Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Primary active transporters in the liver canalicular membrane, which contain the ATP binding cassette (ABC¹), play an important role as efflux pumps in the excretion of endogenous bile constituents or xenobiotics into the bile canaliculi (Hooiveld et al., 2001). As might be expected, certain types of liver diseases have an influence on this hepatobiliary excretion. Experimental hepatic injury (EHI) induced by a single administration of carbon tetrachloride (CCl₄) has been widely used as a pathological model for liver diseases, since it is known that CCl₄ produces acute hepatocellular injury with centrilobular necrosis and steatosis (Recknagel, 1967). The effects of CCl₄-EHI on biochemical characteristics such as increased lipid peroxidation (Recknagel, 1967) and the activities (Mourelle et al., 1987; Romero et al., 1994) of glutamic oxaloacetic transaminase (GOT) and glutamic pyruvic transaminase (GPT) and hepatobiliary excretion of xenobiotics (for example, sulfobromophthalein and leukotriene; Iga et al., 1977; Wettstern et al., 1990) have been widely studied. Our earlier study revealed that transport systems for some organic cations on the sinusoidal membrane are prone to damage as the result of CCl₄-EHI, as demonstrated by a decrease in the in vitro maximal rates of hepatic uptake and efflux without any influence on in vivo canalicular excretion clearance (Hong et al., 2000). In addition, a dose-dependent inhibition of Na⁺/taurocholate cotransport (Ntcp) into sinusoidal membrane vesicles by the presence of trichloroethylene and 1,1,2-trichloro-1,2,2-trifluoroethane (all CCl₄ analogs) has been reported (Neghab et al., 1996). Recently, Geier et al. (2002) reported that hepatobiliary organic anion transporters on the sinusoidal membrane were regulated differently in acute toxic liver injury induced by CCl₄; mRNA levels were significantly decreased for Ntcp, Oatp1 (organic anion transporting polypeptide1), and Oatp2, whereas the level remained unchanged for Oatp4.

Contrary to the cases of transport systems on the sinusoidal membranes, considerably less information is available on the effects of CCl₄-EHI on the canalicular transport system, which is believed to be a key step in the vectorial transport of various xenobiotics from the portal blood to the bile (Bossard et al., 1993; Han et al., 1999). Thus, the objective of this study was to examine the effect of CCl₄-EHI on the expression and functional activity of representative ABC transporters on the canalicular membrane, including P-glycoprotein (P-gp), a bile salt export pump (Bsep), and a multidrug resistance associated protein2 (Mrp2). Western blot analysis was performed to evaluate the expression of the transporters, and in vivo canalicular excretion and in vitro uptake into canalicular liver plasma membrane (cLPM) vesicles for daunomycin, taurocholate, and 17-estradiol-17-D-glucuronide (E₂17G) were measured to estimate the functional activity of the relevant transporters. These compounds were selected because they represent substrates for P-gp (Kamimoto et al., 1989; Hooiveld et al., 2001), Bsep (Gerloff et al., 1998; Hooiveld et al., 2001), and Mrp2 (Morikawa et al., 2000; Hooiveld et al., 2001), respectively, and are extensively excreted into bile via relevant transporters in rats [i.e., approximately 36% of an i.v. dose (10 µmole/kg) for daunomycin (Catapano et al., 1988), 87% of an i.v. dose (54 µmole/kg) for taurocholate (Ring et al., 1994), and 77% of an i.v. dose (180 pmol/kg) for E₂17G (Morikawa et al., 2000)].

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. [³H]Taurocholate (2 Ci/mmol), [³H]17-estradiol-17-D-glucuronide (E₂17G, 44 Ci/mmol) and [³H]daunomycin (4.4 Ci/mmol) were purchased from PerkinElmer Life Science Inc. (Boston, MA). C219 and M₂ III-6, monoclonal antibodies to P-gp and Mrp2, respectively, were purchased from Alexis Biochemical Co. (San Diego, CA), and Antispgp, a polyclonal antibody to Bsep, was purchased from Kamiya Biomedical Co. (Seattle, WA). An enhanced chemiluminescence detection system was purchased from Amersham Biosciences Inc. (Piscataway, NJ). Sucrose was purchased from Junsei Chemical Co. (Tokyo, Japan). All other chemicals including reagents for Western blot analysis and the vesicle uptake study were purchased from Sigma-Aldrich (St. Louis, MO).

