Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

We reported previously that some quinolones are well distributed into most tissues via a common mechanism (Okezaki et al., 1988) and that, after crossing the blood-brain barrier, they induce severe convulsions both by themselves and when coadministered with fenbufen (Tsuji et al., 1988a,b). HSR-903, a new quinolone antibacterial agent with potent antibacterial activity (Takahashi et al., 1997), is well distributed into tissues and has a low toxicity (Murata et al., 1995). Although most quinolones are predominantly excreted into urine after oral administration, HSR-903 was mainly excreted into bile (unpublished observations, Takagi T and Takahara E), and this excretion profile is very similar to those of the recently developed drugs sparfloxacin and grepafloxacin (Matsunaga et al., 1991; Akiyama et al., 1995). It is desirable to elucidate the mechanisms controlling the excretion routes of HSR-903, as was performed previously for -lactam antibiotics (Tamai et al., 1985; Tamai and Tsuji, 1987; Terasaki et al., 1986; Tsuji et al., 1985, 1989) and grepafloxacin (Sasabe et al., 1997, 1998a,b).

The hepatobiliary transport of drugs is generally considered to be a complex process consisting of uptake by hepatic parenchymal cells, intracellular transport, and excretion into the bile across the bile canalicular membrane. Studies on the process of excretion of drugs across the bile canalicular membrane have been carried out with bile canalicular membrane vesicles and with mutant rats having a hereditary defect in the biliary excretion of organic anions. Three types of mutant rats, i.e. the Wistar strains TR (Jansen et al., 1985) and GY (Kuipers et al., 1988) and the Sprague-Dawley strain EHBR¹ (Mikami et al., 1986; Hosokawa et al., 1992), have been reported. Shimamura et al. (1994) demonstrated the existence of multiple systems for the transport of organic anions across the bile canalicular membrane, by showing that the canalicular transport of dibromosulfophthalein is mediated by the primary active transporter that is defective in EHBR, whereas that of indocyanine green is mediated by another transporter, which is present in EHBR (Sathirakul et al., 1993, 1994).

In our preliminary study, it was found that the biliary excretion of HSR-903 was dramatically reduced in EHBR. Therefore, in the present study, we first measured the biliary excretion after iv administration of unlabeled HSR-903 or its stereoisomer, (R)-HSR-903, in SDR and EHBR. We then elucidated the mechanism of the excretion of HSR-903 into bile, by using isolated hepatocytes from the two types of rats.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. HSR-903 [(S)-()-5-amino-7-(7-amino-5-azaspiro[2.4]hept-5-yl)1-cyclopropyl-6-fluoro-1,4-dihydro-8-methyl-4-oxoquinoline-3-carboxylic acid methanesulfonate], [¹⁴C]HSR-903 (specific activity, 256 kBq/mg of base) (fig. 1), and other quinolone derivatives were synthesized by Hokuriku Seiyaku Co. (Fukui, Japan). [³H]Inulin and [³H]H₂O were purchased from New England Nuclear (Boston, MA). Collagenase (used for the preparation of cell suspensions) and 2,4-dinitrophenol were obtained from Wako Pure Chemical Industries (Osaka, Japan). Bovine serum albumin fraction V, BSP, taurocholic acid, and cimetidine were purchased from Sigma Chemical Co. (St. Louis, MO). Sodium azide was obtained from Nacalai Tesque (Kyoto, Japan). All other reagents were commercial products and of reagent grade.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Chemical structure of [14C]HSR-903. *, position of radiolabeling.

Animals. Male SDR (220-250 g) were purchased from Charles River Japan (Kanagawa, Japan) and male EHBR (270-320 g) were obtained from SLC (Shizuoka, Japan). These rats were allowed free access to laboratory chow and water.

In Vivo Study. The study was performed according to the Guidelines for the Care and Use of Laboratory Animals on Takara-machi Campus of Kanazawa University and was approved by the Committee of Ethics of Animal Experimentation of Kanazawa University, Takara-machi Campus. The animals were lightly anesthetized with diethyl ether, and the common bile duct was cannulated with polyethylene tubing to collect bile samples. The rats were allowed to recover from anesthesia before the experiments. HSR-903 or its R-isomer (5 mg/kg), in saline solution, was injected iv through the tail vein. Blood, bile, and urine were obtained at designated times.

