Abstract
It has been reported that in vivo biliary clearance can be predicted using sandwich-cultured rat and human hepatocytes. The predicted apparent biliary clearance (CLbile, app) from sandwich- cultured rat hepatocytes (SCRH) based on medium concentrations correlates to in vivo CLbile, app based on plasma concentrations of angiotensin II receptor blockers (ARBs), HMG-CoA reductase inhibitors (statins), β-lactam antibiotics, and topotecan. However, the predicted biliary clearance from SCRH was 7- to 300-fold lower than in vivo biliary clearance. We speculated that the process of biliary excretion might not have been evaluated using sandwich-cultured hepatocytes. To evaluate this issue, intrinsic biliary clearance (CLbile, int) based on intracellular compound concentrations was evaluated to investigate the in vitro-in vivo correlation of this process among ARBs, statins, β-lactam antibiotics, and topotecan. Intrinsic biliary clearance in SCRH correlated to in vivo values obtained by constant intravenous infusion of six compounds, but not rosuvastatin and cefmetazole, to rats. Moreover, differences between SCRH and in vivo CLbile, int (0.7–6-fold) were much smaller than those of CLbile, app (7–300-fold). Therefore, in vivo CLbile, int is more accurately reflected using SCRH than CLbile, app. In conclusion, to predict in vivo biliary clearance more accurately, CLbile, int should be evaluated instead of CLbile, app between SCRH and in vivo.
Introduction
Hepatic clearance is the most important factor governing drug availability because the liver is the primary organ responsible for eliminating xenobiotics in the body. Metabolism, hepatic uptake, and biliary excretion contribute to the elimination of administered drugs. Influx transporters [e.g., organic anion-transporting polypeptide (Oatp)] and efflux transporters [e.g., multidrug resistance (Mdr), multidrug resistance associated protein (Mrp), and breast cancer resistance protein (Bcrp)] expressed in the liver play important roles in regulating drug disposition and elimination (Hirano et al., 2004, 2005; Nakagomi-Hagihara et al., 2006; Yamashiro et al., 2006; Kitamura et al., 2008). Predicting human pharmacokinetics using in vitro models and preclinical species efficiently accelerates the process of discovering new drug candidates. Through drug candidate discovery and development, highly accurate forecasting of metabolic clearance has become possible using biological materials (Iwatsubo et al., 1997; Ito et al., 1998). Although biliary excretion is an important route for drug elimination (Levine, 1978; Rollins and Klaassen, 1979), evaluating biliary excretion of drug candidates in humans is difficult because of the scarcity of clinical bile samples. Therefore, it is necessary to predict biliary excretion in humans.
Hepatocytes are a widely accepted in vitro tool for evaluating mechanisms of hepatic uptake and metabolism of xenobiotics and hepatotoxicity (Iwatsubo et al., 1997; Kato et al., 2002; Hirano et al., 2005). However, it was reported that cell polarity and liver-specific functions, such as albumin secretion, hepatic uptake, and enzyme activity under conventional monolayer culture conditions were rapidly lost (Foliot et al., 1985; Dunn et al., 1989). In contrast with conventional conditions, sandwich-cultured hepatocytes maintain liver-specific functions for several days and exhibit the formation of bile canaliculi and the localization of efflux transporters on the canalicular membrane (LeCluyse et al., 1994; Talamini et al., 1997). Moreover, the biliary clearance of some drugs, such as angiotensin II receptor blockers, HMG-CoA reductase inhibitors, and β-lactam antibiotics, showed correlation between sandwich-cultured rat or human hepatocytes and in vivo (Liu et al., 1999a; Abe et al., 2008, 2009; Fukuda et al., 2008; Li et al., 2010). Although previous reports have suggested that sandwich-cultured hepatocytes can be used to predict the in vivo biliary clearance of drug candidates, differences in biliary clearance based on the medium or plasma concentration are greater than 10-fold between sandwich-cultured rat or human hepatocytes and in vivo values. This approach may be useful for predicting the rank order of in vivo biliary clearance, but predicting the precision of biliary clearance using sandwich-cultured hepatocytes is questionable. Because the protein expression of rat Oatp decreases in culture (Hoffmaster et al., 2004; Zhang et al., 2005), decreased influx transporter activities result in differences in biliary clearance between sandwich-cultured rat hepatocytes (SCRH) and in vivo for Oatp substrates. In contrast, the protein expression of efflux transporters is relatively maintained during culture (Hoffmaster et al., 2004; Zhang et al., 2005). In several reports, the apparent biliary clearance based on medium concentration was evaluated in SCRH. However, it has not been evaluated whether intrinsic biliary clearance in sandwich-cultured hepatocytes would reflect the process of biliary excretion in vivo.