Induction of Experimental Hepatic Injury by CCl₄. Male Sprague Dawley rats (250-300 g; Dae-Han Biolink, Taejon, Korea) were injected intraperitoneally with a single dose of CCl₄ (1 ml/kg) as a 50% (v/v) solution in olive oil and then fasted for 6, 12, 24, 36, or 48 h. Water was fed ad libitum. Control animals received a corresponding dose of olive oil using the same experimental protocol. The activities of sGPT and sGOT were measured (Reitman and Frankerl, 1957) using a commercial colorimetric determination kit (Yeoung Dong Pharm. Co., Seoul, Korea). Experimental protocols involving animal studies were reviewed by the Animal Care and Use Committee in College of Pharmacy, Seoul National University according to the National Institutes of Health guidelines (National Institutes of Health publication number 86-23, revised 1985) of "Guide for the Care and Use of Laboratory Animals".

Preparation of cLPM Vesicles. cLPM vesicles were prepared from four normal and CCl₄-EHI rats according to the method of Inoue et al. (1983) as described previously (Song et al., 1999). The purity of the vesicle preparations was routinely assessed by measuring the relative enrichment of the activity of alkaline phosphatase (ALP, Pekarthy et al., 1972), a marker enzyme for the canalicular membrane, compared with that of crude homogenates. The protein concentration of the vesicle preparation was measured by the Lowry method (Lowry et al., 1951) using bovine serum albumin as a standard. To estimate the relative proportion of the inside-out vesicles, the concentrations of the exposed sialic acid of the vesicles were determined (Warren, 1959). Immediately after their preparation, the cLPM vesicles were suspended in a membrane suspension buffer (MSB), which contained 250 mM sucrose, 10 mM Hepes, 10 mM Tris, 10 mM MgCl₂, and 0.2 mM CaCl₂ (pH 7.4), to yield a protein concentration of 8 to 10 mg/ml. The suspension was stored at 70 ^cC, for up to 2 weeks, until the uptake studies and Western blot analysis were carried out.

Western Blot Analysis. The vesicle preparations were diluted in a buffer consisting of 2% sodium dodecyl sulfate (SDS), 10% glycerol, 5% 2-mercaptoethanol, and 50 mM Tris-HCl (pH 6.8) to yield a total protein concentration of 1.25 µg/µl. A 16-µl aliquot of the suspension (equivalent to 20 µg of total protein) and 10 µl of a solution of molecular weight markers (High Mw-SDS calibration kit, Amersham Biosciences Inc.) were subjected to electrophoresis on a 6% polyacrylamide gel with 0.1% SDS, and electrotransferred to a nitrocellulose membrane (0.45 µm pore size; Amersham Biosciences Inc.). The membrane was blocked with a phosphate-buffered saline solution (pH 7.4) containing 0.1 (w/v) % Tween 20 (PBST) and 5% (w/v) nonfat dry milk (Seoul Milk Co., Seoul, Korea) for 1 h at 37°C, and was probed for 18 h at 4°C with the primary antibodies (dilution 1: 1000 each) C219 (Lee et al., 1996), M₂ III-6 (Kool et al., 1997), and Antispgp (Lecureur et al., 2000) that recognize P-gp, Mrp2, and Bsep, respectively. After washing three times with 60 ml of PBST, the membrane was incubated with secondary antibodies [i.e., horseradish peroxidase-conjugated goat anti-mouse IgG for C219 and M₂ III-6 (dilution 1:1000; Zymed Laboratories, South San Francisco, CA), and horseradish peroxidase-conjugated goat anti-rabbit IgG for Antispgp (dilution 1:1000; Amersham)] for 1 h at room temperature. After washing three times with 60 ml of PBST, it was possible to visualize all proteins recognized by the antibodies using an enhanced chemiluminescence detection system. Band intensities were analyzed by densitometry using a software program, Quantity One (version 4.1; Bio-Rad, Hercules, CA).