Integration Plot Analysis. After iv bolus administration of [¹⁴C]HSR-903 and [³H]inulin (5 mg/kg each) to rats, blood samples were collected at designated times. After an appropriate time, the rats were killed, liver tissues were excised, and tissue weight and the associated radioactivity were measured. When the tissue uptake was measured within a short period, during which the efflux and/or elimination of radioactivity from the tissue were negligible, the tissue uptake at time t was proportional to AUC_0-t. By dividing C_liver,t and AUC_0-t by C_p,t, CL_uptake was obtained from the slope of the plot of C_liver,_t/C_p,t (i.e. liver/plasma concentration ratio at time t, in milliliters per gram of tissue) (on the ordinate) vs. AUC_0-t/C_p,t (on the abscissa) (Kim et al., 1988; Liu et al., 1992).

In Vitro Studies with Isolated Rat Hepatocytes. Hepatocytes were isolated from SDR or EHBR as described previously (Tsuji et al., 1985). Cell viability was routinely checked by the trypan blue exclusion method and lactate dehydrogenase-latency test (Moldeus et al., 1978). The hepatocytes were suspended in Krebs-Henseleit buffer and used at a concentration of 1 × 10⁶ cells/0.2 ml. Protein concentrations were determined by the method described by Lowry et al. (1951), with bovine serum albumin as a standard.

Analytical Method. The concentrations of unchanged HSR-903 and its R-isomer were determined by HPLC assay. Briefly, bile (0.5 ml), urine (0.5 ml), or plasma (0.1 ml) was mixed well with 0.1 ml of 0.067 M phosphate buffer (pH 7.0). This sample was mixed well with 0.1 ml of 1 N NaOH and 3 ml of diethyl ether and then centrifuged at 3000 rpm for 5 min. To the resultant aqueous layer, 0.5 ml of 1 M phosphate buffer (pH 7.0) and 6 ml of chloroform/isoamyl alcohol (95:5, v/v) were added, and the mixture was vigorously shaken for 10 min. After centrifugation of the mixture at 3000 rpm for 10 min, a 5-ml aliquot of the organic layer was obtained and evaporated to dryness at 37°C under reduced pressure. The residue was dissolved in 0.5 ml of 0.1 M citrate buffer (pH 4.0)/acetonitrile (3:1, v/v) for HPLC assay. For the determination of conjugates of HSR-903 or its R-isomer, bile (0.5 ml), urine (0.5 ml), or plasma (0.1 ml) was mixed well with 0.1 ml of 0.5 M acetic acid buffer (pH 5.0) containing -glucuronidase (4000 units in 0.5 ml) and was incubated at 37°C for 24 hr. The resultant mixture (0.5 ml of bile, 0.5 ml of urine, or 0.1 ml of plasma) was mixed well with 0.1 ml of 0.067 M phosphate buffer (pH 7.0), and the subsequent procedure was as described above.

Data Analysis. Kinetic parameters (K_t, J_max, and k_d) for the concentration-dependent uptake study were estimated by iterative nonlinear least-squares analysis using the MULTI program (Yamaoka et al., 1981), according to the following equation (1):

(1)

where v is the initial uptake rate of the drug (nanomoles per minute per milligram of protein), K_t is in millimolar, J_max is in nanomoles per minute per milligram of protein, and k_d is in microliters per minute per milligram of protein. PS_u,influx (in microliters per minute per milligram of protein) was calculated according to the following equation (2):

(2)

The PS_u,influx value was then converted to milliliters per minute per kilogram using following values: 10⁶ cells = 1.7 mg of protein, 1 g of liver = 1.25 × 10⁸ cells (Nakamura et al., 1994), and liver weight = 44 g/kg of body weight.