To evaluate this issue, intrinsic biliary clearance based on intracellular concentrations evaluated in SCRH was compared with that based on the concentrations in liver under steady-state conditions by constant intravenous infusion to rats.
Materials and Methods
Chemicals.
Rosuvastatin and valsartan were purchased from Toronto Research Chemicals, Inc. (North York, ON, Canada). Cefoperazone, diclofenac, and collagenase type I were purchased from Wako Pure Chemicals (Osaka, Japan). Cefmetazole and cefpiramide were purchased from the United States Pharmacopeial Convention (Rockville, MD). Pravastatin was purchased from Cayman Chemical (Ann Arbor, MI). Topotecan was purchased from Enzo Life Sciences AG (Lausen, Switzerland). Olmesartan was synthesized by Takeda Pharmaceutical Co. Ltd. (Kanagawa, Japan). Hanks' balanced salt solution (HBSS) and Ca2+/Mg2+-free HBSS were purchased from Invitrogen (Carlsbad, CA). Matrigel and collagen type I-coated 24-well BioCoat plates were obtained from BD Biosciences (San Jose, CA). The hepatocyte plating medium (InVitroGRO CP medium), culture medium (InVitroGRO HI medium), and Torpedo Antibiotic Mix were purchased from Celsis In Vitro Technologies (Baltimore, MD). BCA protein assay kits were purchased from Thermo Fisher Scientific (Waltham, MA). All other reagents and solvents were purchased from Wako Pure Chemicals and Sigma-Aldrich (St. Louis, MO).
Animals.
Male Sprague-Dawley rats were purchased from Charles River Japan Inc. (Shiga, Japan) and acclimatized for more than 7 days before the experiment. Rats were housed under controlled temperature and humidity with a 12-h light/dark cycle. Laboratory chow (CE-2; CLEA Japan Inc., Tokyo, Japan) and water were available with free access. All animal experiments were conducted in accordance with the guidelines of the Experimental Animal Care and Use Committee of Takeda Pharmaceutical Co., Ltd.
Hepatocyte Isolation and Culture.
Hepatocytes were isolated from male Sprague-Dawley rats by a modification of the two-step collagenase digestion method described previously (Nakakariya et al., 2008). Rats (7–8 weeks old) were anesthetized by intraperitoneal injection of pentobarbital. Rat livers were perfused with Ca2+/Mg2+-free HBSS containing 1 mM EGTA for approximate 10 min at a flow rate of 30 ml/min, which was followed by perfusion with HBSS containing 1 mg/ml collagenase type I for approximate 10 min at a flow rate of 15 ml/min. Hepatocytes were dispersed from the digested liver in Krebs-Henseleit buffer with 2% bovine serum albumin and rinsed by repeated low-speed centrifugation at 50g for 5 min at 4°C. The cell pellet was resuspended in 35% (v/v) isotonic Percoll in Krebs-Henseleit buffer and centrifuged at 100g for 20 min at 4°C. After the resultant pellet was resuspended in plating medium, the number of viable cells was determined using trypan blue staining. Only cells with viability greater than 90% were used for further studies. Hepatocytes were seeded on 24-well collagen-coated culture plates at a density of 5 × 105 cells/well in 0.5 ml and allowed to attach for 2 to 3 h at 37°C in a humidified incubator with 95%/5% of air/CO2. Next, the medium was aspirated, and the hepatocytes were washed once with plating medium. Fresh plating medium was added to the cultures. On the 2nd day after hepatocyte plating, hepatocytes were overlaid with BD Matrigel at a concentration of 0.25 mg/ml in ice-cold culture medium. The culture medium was refreshed every 24 h for 72 h.