In Vivo Biliary Excretion across Canalicular Membranes. After light anesthesia with ketamine (50 mg/kg as i.p. dose; Yuhan Pharm. Co., Kyounggido, Korea), the femoral arteries and veins of rats were cannulated with PE-50 polyethylene tube and bile ducts with PE-10 polyethylene tube. Normal rats and CCl₄-pretreated (for 24 h) rats (i.e., CCl₄-EHI_{24 h} rats) were used throughout the in vivo excretion experiments. For the estimation of canalicular excretion clearance (CL_exc) of daunomycin, [³H]taurocholate and [³H]E₂17G, rats received i.v. injection at bolus doses of 10 µmole/kg, 400 pmol/kg, and 160 pmol/kg, followed by i.v. infusion at rates of 10 µmole/h/kg, 400 pmol/h/kg (12 µCi), and 160 pmol/h/kg (48 µCi), for respective compounds. Blood samples (0.3 ml) were taken from the femoral artery at 30-min intervals over a 3-h period, and bile was collected for 30-min period up to 3 h. Plasma samples were separated from the blood samples by centrifuging at 3,000 rpm for 10 min. At 3 h (i.e., at the steady state) after each infusion start, rats were sacrificed and the liver was isolated immediately. A 20% liver homogenate was then prepared using normal saline, centrifuged at 3,000 rpm for 10 min, and aliquots (100 µl) of the supernatant were collected for the determination of the respective compounds. The in vivo CL_exc of a compound was calculated by dividing the biliary excretion rate by the liver substrate concentration at the steady state.

Assay of Taurocholate, Daunomycin, and E₂17G in In Vivo Study. The concentrations of taurocholate in plasma, liver (homogenates), and bile samples were quantified by liquid scintillation counting (LSC, Wallac 1409; PerkinElmer Life Science Inc.). The concentration of daunomycin, which is known to be metabolized in the body to daunomycinol (Pea et al., 2000), was determined by high performance liquid chromatography (HPLC). A 100-µl aliquot of the biological samples (i.e., plasma, bile, and supernatant of the liver homogenate) was deproteinized by the addition of MeOH (250 µl) containing Adriamycin (internal standard, 1 µM). Ethyl acetate (1 ml) was then added, the suspension was vigorously mixed for 5 min and then centrifuged at 10,000 rpm for 5 min. An aliquot (1.2 ml) of the supernatant was transferred and evaporated using a Spinvac (Hanil Science, Seoul, Korea), and the residue was reconstituted with the mobile phase (150 µl) used for HPLC (acetonitrile: 0.01 M phosphoric acid = 3: 7 v/v, pH 3.0). A 50-µl aliquot of the reconstituted solution was injected into the HPLC system, which consisted of a Hitachi L-7110 pump (Hitachi, Japan), a Shimadzu RF 535 fluorescence detector (Shimadzu, Japan), a Hitachi L-7200 autosampler, a Hitachi D-7500 integrator, and a C₁₈ column (Shisheido Capcell pak, 4.6 × 250 mm, 5-µm particle size). The flow rate of the mobile phase was set at 1 ml/min. The chromatogram was monitored by fluorescence detection at an excitation wavelength of 470 nm and an emission wavelength of 565 nm. The eluent resulted in sharp and well resolved peaks corresponding to daunomycin, Adriamycin (internal standard), and possible metabolites. The retention times of Adriamycin and daunomycin were 4.9 and 8.3 min, respectively, under the conditions used. Calibration curves of daunomycin were linear over the concentration range of 0.1-4 µM for plasma, bile, and liver samples, with respective correlation coefficients of over 0.999.