(3)

(4)

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Plasma Profiles for HSR-903 (S-Isomer), Its R-Isomer, and Their Conjugates. Fig. 2 shows the plasma profiles for HSR-903, its R-isomer, and their glucuronides after iv bolus administration of HSR-903 or the R-isomer to both SDR and EHBR. The pharmacokinetic parameters obtained are summarized in table 1. The plasma concentrations of unchanged drug were higher in EHBR than in SDR, indicating a greater CL_tot in SDR than in EHBR for both HSR-903 and the R-isomer. In addition, the CL_tot of the R-isomer was approximately 2 times greater than that of HSR-903.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Plasma concentration-time profiles for HSR-903, its R-isomer, and their glucuronides after a single iv bolus administration to SDR () and EHBR (), at a dose of 5 mg/kg. Each point and vertical bar represent the mean ± SE from three experiments.

Cumulative Biliary and Urinary Excretion of HSR-903 and Its Conjugate. The cumulative amounts of HSR-903 and its glucuronide excreted into bile up to 4 hr after iv bolus administration of HSR-903 to EHBR and SDR are shown in fig. 3, a and b. The cumulative biliary excretion of HSR-903 glucuronide, expressed as a percentage of the dose, in EHBR 4 hr after dosing (0.36 ± 0.03%, mean ± SE, N = 3) was much lower than that in SDR (13.52 ± 0.02%, N = 4), and the cumulative biliary excretion of intact HSR-903 was also lower in EHBR (1.85 ± 0.15%, N = 3) than in SDR (3.32 ± 0.36%, N = 4). The CL_bile values of HSR-903 and its glucuronide in EHBR were decreased to approximately 40 and 2% of those in SDR, respectively (table 2). In contrast, the CL_urine values of HSR-903 and its glucuronide were comparable for SDR and EHBR.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Biliary and urinary excretion profiles for HSR-903, its R-isomer, and their glucuronides after a single iv bolus administration to SDR and EHBR, at a dose of 5 mg/kg. Symbols show the biliary (circles) and urinary (triangles) excretion by SDR (closed symbols) and EHBR (open symbols). Each point and vertical bar represent the mean ± SE from three experiments.

Comparison of Biliary and Urinary Excretion of HSR-903 (S-Isomer) and Its R-Isomer. The cumulative amounts of the R-isomer and its glucuronide excreted into bile up to 4 hr after iv bolus administration of the R-isomer to both SDR and EHBR are shown in fig. 3, c and d. As with HSR-903, the cumulative amount of the R-glucuronide excreted in bile in EHBR (0.40 ± 0.12%, N = 3) was much lower than that in SDR (37.21 ± 3.04%, N = 4), and that of the intact R-isomer was decreased significantly in EHBR (1.30 ± 0.23%, N = 3), compared with that in SDR (5.78 ± 0.84%, N = 4). The CL_bile values of the R-isomer and its glucuronide in EHBR were decreased to approximately 15 and 1% of those in SDR, respectively (table 2). The CL_urine values of HSR-903 and the R-isomer were not significantly different, whereas the CL_bile value of the R-isomer was approximately 3 times greater than that of HSR-903.

Integration Plot Analysis. The in vivo hepatic uptake of [¹⁴C]HSR-903 was evaluated by integration plot analysis (Kim et al., 1988; Liu et al., 1992). As shown in fig. 4, the plot was a straight line, with the slope corresponding to the early-phase uptake clearance CL_uptake. The slope for [¹⁴C]HSR-903 was significantly larger than that for [³H]inulin, an extracellular marker (fig. 4). The CL_uptake of [¹⁴C]HSR-903 was estimated to be 32 ml/min/kg. Based on the CL_uptake value obtained in vivo, the hepatic blood flow rate (58.8 ml/min/kg), the blood/plasma concentration ratio (1.3), and the plasma unbound fraction (0.45), CL_int,liver was estimated to be 102 ml/min/kg, i.e. greater than the hepatic blood flow rate.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Integration plot of hepatic uptake of [14C]HSR-903 and [3H]inulin after a single iv bolus administration of [14C]HSR-903 at a dose of 5 mg/kg. , Uptake of [14C]HSR-903; , uptake of [3H]inulin.