Transport Studies in SCRH.
The uptake study was performed using a previously reported method (Liu et al., 1999a). In brief, SCRH were rinsed three times with 0.5 ml of HBSS (standard buffer) or Ca2+/Mg2+-free HBSS with 1 mM EGTA [(−) HBSS] and preincubated in 0.5 ml of standard buffer or (−) HBSS at 37°C for 10 min. After the buffer was removed by aspiration, an uptake reaction was initiated by addition of 0.5 ml of substrate-containing standard buffer and terminated by rinsing three times with 0.5 ml of ice-cold standard buffer. Cells were lysed with 0.25 ml of 0.5% Triton X-100 in 50 mM phosphate buffer by shaking for approximately 30 min at room temperature. Cell lysate samples were mixed with acetonitrile containing alprenolol and diclofenac at a concentration of 100 ng/ml, which was used as an internal standard, and then centrifuged. The supernatants were diluted with an appropriate volume of 10 mM acetic ammonium or 0.2% (v/v) formic acid in 10 mM ammonium formate and analyzed by liquid chromatography with tandem mass spectrometry (LC-MS/MS). A BCA protein assay kit was used to determine the protein concentration in the cells lysed with 0.5% Triton X-100.
Fluorescence Microscopy.
Retention of 5 (and 6)-carboxy-2′,7′-dichlorofluorescein (CDF) in bile canalicular lumen was examined by fluorescence microscopy. A transport assay was performed using the method described under Transport Studies in SCRH. The cells and bile canaliculi were imaged with an inverted fluorescence microscope (CKX41; Olympus, Tokyo, Japan).
Determination of Plasma Protein Binding.
Plasma protein binding of test compounds was determined using the equilibrium dialysis method. In brief, 10 μl of 100 μM test compound in dimethyl sulfoxide was added to 1 ml of pooled rat plasma to a final concentration of 1 μM. Plasma (150 μl) was placed on one side (plasma side) of a dialysis membrane, and 50 mM phosphate buffer (150 μl) was placed on the opposite side (buffer side). After the tops of both sides were sealed, the plate was placed on a single plate rotator (Harvard Apparatus, Inc., Holliston, MA) set at 20 rpm for 24 h at room temperature and protected from light. The supernatants were mixed with acetonitrile containing alprenolol and diclofenac at a concentration of 100 ng/ml, which was used as an internal standard, and then centrifuged. Supernatants were diluted with an appropriate volume of 10 mM acetic ammonium or 0.2% (v/v) formic acid in 10 mM ammonium formate and analyzed using LC-MS/MS.
The unbound fraction in rat plasma (fu, p) was calculated using eq. 1: where Cbuffer and Cplasma represent the drug concentration on the buffer and plasma sides, respectively.
In Vivo Pharmacokinetic Studies in Rats during Continuous Intravenous Infusion.
While under anesthesia induced by isoflurane, 9-week-old rat femoral veins and common bile ducts were cannulated using 3.0-French (i.d. 0.61 mm; o.d. 0.97 mm) and 2.0-French (i.d. 0.33 mm; o.d. 0.66 mm) catheters, respectively. Animals were held in Bollman cages during experiments. Test compounds were administered at a constant rate (0.2 ml/h) by intravenous infusion through the femoral vein to achieve a steady-state plasma concentration of compound. Rosuvastatin, pravastatin, topotecan, cefmetazole, and cefoperazone were dissolved in saline at a concentration of 1 mg/ml. Valsartan, olmesartan, and cefpiramide were dissolved in a mixture of saline and dimethylacetamide (1:1, v/v) at a concentration of 1 mg/ml. Blood and bile were collected at designated times, and the blood was centrifuged to obtain plasma. Livers were collected from sacrificed rats 5 h after the infusion was begun. Plasma, bile, and liver samples were frozen at −80°C until analysis. Plasma, bile, and liver samples were mixed with acetonitrile containing alprenolol and diclofenac at a concentration of 100 ng/ml as an internal standard and then centrifuged. The supernatants were diluted with an appropriate volume of 10 mM acetic ammonium or 0.2% (v/v) formic acid in 10 mM ammonium formate and analyzed using LC-MS/MS.