In Vitro Vesicle Uptake Studies. The uptake of [³H]daunomycin, [³H]taurocholate, and [³H]E₂17G into cLPM vesicles was measured by a rapid filtration technique (Song et al., 1999) using vesicles prepared from normal and CCl₄-EHI_{24 h} rats. The uptake of [³H]daunomycin was measured as follows. A frozen vesicle suspension was quickly thawed by immersion in a 37 °C water bath, re-vesicularized by passing it through a 25 gauge needle 20 times, and appropriately diluted with MSB to give 3 to 4 mg/ml of protein. Ten microliters of the diluted suspension was preincubated in a test tube at 37 ^cC for 4 min, and 40 µl of MSB, which contained 0.2 µM daunomycin (0.035 µCi) with or without the ATP-regenerating system (1.2 mM ATP, 3 mM phosphocreatine, and 3.6 µg/100 µl creatine phosphokinase), was then added to the diluted vesicle suspension. At predetermined times, the uptake was quenched by the addition of 1 ml of an ice-cold solution of MSB containing 20 µM daunomycin. The entire sample was then rapidly filtered through a MF-MEMB filter (0.45-µm pore size, 25-mm diameter; Seoul Science, Seoul, Korea), which had been presoaked for 2 h in ice-cold MSB. The tube was rinsed again with 1 ml of ice-cold MSB and then filtered. After washing with 10 ml of ice-cold MSB, the filter was dissolved in 4 ml of scintillation cocktail (Ultimagold; PerkinElmer Life Science Inc.), and the radioactivity of the mixture was determined by LSC. Presoaking and rinsing the filter with the ice-cold MSB, which contains 20 µM daunomycin, resulted in a very small level of nonspecific binding of daunomycin to the filter (i.e., negligible radioactivity in the filter, data not shown). The amount of daunomycin in the vesicles (expressed as picomoles per milligrams of protein) was plotted against time. The ATP-dependent fraction that was taken up was estimated by the difference in uptake between the two conditions (i.e., in the presence and absence of ATP). The initial rate of uptake of daunomycin (i.e., ATP-dependent or -independent) was obtained from the linear portion (generally up to 1 min) of the temporal uptake profile. The concentration dependence of the ATP-dependent initial uptake rate of daunomycin was examined for the concentration range of 5 to 500 µM.

Data Analysis. All data are expressed in the form of the mean ± S.D. A two-way analysis of variance was performed to test differences in the temporal uptake of each substrate between the transport conditions and treatments (i.e., normal and CCl₄-EHI). The student's t test was used to test differences in the mean kinetic parameters (i.e., for in vitro and in vivo experiments) between the treatments. In all cases, P < 0.01 was accepted as denoting a statistical difference.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Pathophysiological and Biochemical Changes by CCl₄-EHI. The effects of CCl₄-EHI_{24 h} (i.e., EHI at 24 h after the CCl₄ administration) on various pathophysiological parameters are summarized in Table 1. While the body weights were similar in both rats, the liver weight was increased in CCl₄-EHI_{24 h} rats by 15%. EHI was confirmed by 11 and 14-fold increases in the activities of sGOT and sGPT, respectively. The yield of protein from liver homogenates was decreased by 19% as a result of the CCl₄-EHI_{24 h}, which is consistent with the previously reported decrease in protein synthesis (Romero et al., 1994). The ALP activity of the homogenate was unaffected by the CCl₄-EHI_{24 h}.

Western Blot Analysis. The expression of ABC transporters that contain ATP binding cassettes (ABC) on canalicular membranes was measured for cLPM vesicles by a Western blot analysis (Fig. 1). Representative immunoblots for P-gp in cLPM vesicles, prepared from livers of normal and CCl₄-EHI rats, are shown in Fig. 1A. Bands of approximately 170 kDa, the molecular size of P-gp (Kamimoto et al., 1989), was observed for all the vesicle samples. The intensity of the band varied with time (6, 12, 24, 36, and 48 h) after the CCl₄ administration. Densitometric analysis (Fig. 1, A and A') shows a significant decrease in P-gp expression at 6 h after CCl₄ administration, compared with cLPM vesicles from the normal liver (i.e., control). The expression levels at 12 h, however, exhibited a marked increase (i.e., 2.1-fold of the control value). A maximal expression (i.e., 3.6-fold increase) was observed at 24 h after the CCl₄ administration. In 36 h, the expression decreased to levels similar to those of the control. Representative immunoblots for Bsep in cLPM vesicles are shown in Fig. 1B. Bands of approximately 150 kDa, the molecular size of Bsep (Gerloff et al., 1998), were observed for all the vesicle samples. Contrary to the case for P-gp, no change in the expression of Bsep was observed by the CCl₄ treatment up to 48 h (Fig. 1, B and B'). The strongest band of approximately 190 kDa (Mottino et al., 2000) was observed for Mrp2 (Fig. 1C). The density of the band decreased as a function of time after CCl₄ administration, exhibiting a maximal decrease to 27% of control at 24 h and partial recovery to 60% in 48 h (Fig. 1, C and C').