Time Course of HSR-903 Uptake by Isolated Rat Hepatocytes. Fig. 5 shows the time courses of the uptake of [¹⁴C]HSR-903 by isolated hepatocytes from SDR and EHBR. The uptake of [¹⁴C]HSR-903 at 37°C increased linearly until 15 sec in both strains of rats. The apparent uptake of [¹⁴C]HSR-903 at 5 min was 21.5 µl/mg of protein, and the drug was accumulated approximately 21-fold against the concentration gradient, as calculated using the cell volume of 1.03 µl/mg of protein obtained in the present study as the [³H]H₂O space. Furthermore, the uptake of [¹⁴C]HSR-903 showed a marked temperature dependence (fig. 5). The initial uptake was similar between SDR and EHBR, whereas uptake by hepatocytes from EHBR was greater than that by cells from SDR under steady-state conditions.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Time course of [14C]HSR-903 (10 µM) uptake by isolated hepatocytes from SDR () or EHBR () at 37°C and from SDR at 4°C (). Each point and vertical bar represent the mean ± SE from three to six experiments. Inset, initial values on an expanded scale.

Concentration Dependence of HSR-903 Uptake by Isolated Hepatocytes from SDR. The concentration dependence of the hepatic uptake of HSR-903 was determined in the presence of increasing concentrations of unlabeled HSR-903. As shown in fig. 6, HSR-903 exhibited saturable uptake with the following kinetic parameters: K_t = 2.26 mM, J_max = 24.9 nmol/min/mg of protein, and k_d = 7.71 µl/mg of protein/min. Using eqs. 2 and 3 and the parameters obtained in in vitro experiments, the CL_int,liver value was calculated to be 165 ml/min/kg, i.e. 3 times greater than the blood flow rate.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Concentration-dependent uptake of HSR-903 by isolated rat hepatocytes. Each symbol and veritical bar represent the mean ± SE from three experiments. , least-squares fit of the data to eq. 1; - - -, estimated nonsaturable uptake; - - - -, theoretical curve for the saturable uptake process. Inset, Eadie-Hofstee plot.

Effects of Various Compounds and Sodium Ion on HSR-903 Uptake by Isolated Hepatocytes from SDR. The effects of metabolic inhibitors, organic cations, anions, and sodium ion were studied. The addition of a metabolic inhibitor (sodium azide or 2,4-dinitrophenol) did not inhibit the uptake of [¹⁴C]HSR-903. Replacement of sodium by potassium had no effect on the uptake of [¹⁴C]HSR-903 (table 3).

Effects of Quinolones on HSR-903 Uptake by Isolated Rat Hepatocytes. To determine the structural specificity of the hepatic uptake of HSR-903, the effect of several quinolone antibacterial agents was examined (table 4). Unlabeled HSR-903 and its R-isomer inhibited the uptake of [¹⁴C]HSR-903 in a concentration-dependent and stereospecific manner. In addition, grepafloxacin and sparfloxacin reduced the uptake of [¹⁴C]HSR-903, whereas lomefloxacin was not inhibitory (table 4).

Comparison of SDR and EHBR with Respect to Uptake and Efflux of [¹⁴C]HSR-903. To clarify the difference between SDR and EHBR in biliary excretion of HSR-903, in vitro uptake and efflux of HSR-903 were examined. As shown in fig. 5, the initial rates of uptake of [¹⁴C]HSR-903 into hepatocytes from SDR (18.7 nmol/min/mg of protein) and EHBR (17.1 nmol/min/mg of protein) showed no significant difference. However, decreased efflux of [¹⁴C]HSR-903 was observed in hepatocytes from EHBR, compared with those from SDR (fig. 7). In SDR the efflux was rapid and the amount of [¹⁴C]HSR-903 remaining fell to 46% at 5 min, whereas in EHBR the efflux was slow and the amount of [¹⁴C]HSR-903 remaining was 43% at 30 min.

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Time courses of [14C]HSR-903 (10 µM) efflux by isolated hepatocytes from SDR () or EHBR (). Each symbol and vertical bar represent the mean ± SE from three to six experiments.