LC-MS/MS Analysis.
LC-MS/MS analysis was conducted using an API 4000 triple quadruple mass spectrometer (Applied Biosystems, Foster City, CA) coupled with a TurboIonSpray interface in positive or negative ion mode and connected to a ultra-fast liquid chromatograph (Shimadzu, Kyoto, Japan). Reverse-phase chromatography [mobile phase A, 10 mM acetic ammonium for negative ion mode or 0.2% (v/v) formic acid in 10 mM ammonium formate for positive ion mode; mobile phase B, methanol] was used to elute and separate the various substrates over a Shim-pack XR-ODS C18 column (20 × 2.0 mm, 5 μm; Shimadzu). Because the chemical structure of topotecan is pH-dependent, the total amounts of topotecan (lactone and carboxylate form) were determined at an acidic pH. Injections of 10 μl were analyzed at a flow rate of 0.3 ml/min. The following transitions (precursor ion m/z > product ion m/z) were monitored: 480.2 > 243.9 for rosuvastatin, 423.2 > 101.1 for pravastatin, 434.0 > 179.0 for valsartan, 644.1 > 114.9 for cefoperazone, 472.2 > 215.0 for cefmetazole, 613.1 > 122.3 for cefpiramide, 447.3 > 207.2 for olmesartan, 422.0 > 377.0 for topotecan, 250.2 > 116.3 for alprenolol, and 293.8 > 249.9 for diclofenac. Peak areas of all analytes were integrated and quantified using Analyst 1.4.2 (Applied Biosystems).
Data Analysis.
The biliary excretion index (BEI) (percentage) and in vitro apparent biliary clearance (CLbile, app in milliliters per hour per kilogram) were calculated in hepatocytes based on eqs. 2 and 3: where accumulation [(+)Ca2+/Mg2+] and accumulation [(−)Ca2+/Mg2+] represent the accumulation of test compounds in SCRH preincubated in standard buffer and (−) HBSS, respectively. The BEI and CLbile, app were determined after a 15-min incubation. Drug concentrations in the medium were defined as initial substrate concentration in the incubation medium. The units of in vitro CLbile, app were converted from microliters per minute per milligram of protein to milliliters per hour per kilogram using the following parameters: rat liver weight and protein content in liver tissue were assumed to be 40 g/kg b.wt. and 200 mg protein/g liver wt., respectively (Seglen, 1976; Davies and Morris, 1993).
The predicted CLbile, app from in vitro values were estimated according to the eqs. 4 and 5 below based on the well stirred model of hepatic disposition. where Qp represents the hepatic plasma flow rate (2268 ml · h−1 · kg−1). In eq. 4, fu, p was assumed to be unity. When the unbound fraction is taken into consideration,
In vivo CLbile, app was calculated according to eq. 6: where AUCplasma represents the area under the plasma concentration-time curve. AUCplasma and percentage dose in bile values were obtained from references.
The intrinsic biliary clearance (CLbile, int), based on the compound concentration in liver and hepatocytes, was calculated according to eqs. 7 (in vivo) and 8 (in vitro), respectively. Compound concentration in hepatocytes was calculated based on eq. 9. where Xbile, 4 h–5 h and AUCliver, 4 h–5 h represent the amount of compounds excreted into bile between 4 and 5 h after the infusion was begun and the area under the liver concentration-time curve between 4 and 5 h after infusion was begun, respectively. When CLbile, int was compared between SCRH and in vivo, the unbound fraction was presumed to be equal between SCRH and liver. Therefore, CLbile, int was calculated based on the total concentration of compounds in hepatocytes and liver. In vitro CLbile, int was determined after 15 min of incubation. Intracellular space (5.2 μl/mg protein) was obtained from reference values (Yamano et al., 1999).
Results
Cell Morphology and Construction of Bile Canaliculi in SCRH.
Figure 1 shows a phase-contrast image of SCRH cultured for 4 days and fluorescence images of SCRH after a incubation with CDF diacetate in standard buffer and (−) HBSS. Fluorescent CDF, an Mrp2 substrate, was localized in the bile canalicular lumen after incubation in standard buffer (Fig. 1B). In contrast, CDF was not localized in the bile canalicular lumen, but only intracellular accumulation of CDF was observed in (−) HBSS (Fig. 1C).
