has not been thoroughly investigated. In the present study, we attempted to explain the species differences in HR011303 excretion. The molecular weight of M5 is 513 Da, which is greater than 475 Da, M5 was effluxed into rat bile and was not well effluxed into human bile. Moreover, the carboxylic acid moiety of glucuronic acid is ionized at physiologic pH and then recognized by the renal organic anion transporter systems in humans, thus enabling glucuronides to be secreted in urine. The pH value of human urine was generally in the range of 4.6–8.4 (Klette et al., 2002), which was not obviously higher than that of rats (6.5–7.0). However, the proportion of HR011303-related compounds excreted in urine still largely differed between rats (10.06%) and humans (81.50%). Therefore, the remarkable difference in mass balance results in rats and humans was not well explained by the molecular weight and pH value. Other factors may be involved in the HR011303 disposition in humans and rats.
After HR011303 entered the human body, it was metabolized by UDP-glucuronosyltransferases to produce glucuronic acid conjugate M5 in both liver and kidney microsomes (companion manuscript, DMD-AR-2021-000581). Similarly, we studied the metabolic sites of HR011303 in rats. HR011303 can be metabolized to M5 only in the RLMs. M5 could not be generated from HR011303 in the RKMs. HR011303 can be metabolized to M5 in both human and rat hepatocytes, but urine is the major means of excretion in humans. Thus, M5 or HR011303 in human hepatocytes might have first returned to the blood and then into the kidney. The transport process of HR01103 and M5 from hepatocytes into the blood may exhibit species differences.
Glucuronides can be excreted into bile following their canalicular transport by MRP2 or urine following their sinusoidal transport into the blood by MRP3 (Parkinson et al., 2001). MRP3 is an ATP-dependent, unidirectional efflux pump localized in the basolateral (sinusoidal) membrane. It mediates the efflux of substrates, particularly glucuronides and glutathione conjugates, into sinusoidal blood at the expense of ATP hydrolysis (Borst and Elferink, 2002; Hillgren et al., 2013). Wang quantified and compared a range of hepatic transporters in various species and found that MRP3 expression level in rat liver tissue was much lower than that in human liver and was even lower than the LLOQ (0.23 fmol) (Wang et al., 2015). Based on this distribution difference, MRP3 might play an important role in the distribution and excretion of HR011303 and its major metabolite M5. HR011303 and M5 efflux were studied in human MRP3-expressing membrane vesicles. As we speculated, M5 rather than HR011303 can be effluxed into the blood via MRP3.
Although M5 is transported into the blood through MRP3, the abundance of M5 in plasma is low. On the one hand, M5 is unstable in plasma at the physiologic temperature of the human body (Fig. 6). When M5 is transported from the liver to the blood, most of it is hydrolyzed at 37°C. The sample preparation of quantifying M5 in human plasma by LC-MS/MS needs to be performed on ice water and requires the addition of protease inhibitors in freshly collected plasma before aliquoting. On the other hand, M5 can be taken up by the transporters OATP1B1, OATP1B3, and OAT3. Hence, part of the M5 in blood was taken up into the liver or kidney. All of these factors contributed to the low abundance of M5 in human plasma.
In humans and rats, HR011303 is mostly metabolized to M5 in the liver. In human liver, the formed M5 is effluxed into the blood by MRP3. By contrast, in rat liver, M5 is excreted into bile. Transporters on the bile duct side may be involved in this process. Humans are poorer biliary excreters compared with rats (Mahmood and Sahajwalla, 2002). The elimination rates of MRP2/Mrp2 specific substrates in isolated rat liver cells were 4–6 times faster than in human hepatocytes(Li et al., 2008). The amount of MRP2/Mrp2 in rat liver was approximately 10 times higher than those of other species, including humans. (Li et al., 2009). We used SCRH and SCHH to study the bile efflux of drug-related substances after administration of HR011303. Transporter inhibitors were used to explore the contribution of different transporters to the efflux process. In rats, HR011303 was metabolized to M5 in the liver and excreted as M5 into bile. In contrast, HR011303 and M5 did not show any tendency to be effluxed into bile in humans. The addition of MK571 (an inhibitor of Mrp2) can significantly inhibited the bile efflux of M5 in rats. Mrp2 may be the main transporters involved in the efflux of M5 into bile in rats. Considering the totally different distribution of MRP2/Mrp2 and MRP3/Mrp3 in human and rat hepatocytes and their respective roles in drug disposition, we proposed that the competition for M5 between MRP2/Mrp2 and MRP3/Mrp3 leads to a reverse excretion pathway in humans and rats.
By using human MRP2 and rat Mrp2-expressing membrane vesicles, we came to the conclusion that M5 was the substrate of both MRP2 and Mrp2. Combining that MRP2/Mrp2 expression level in rat liver is about 10 times that of a human, we speculate that M5 in rat liver is mainly excreted into bile through Mrp2, while in human hepatocytes, we conjecture that MRP3 performs better than MRP2 in the competitive transport of M5.
This study investigated the key transporters responsible for the species differences in HR011303 excretion and explored the reasons for the low plasma exposure of the main metabolite M5. The metabolism and distribution process after the oral administration of HR011303 in humans and rats are shown in Figs. 10 and 11. In humans, HR011303 enters into the liver and kidney through passive diffusion and is metabolized by UDP-glucuronosyltransferases to produce the major metabolite M5. The majority of M5 produced in the liver is effluxed into the blood by MRP3 and then enters the kidney as M5 or HR011303. In rats, HR011303 is metabolized in the liver to the main metabolite M5, which is then excreted into bile by Mrp2 and undergoes further metabolism in the gut before excretion in the feces.
Proposed scheme of the disposition and excretion processes of HR011303 and M5 in humans. The width of the arrows represents the relative amount of substrate undergoing a certain process.
Proposed scheme of the disposition and excretion processes of HR011303 and M5 in rats. The width of the arrows represents the relative amount of substrate undergoing a certain process.
Authorship Contributions
Participated in research design: Guo, Liu, Huang, Yu, Zhong, Li, Chen XY, Diao.
Conducted experiments: Guo, Liu, Meng, Xue.
Performed data analysis: Guo, Liu, Xue, Wu, Chen ZD, Diao.
Wrote or contributed to the writing of the manuscript: Guo, Liu, Zheng, Wu, Chen XY, Diao.
Footnotes
- Received June 14, 2021.
- Accepted October 28, 2021.
↵1 Z.G. and M.L. contributed equally to this work.
This research was financially supported by a grant from the National Natural Science Foundation of China [Grant 81903701].
G.Z.L. and Q.H. are employees of Jiangsu Hengrui Medicine Co., Ltd., and the other authors declare that they have no competing interest.
Abbreviations
- BCRP
- breast cancer resistance protein
- BEI
- bile excretion index
- HBSS
- Hank’s balanced salt solution
- HEK
- human embryonic kidney
- LC-MS/MS
- liquid chromatography/tandem mass spectrometry
- MRP
- multidrug resistance-associated protein
- OAT
- organic anion transporter
- OATP
- organic anion-transporting polypeptide
- RKM
- rat kidney microsome
- RLM
- rat liver microsome
- SCHH
- sandwich-cultured human hepatocytes
- SCRH
- sandwich-cultured rat hepatocytes
- SD
- Sprague-Dawley
- UDPGA
- 5′-diphosphoglucuronic acid
- Copyright © 2022 by The American Society for Pharmacology and Experimental Therapeutics