Inhibitory Effects on HSR-903 Efflux by Isolated Rat Hepatocytes. The effects of a metabolic inhibitor (sodium azide) and an organic anion (BSP) on the efflux of [¹⁴C]HSR-903 were studied (data not shown). The addition of 10 mM sodium azide significantly reduced the initial rate of efflux of [¹⁴C]HSR-903 from hepatocytes (84.2 ± 7.6% of control, mean ± SE, N = 3). BSP also inhibited the initial efflux rate of [¹⁴C]HSR-903 (56.7 ± 2.4% of control).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Many quinolone antibacterial agents have been reported to be excreted mainly in urine. However, recently developed derivatives such as grepafloxacin (Akiyama et al., 1995; Sasabe et al., 1997), sparfloxacin (Matsunaga et al., 1991), and HSR-903 (Takagi T and Takahara E, unpublished observations), which have potent antibacterial activities and fewer side effects, are predominantly excreted in bile. Moreover, after comprehensive studies of the hepatobiliary transport mechanism for grepafloxacin, we speculated that grepafloxacin and its glucuronides might be excreted into bile by cMOAT (Sasabe et al., 1998a,b). These findings prompted us to examine the biliary excretion mechanism for HSR-903 in detail.

HSR-903 has two optical isomers, the S-isomer (HSR-903) and the R-isomer; the former has approximately 10 times greater antibacterial activity than the latter (Takahashi Y and Okezaki E, unpublished observations). Accordingly, we mainly focused on the hepatic disposition of HSR-903 and compared it with that of the R-isomer. There was no significant difference in rat plasma binding between the two isomers (table 1). The pharmacokinetic parameters determined from the plasma concentration-time profiles of the S-isomer, the R-isomer, and their glucuronides after iv bolus administration in both SDR and EHBR revealed significant differences in CL_tot values between SDR and EHBR and between the enantiomers, whereas no clear differences in volume of distribution values were observed (table 1). These results indicate that the pharmacokinetic differences between HSR-903 and its stereoisomer and between SDR and EHBR are not the result of tissue distribution but are primarily the result of elimination clearance. Although clearance of the R-isomer in SDR that is greater than the hepatic blood flow may be ascribed to extrahepatic glucuronidation (e.g. in intestinal tissue) and renal excretion, more study on the pharmacokinetics of the R-isomer is required.

The cumulative amount of HSR-903 glucuronide excreted into bile in EHBR was much less than that in SDR, resulting in the dramatically reduced CL_bile values for HSR-903 and its conjugate in EHBR, without any change in the CL_urine values. Such reduced biliary excretion in EHBR has also been observed for a nonconjugated organic anion (dibromosulfophthalein), glutathione conjugates, leukotriene C₄, BSP-glutathione, and glucuronides of bilirubin, liquiritigenin, and E-3040 (a dual inhibitor of 5-lipoxygenase and thromboxane A₂) (Huber et al., 1987; Jansen et al., 1987, 1993; Kobayashi et al., 1991; Nishida et al., 1992; Shimamura et al., 1994; Sathirakul et al., 1993; Takenaka et al., 1995a,b).

Stereospecificity is good evidence for the participation of a specialized transport mechanism, so the CL_bile values of the S- and R-isomers were compared (table 2, fig. 3). The CL_bile value of the intact R-isomer was approximately 3 times greater than that of HSR-903. Furthermore, the CL_bile value of the conjugated R-isomer was approximately 3 times greater than that of conjugated HSR-903, with the ratio between the stereoisomers being similar to that for the intact drugs, in SDR but not in EHBR. These differences between the CL_bile values of the stereoisomers of the intact forms and their metabolites suggest that cMOAT, which is defective in EHBR (Ito et al., 1997), is stereospecific in discriminating between HSR-903 and its stereoisomer. The processes of hepatic uptake of the intact forms and the glucuronidation steps in hepatocytes are also possibly stereospecific. Although the use of intact conjugates of the two isomers would be helpful in clarifying the hepatic handling of the conjugates, we could not perform these experiment because the conjugates are not available as reagents.

To identify the rate-limiting process in the biliary excretion of HSR-903, the hepatic uptake rate was determined both in vivo and in vitro. The uptake of [¹⁴C]HSR-903 in vivo was evaluated by integration plot analysis (fig. 4). The estimated CL_int,liver of 102 ml/min/kg for [¹⁴C]HSR-903 was close to the value of PS_u,influx (165 ml/min/kg) calculated from the in vitro hepatic uptake study, being greater than the blood flow rate. The in vivo and in vitro results both suggest that the hepatic uptake of [¹⁴C]HSR-903 is very fast and that the uptake is blood flow-limited.