Transport Study of Compounds in SCRH.
Eight compounds (rosuvastatin, pravastatin, valsartan, olmesartan, cefmetazole, cefoperazone, cefpiramide, and topotecan) secreted into bile and reportedly observed in in vitro-in vivo correlation (IVIVC) were selected to confirm the IVIVC of CLbile, app in SCRH (Abe et al., 2008; Fukuda et al., 2008; Li et al., 2010). Compound accumulations in SCRH are shown in Fig. 2. Accumulation of the test compounds in standard buffer was higher than that in (−) HBSS, except for cefmetazole after a 15-min incubation. Table 1 summarizes the BEI and CLbile, app of the test compounds in SCRH. The CLbile, app of the test compounds determined in SCRH was converted to milliliters per hour per kilogram by determining total protein in each well and using the physiological parameters of the rat (40 g liver/kg b.wt. and 200 mg protein/g liver). Nearly all test compound data were similar to reported data (Abe et al., 2008; Fukuda et al., 2008; Li et al., 2010).
IVIVC of Apparent Biliary Clearance in Compounds.
A well stirred hepatic model was applied to the predicted CLbile, app from in vitro data by eqs. 4 and 5, and in vivo data were obtained from published data (Tables 1 and 2). When CLbile, app was predicted on the basis of the plasma total concentration of the compound using eq. 4, the predicted CLbile, app did not correlate to CLbile, app in vivo (Fig. 3A). However, the predicted CLbile, app based on the plasma unbound concentration of the compound using eq. 5 was correlated to CLbile, app in vivo (Fig. 3B), even though the predicted CLbile, app was 7- to 300-fold lower than in vivo CLbile, app.
Plasma and Liver Concentration and Cumulative Biliary Excretion of Drugs in Rats.
To evaluate the CLbile, int on the basis of liver concentration in vivo, eight drugs were administered to rats by continuous intravenous infusion. The plasma compound concentration and the biliary excretion-time curves of the test compounds are shown in Fig. 4. Plasma concentration (Fig. 4A) and biliary excretion (Fig. 4B) in 1 h of all the compounds reached steady-state by 4 h after the infusion was begun. As shown in Table 3, the CLbile, int in vivo calculated using eq. 7 was relatively high for pravastatin and cefoperazone and low for topotecan.
IVIVC of Intrinsic Biliary Clearance in Drugs.
The CLbile, int of the test compounds was also evaluated in SCRH using eq. 8, and are summarized in Table 4. The CLbile, int of the test compounds determined in the SCRH as well as CLbile, app was converted to milliliters per hour per kilogram. In vitro CLbile, int based on the intracellular total concentration of compound was plotted against in vivo CLbile, int (Fig. 5). The CLbile, int in SCRH correlated to that in vivo for six compounds tested (except rosuvastatin and cefmetazole). Moreover, compared with the CLbile, app based on the plasma unbound concentration of the compound (7–300-fold), differences in CLbile, int between SCRH and in vivo were low (0.7–6-fold).
Discussion
Biliary excretion is one of the primary elimination routes of xenobiotics and their conjugates, along with metabolism and renal secretion. However, the prediction of biliary excretion in humans using in vitro tools has not been established. It was reported that in vivo biliary clearance can be predicted from sandwich-cultured hepatocytes in rats and humans (Liu et al., 1999a; Abe et al., 2008, 2009; Fukuda et al., 2008; Li et al., 2010). Although apparent biliary clearance using sandwich-cultured rat or human hepatocytes correlates to in vivo values, the differences in biliary clearance between in vitro and in vivo are more than 10-fold. Therefore, sandwich-cultured hepatocytes may not be an appropriate method of examining biliary excretion. In the present study, intrinsic biliary clearance based on intracellular concentration was examined in SCRH and in vivo, and we found that intrinsic biliary clearance in SCRH could reflect in vivo processes.