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Representative immunoblots of P-gp (A), Bsep (B), and Mrp2 (C) proteins, and the relative densities of immunostained bands of P-gp (A'), Bsep (B'), and Mrp2 (C') in cLPM vesicles from normal and CCl4-EHI rats. cLPM vesicles were prepared from pooled liver homogenates from normal (n = 4) and CCl4-EHI rats (n = 4) at 6, 12, 24, 36, and 48 h after an intraperitoneal injection of CCl4 (1 ml/kg). Three different batches of vesicles were prepared using different liver homogenate pools (i.e., the number of rats at each time point = 12, the total number of rats used = 72). Lanes were loaded with each membrane vesicle preparation (20 µg of total protein). The right lane was loaded with a high molecular weight size marker. The target protein was detected using antibodies such as C219, M2III-6, and Antispgp that recognize P-gp, Mrp2, and Bsep, respectively. Each data point in the right column is expressed as the mean ± S.D. for three different batches of cLPM vesicle preparations.

In Vivo Canalicular Excretion Clearance. To address the issue of whether the altered expression of ABC transporters (Fig. 1) is reflected on the in vivo functional activity of the transporters, the effect of CCl₄-EHI on the canalicular excretion of representative substrates of the transporters was examined. Since the rats exhibited maximal change in the expression of P-gp and Mrp2 at 24 h after the CCl₄ pretreatment (Fig. 1), subsequent in vivo experiments were performed using rats at 24 h after the pretreatment. An i.v. infusion of taurocholate and E₂17G was performed at tracer doses to avoid their possible interactions with endogenous taurocholate and E₂17G (Morikawa et al., 2000) in canalicular excretion. As the result, the plasma concentration of these compounds remained in the nanomolar range (Table 3). The well known cholestatic effect of E₂17G (Meyers et al., 1980; Morikawa et al., 2000) was not apparent at this plasma level, demonstrated by comparable bile flow rates among normal rats [i.e., 65.5 ± 11 µl/min/kg (mean ± S.D., n = 6) for daunomycin, 55.5 ± 9.5 µl/min/kg (mean ± S.D., n = 6) for taurocholate, 52.2 ± 14 µl/min/kg (mean ± S.D., n = 6) for E₂17G, and 58.4 ± 6.7 µl/min/kg (mean ± S.D., n = 3) for control]. Moreover, bile flow was not influenced by the CCl₄-EHI_{24 h} regardless of the compounds infused (data not shown).

View this table: [in this window] [in a new window]   TABLE 3 Summary of the Effects of CCl4 on the Pharmacokinetic Parameters of daunomycin, taurocholate, and E217 G at steady-state conditionsa.

View larger version (24K): [in this window] [in a new window]   Fig. 2.   Plasma concentrations and biliary excretion rates of daunomycin (A and A', respectively), taurocholate (B, B') and E217G (C, C') in normal () and CCl4-EHI24 h () rats following i.v. bolus dose and i.v. infusion. Doses of daunomycin, taurocholate, and E217G were 10 µmole/kg and 10 µmole/h/kg, 400 pmol/kg and 400 pmol/h/kg (12 µCi), and 160 pmol/kg and 160 pmol/h/kg (48 µCi) (for respective i.v. bolus doses and infusion rates), respectively. Each point represents the mean ± S.D. of six different rats.