To elucidate the mechanism of such rapid hepatic uptake of HSR-903, isolated rat hepatocytes were used. The steady-state uptake of [¹⁴C]HSR-903 exhibited an apparent 21-fold accumulation against the concentration gradient. This apparent accumulation may include metabolites of HSR-903 as well as unchanged HSR-903 at steady state and even in the early stages of uptake. Accordingly, analysis of intracellular radiolabeled compounds by HPLC or other methods is needed, to clarify the extent of accumulation of unchanged HSR-903. However, approximately 10-fold concentrative accumulation of [¹⁴C]HSR-903 was observed within a few minutes. The uptake of HSR-903 showed marked temperature dependence (fig. 5) but was not inhibited by the substitution of sodium by potassium or by the addition of a metabolic inhibitor (sodium azide). These results suggest that a concentrative uptake mechanism is involved in the entry of HSR-903 into hepatocytes, although the driving force is presently unclear. The initial uptake rates of HSR-903 were comparable between SDR and EHBR. However, at steady state HSR-903 was accumulated to a greater extent in hepatocytes from EHBR. This may be ascribed to decreased efflux in EHBR-derived hepatocytes, because of the deficiency of the bile cMOAT.

The structural specificity of the hepatic uptake of HSR-903 was examined by measuring the effects of several quinolone antibacterial agents on [¹⁴C]HSR-903 uptake (table 4). Unlabeled HSR-903 and its R-isomer inhibited the uptake of [¹⁴C]HSR-903 in a concentration-dependent manner, but there was no significant difference in inhibitory effects between the stereoisomers. In addition, grepafloxacin and sparfloxacin reduced the uptake of [¹⁴C]HSR-903, and lomefloxacin had a slight but significant inhibitory effect (table 4). The octanol/water partition coefficients of HSR-903, grepafloxacin, and sparfloxacin are 2.58, 5.91, and 1.14, respectively (Fukuoka H and Yamazaki M, unpublished observations), being greater than that of lomefloxacin (<0.6). These findings indicate that several quinolone antibacterial agents, including HSR-903, are taken up by liver cells via a common transporter and that their lipophilicity may contribute to the affinity for the transporter.

The addition of a metabolic inhibitor and the replacement of sodium with potassium had no effect on the uptake of [¹⁴C]HSR-903. Furthermore, organic anions, an organic cation, a bile acid, and amino acids had no effect on [¹⁴C]HSR-903 uptake. Therefore, this transporter in the sinusoidal membrane seems to be specific for certain quinolones. Sasabe et al. (1997) reported Na⁺-independent and carrier-mediated active uptake of grepafloxacin by isolated rat hepatocytes. None of the known transporters for bile acids, organic anions, organic cations, or ouabain is responsible for the uptake of grepafloxacin (Sasabe et al., 1997, 1998a), the transport mechanism for which seems to be very similar to that for HSR-903.

In vitro efflux of [¹⁴C]HSR-903 from hepatocytes was different in SDR and EHBR in the later stage (fig. 7), strongly indicating that the efflux process across the canalicular membrane from the intracellular space is rate-limiting in the biliary excretion of HSR-903. Furthermore, a metabolic inhibitor (sodium azide) and an organic anion (BSP) reduced the efflux of [¹⁴C]HSR-903. Therefore, the efflux is considered to be ATP-dependent and sensitive to anions, suggesting that cMOAT, which is functionally deficient in EHBR, is responsible for the efflux of HSR-903 and/or its glucuronide.

In conclusion, this comprehensive study of the hepatobiliary transport of HSR-903 in vivo and in vitro demonstrated that HSR-903 is efficiently taken up from blood by hepatic parenchymal cells via an Na⁺-independent, carrier-mediated, active transport system, which plays a key role in the effective hepatic uptake of HSR-903. Furthermore, HSR-903 is actively excreted into bile from the cells via cMOAT. Both the uptake and efflux mechanisms in hepatocytes are common for HSR-903 and certain other quinolones.