To evaluate biliary excretion in SCRH, rosuvastatin, pravastatin, valsartan, olmesartan, cefmetazole, cefoperazone, cefpiramide, and topotecan were selected as model compounds. The compounds were excreted into bile canalicular lumen in SCRH, suggesting that functional canalicular transporters expressed and contributed compounds to excrete into bile canaliculi in SCRH (Fig. 2). Next, the IVIVC of apparent biliary clearance based on the medium or plasma concentration of these compounds was evaluated. The correlation between BEI obtained from SCRH and the in vivo biliary excretion ratio was not observed (data not shown). Meanwhile, the predicted CLbile, app with plasma protein binding normalization was correlated with the corresponding CLbile, app in vivo values (Fig. 3B). These results are consistent with published data (Abe et al., 2008; Fukuda et al., 2008; Li et al., 2010). However, the CLbile, app in vivo values were 7- to 300-fold of the predicted CLbile, app from SCRH (Tables 1 and 2). It was reported that the hepatic clearance of angiotensin II receptor blockers, HMG-CoA reductase inhibitors, and β-lactam antibiotics predicted based on uptake clearance in rat isolated hepatocytes was almost comparable to in vivo biliary clearance (Watanabe et al., 2009, 2010a,b). These results suggested that the rate-limiting step of hepatic clearance in these compounds is the hepatic uptake process. In SCRH, protein expression and uptake activity of Oatp substrates decreased during the culture period (Liu et al., 1999b; Hoffmaster et al., 2004). In contrast, the protein expression of efflux transporters was relatively maintained during culture (Hoffmaster et al., 2004; Zhang et al., 2005). Therefore, differences in apparent biliary clearance between SCRH and in vivo may be caused by decreased influx transporter activity during the culture period. Thus, the IVIVC of CLbile, app in test compounds may be reflected in the uptake clearance, and SCRH may not accurately reflect the biliary excretion process. To investigate whether CLbile, int is correlated between SCRH and in vivo, the CLbile, int determined using hepatocytes and liver concentration under steady-state conditions was evaluated in vitro and in vivo, respectively. The intracellular unbound fraction of test compounds was presumed to be comparable between in vitro and in vivo when similar concentrations of compounds were used; CLbile, int was calculated on the basis of the total concentration of compounds in liver and hepatocytes (eqs. 7 and 8). The CLbile, int in SCRH correlated to that in rats for six compounds tested (except rosuvastatin and cefmetazole) (Fig. 5). Moreover, compared with CLbile, app based on medium or plasma concentration of the compounds (7–300-fold), differences of CLbile, int between SCRH and in vivo were low (0.7–6-fold) (Table 4).
There are three possible explanations for the poor correlation of rosuvastatin and cefmetazole in CLbile, int and the differences of CLbile, int between SCRH and in vivo. First, the intracellular unbound concentration of compounds in SCRH was different from that in vivo. Second, there was no dynamic blood and bile flow in SCRH. Finally, efflux transporter activities changed in SCRH. For the first possibility, binding components were determined to be comparable between hepatocytes and liver tissue, and the unbound fraction was presumed to be equivalent both in vitro and in vivo. It was reported that the liver to plasma concentration ratio in vivo was predicted using rat hepatocytes among various compounds in vitro (Yamano et al., 1999). These results indicated that unbound fraction in vitro could be comparable with that in vivo and supported our presumption. Moreover, there was no possibility of nonlinear unbound fractions under conditions in which the intracellular compound concentration was nearly equal in vitro and in vivo (Table 4). These factors rule out the first possibility. In contrast, the other two possibilities may have to be taken into account. The lack of dynamic flow and limited space of bile canaliculi in the SCRH should result in quick saturation of biliary secretion and underestimation of biliary clearance. Figure 6 shows accumulation profiles of rosuvastatin and pravastatin. Because the accumulation of these compounds did not change from 3 min after incubation was begun, a time-dependent change of biliary excretion was not observed in SCRH. However, it would be difficult to evaluate stably biliary excretion within 3 min for all compounds in practical, especially valsartan and olmesartan, for