Alterations in the ATP-dependent Uptake of Daunomycin, Taurocholate and E₂17G into cLPM Vesicles. Certain pathophysiological changes by EHI, such as decreased metabolic activity (Olatunde Farombi, 2000) and an increased plasma level of endogenous compounds (e.g., corticosterone; Huang et al., 2001), might interfere with the in vivo transport of relevant substrates. To exclude this possibility, the uptake of these substrates into cLPM vesicles from normal and CCl₄-EHI_{24 h} rats was examined. Figure 3A shows the result for daunomycin. The uptake of daunomycin was significantly increased by the presence of ATP. The uptake in the absence of ATP was not influenced by the CCl₄-EHI_{24 h}, whereas the uptake in the presence of ATP was increased significantly for cLPM vesicles from CCl₄-EHI_{24 h} rats. Figure 3A' shows the concentration (5-500 µM) dependence of the ATP-dependent uptake rate of daunomycin. Consistent with Fig. 3A, a much higher rate of uptake was observed by the CCl₄-EHI_{24 h}. A kinetic analysis revealed a 1.8-fold increase in the value of V_max for the uptake of daunomycin into cLPM vesicles from the CCl₄-EHI_{24 h} rats, compared with those from normal rats, whereas the values of K_m for vesicles from both rats were comparable (Table 4). As a result, a 1.9-fold increase in the value of CL_int was observed by the CCl₄-EHI_{24 h} for daunomycin (Table 4), highly consistent with the increase in CL_exc in the in vivo study (i.e., a 1.8-fold increase). The increase in the value of V_max under a constant value of K_m by the CCl₄-EHI_{24 h} appears to be consistent with the increase in the expression of P-gp by the CCl₄-EHI_{24 h} (Fig. 1, A and A'), although it is less prominent than expected from the 3.6-fold increase in the expression of P-gp (Fig. 1)

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Effects of CCl4-EHI on the ATP-dependent uptake of 0.2 µM [3H]daunomycin (A), 1 µM [3H]taurocholate (B), and 1 µM [3H]E217G (C) into cLPM vesicles from normal (, ) and CCl4-EHI rats (, ) in the presence (, ) or absence (, ) of ATP. Concentration dependence in the ATP-dependent uptake of daunomycin (A', 5-500 µM), taurocholate (B', 5-500 µM) and E217G (C', 0.5-200 µM). ATP-dependent fraction of the uptake (A', B', and C', ) was obtained by subtracting the uptake rate of a substrate in the absence of ATP from that in the presence of ATP. Curved lines were generated using the estimated kinetic parameters in Table 4. Each data point is expressed as the mean ± S.D. of triplicate measurements from three different batches of membrane vesicle preparations. , statistically different from normal cLPM vesicles (P < 0.01).

Effect of Direct Contact of CCl₄ on the Uptake of Substrates into cLPM Vesicles. To examine the issue of whether the above observed effects of CCl₄-EHI_{24 h} are attributable to direct interactions of the responsible transport systems with CCl₄, the uptake of 0.2 µM daunomycin, 1 µM taurocholate, and 1 µM E₂17G into cLPM vesicles, following the incubation of normal cLPM vesicles with 1.5 mM CCl₄ at 37°C for 15 min, was assessed in the presence and absence of ATP. The uptake of daunomycin into normal vesicles was decreased significantly by incubation in the presence of CCl₄ and ATP (Fig. 4A). This is contrary to the results obtained for the uptake of daunomycin into cLPM vesicles that were prepared from CCl₄-EHI_{24 h} rats (Fig. 3A), indicating that CCl₄, in the body, affects the ABC transporters on the canalicular membrane in an indirect manner. On the other hand, the effect of direct contact of CCl₄ on the uptake of taurocholate and E₂17G into cLPM vesicles was similar to that of CCl₄-EHI_{24 h} on the uptake of these substrates, regardless of the presence of ATP (Fig. 3, B and C). Despite this similarity, however, CCl₄-EHI seems to influence the ABC transporters on the canalicular membrane in a manner that is distinct from the direct contact of CCl₄ with the membrane, based on the result of the uptake of daunomycin.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Effect of the direct addition of 1.5 mM CCl4 on the ATP-dependent uptake of 0.2 µM [3H]daunomycin (A), 1 µM [3H]taurocholate (B), and 1 µM [3H]E217G (C) in the absence (, ) and presence (, ) of ATP. cLPM vesicles from normal rats were preincubated with (, ) or without CCl4 (, ) at 37°C for 15 min. Each point represents the mean ± S.D. of triplicate measurements from three different batches of membrane vesicle preparations. , statistically different from control (P < 0.01).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Western blot analysis (Fig. 1) indicates that the expression level of ABC transporters on the canalicular membrane, except for Bsep, varies as a function of time after the administration of CCl₄ to rats, exhibiting maximal changes at 24 h (Fig. 1). Most interestingly, the effect of CCl₄-EHI_{24 h} on the expression levels differed significantly depending on the canalicular transporters; the expression level was increased for P-gp, remained unchanged for Bsep, and was decreased for Mrp2. These results for P-gp and Bsep are consistent with previous reports (Huang et al., 2001; Geier et al., 2002), whereas the result for Mrp2 is contrary to data reported by Geier et al. (2002), in which no change in the expression level of Mrp2 was reported for the liver microsomes of CCl₄-EHI_{24 h} rats. In the present study, Western blot analysis of transporters was performed for cLPM vesicles instead of liver microsomes (Geier et al., 2002). This difference (i.e., vesicles and microsomes) might be involved in the discrepancy between the two studies in the expression level of Mrp2.