which BEI is relatively low. Therefore, a lack of dynamic flow may cause the differences in CLbile, int between SCRH and in vivo. Transporter expression levels may also be a determining factor for predicting biliary clearance because transporter-mediated hepatobiliary secretion is the predominant route for compound biliary clearance. Although the decreased protein expression of efflux transporters was lower than that of influx transporters in SCRH (Hoffmaster et al., 2004; Zhang et al., 2005), the protein expression of efflux transporters changed by 0.3- to 3-fold during the culture period (Hilgendorf et al., 2007; Olsavsky et al., 2007). It was reported that Mrp2 and primarily Bcrp contributes to biliary secretion of rosuvastatin (Kitamura et al., 2008). Because Bcrp protein expression increased during culture, in contrast to the decrease in Mrp2 and bile salt export pump expression (Li et al., 2010), intrinsic biliary clearance for rosuvastatin is relatively higher than that of other compounds in SCRH. Meanwhile, it was reported that Mrp2 mainly contributes to biliary secretion of pravastatin, angiotensin receptor blockers (valsartan and olmesartan), and β-lactam antibiotics (cefoperazone and cefpiramide) (Muraoka et al., 1995; Sasaki et al., 2004; Takayanagi et al., 2005; Yamashiro et al., 2006; Kato et al., 2008). Because the main contributor that excreted rosuvastatin into bile was different from that for the other compounds, a deviation of rosuvastatin from the correlation among these compounds would be observed. Meanwhile, it was reported that cefmetazole was not a substrate of Mrp2 and Bcrp (Kato et al., 2008). Although the primary transporter for biliary excretion of cefmetazole is unknown, it may be possible that the activity change of the main contributors for biliary excretion of cefmetazole caused the deviation of cefmetazole from the correlation. To overcome these issues, it has been reported that biliary clearance can be corrected by using the ratio of the protein amount of efflux transporters in rat liver to that in SCRH, improving the IVIVC of CLbile, app (Li et al., 2010). This correction is more suitably applied to CLbile, int than CLbile, app because CLbile, int is focused on the process from liver to bile and involved in efflux transporter activity. In the future, the correction of efflux transporter activity between SCRH and isolated hepatocytes before seeding should be evaluated. To evaluate the whole hepatobiliary disposition, only SCRH is not appropriate, and elementary steps of apparent biliary clearance should be evaluated using both SCRH and freshly isolated hepatocytes. Because it was reported that uptake clearance using freshly isolated hepatocytes would be comparable to hepatic clearance in vivo (Watanabe et al., 2009), the evaluation of uptake clearance using freshly isolated hepatocytes could be reasonable to cover the defect in SCRH.
In summary, compared with CLbile, app (7–300-fold), the absolute values of CLbile, int in vitro were very similar to in vivo values (0.7–6-fold). This finding suggests that CLbile, int in sandwich-cultured hepatocytes can more accurately reflect the process of biliary excretion in vivo than CLbile, app. For accurate in vivo prediction, CLbile, int should be evaluated instead of CLbile, app between SCRH and in vivo.
Authorship Contributions
Participated in research design: Nakakariya, Ono, Amano, Moriwaki, Maeda, and Sugiyama,
Conducted experiments: Nakakariya.
Performed data analysis: Nakakariya.
Wrote or contributed to the writing of the manuscript: Nakakariya, Ono, Amano, Moriwaki, Maeda, and Sugiyama.
Acknowledgments
We thank Yoshiaki Kimura and Atsutoshi Furuta (Takeda Pharmaceutical Co., Ltd., Kanagawa, Japan) for technical support in the animal study.
Footnotes
This work was supported in part by the Japan Health Science Foundation [Grant “Research on Publicly Essential Drugs and Medical Devices”].
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
ABBREVIATIONS:
- Oatp
- organic anion-transporting polypeptide
- Mdr
- multidrug resistance
- Mrp
- multidrug resistance-associated protein
- Bcrp
- breast cancer resistance protein
- SCRH
- sandwich-cultured rat hepatocytes
- HBSS
- Hanks' balanced salt solution
- LC-MS/MS
- liquid chromatography tandem mass spectrometry
- CDF
- 5 (and 6)-carboxy-2′,7′-dichlorofluorescein
- BEI
- biliary excretion index
- IVIVC
- in vitro-in vivo correlation.
- Received August 1, 2011.
- Accepted December 21, 2011.
- Copyright © 2012 by The American Society for Pharmacology and Experimental Therapeutics