In the present study, Antispgp (Lecureur et al., 2000), M₂III-6 (Kool et al., 1997), and C219 (Lee et al., 1996) were used as antibodies for the immunostaining of Bsep, Mrp2, and P-gp, respectively. Antispgp does not cross react with the mdr1a gene product (i.e., P-gp) (Lecureur et al., 2000), and M₂III-6 does not with gene products of mdr1a and mrp2 isoforms (i.e., Mrp1, Mrp3, and Mrp5) (Kool et al., 1997). C219, the most widely used antibody for P-gp (Lee et al., 1996), on the other hand, exhibits a partial cross reactivity to other proteins such as Mdr2 and Bsep (Van Den Elsen et al., 1999). The fact that the level of Mdr2 appeared to be increased in 24 h after the CCl₄ pretreatment (from the mRNA level of mdr2; Nakasukasa et al., 1993) suggests that immunostaining using C219 might lead to an overestimation of P-gp levels. In other words, a 3.6-fold increase in the level of P-gp by the CCl₄-EHI_{24 h} (Fig. 1) might be an overestimation, at least in part.

The effect of the CCl₄-EHI_{24 h} (i.e., changes in the expression of the ABC transporters on the canalicular excretion of representative substrates of the transporters was examined in vivo. The CL_exc values for the substrates were altered multiply (Table 3), consistent with the changes in the expression level of relevant transporters (Fig. 1). However, the change in the CL_exc value for daunomycin (i.e., 1.9-fold increase; Table 3) was not as significant compared with the changes in the expression of P-gp (i.e., 3.6-fold increase; Fig. 1A'). This discrepancy could be related to the cross reactivity of C219 (Van Den Elsen et al., 1999) or to interference by certain endogenous factors in CCl₄-EHI_{24 h} rats on canalicular excretion (Olatunde Farombi, 2000; Huang et al., 2001).

To exclude possible interference by endogenous factors, the in vitro uptake of relevant substrates into cLPM vesicles was examined. Results from the in vitro uptake study were fairly consistent with the expression of relevant transporters (Fig. 1) as demonstrated by V_max and K_m values in Table 3. The CL_int was 1.9-fold increased for daunomycin, remained unchanged for taurocholate, and was 39% decreased for E₂17G. An analysis of the uptake studies revealed that the changes in the CL_int are solely attributable to changes in the V_max and independent of changes in corresponding K_m values (Table 3), suggesting that the alteration in the quantity of relevant transporters, rather than the alteration of their affinity for their substrates, is responsible for the altered CL_int of the relevant substrates. Neither the inside-out proportion nor leakage (mannitol uptake) of the vesicles (Table 2) appears to be related to the altered CL_int since they were not influenced by the CCl₄-EHI. In CCl₄-EHI rats, the composition of plasma membrane lipid is changed, possibly increasing the fluidity of the lipid bilayer (Recknagel, 1967; Mourelle et al., 1987). The increase in the membrane fluidity might accelerate the functional activity of ABC transporters (Hyogo et al., 2001). However, the contribution of membrane fluidity-associated acceleration is likely to be minimal considering the unchanged or decreased CL_int values for taurocholate and E₂17G (Table 3). The above results suggest that the altered expression of relevant transporters on the canalicular membrane is reflected on the canalicular excretion of the relevant substrates in vivo.

The effect of CCl₄-EHI_{24 h} on the expression and functional activity of relevant transporters (in vivo and in vitro) is summarized in Fig. 5. In general, the effect of CCl₄-EHI_{24 h} on the ABC transporters, with respect to the expression and functional activity, were different depending on the transporters. The alteration in the expression and functional activity of each transporter was in an identical direction for all the three transporters. However, a closer relationship was observed between CL_int (or V_max) and CL_exc, compared with the relationship between the expression of transporters and CL_int or CL_exc. This might be related to the cross reactivity of primary antibodies used in the Western blot analysis (especially for P-gp). It is also consistent with the hypothesis that interference by the endogenous factors in CCl₄-EHI_{24 h} on the relevant transport is not so profound.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Summary of the effect of the CCl4-EHI24 h on the expression of the primary active transporters on the canalicular membrane and CLint and CLexc of relevant substrates.

The above effects of the CCl₄-EHI_{24 h} appear to be distinct from the result of direct interactions of the membrane or transport systems with CCl₄, since the incubation of normal cLPM vesicles with 1.5 mM CCl₄ decreased the uptake of daunomycin to the vesicles in the presence of ATP, for example (Fig. 3), contrary to the increased uptake of the compound into the cLPM vesicles that were prepared from the CCl₄-EHI_{24 h} rats (Fig. 2A). Thus the CCl₃ · radical, which is formed from CCl₄ in the body and attacks macromolecules such as proteins and lipid bilayers (Recknagel, 1967; Nelson, 1995), appears to represent the mechanism involved in the indirect effect of the CCl₄ pretreatment.

Although the expression of transporters was changed by the CCl₄-EHI (Fig. 1), the expression does not appear to be necessarily parallel with the level of relevant mRNAs, since the level of mdr 1a and mdr 1b transcripts was reported to be continuously increased for up to 5 days after the CCl₄ administration (Nakasukasa et al., 1993), whereas the expression of the relevant transporter, P-gp, exhibited a maximal expression at 24 h after the CCl₄ administration in the present study (Fig. 1, A and A'). Thus, post-transcriptional regulation appears to influence, at least in part, the level of transporters on the canalicular membrane. Similar discrepancies in the expression of transporters and responsible mRNAs have been reported for Bsep and Mrp2 in ethynylestradiol-induced cholestasis rats (Trauner et al., 1997; Lee et al., 2000).

One of the most important findings of the present study is the multiple alterations in the expression and activities of canalicular membrane transporters by the CCl₄-EHI_{24 h} (Fig. 5). A multiple effect of CCl₄ has also been demonstrated for the expression of mRNAs and proteins of the sinusoidal transporters (i.e., Ntcp, Oatp1, Oatp2, and Oatp4, Geier et al., 2002) as described in the Introduction. The physiological meaning of this type of influence of CCl₄-EHI, especially as host defense mechanisms of biological systems against diseases and xenobiotics, should be carefully considered. Regardless of the underlying mechanisms, the multiple (i.e., transporter and membrane selective) effects of the CCl₄-EHI should be taken into account in understanding of hepatic injuries. Special care should be taken concerning the increase in the expression and functional activity of certain transporters (i.e., P-gp, in this study), because it might be contrary to our general understanding that hepatic injuries may damage membrane transporters both in terms of expression and function. In addition, it is noteworthy that the expression of ABC transporters on the canalicular membrane may vary depending on the types of EHI, as evidenced by the decrease in the level of both Bsep and Mrp2 by ethynylestradiol-induced EHI (Bossard et al., 1993; Lee et al., 2000).