Genipin enhances Mrp2 (Abcc2)-mediated bile formation and organic anion transport in rat liver

Authors


Abstract

Inchin-ko-to (ICKT), an herbal medicine, and its ingredients exert potent choleretic effects by a “bile acid-independent” mechanism. The current study was designed to determine whether ICKT or its ingredients potentiate multidrug resistance-associated protein 2 (Mrp2; Abcc2)-mediated choleresis in vivo. Biliary secretion of Mrp2 substrates and the protein mass, subcellular localization, and messenger RNA (mRNA) level of Mrp2 were assessed in rat liver after infusion of genipin, an intestinal bacterial metabolite of geniposide, a major ingredient of ICKT. The function of Mrp2 was also assessed by the adenosine triphosphate (ATP)-dependent uptake of Mrp2-specific substrates using canalicular membrane vesicles (CMVs) from the liver. Infusion of genipin increased bile flow by 230%. It also increased biliary secretion of bilirubin conjugates and reduced glutathione (GSH) by 513% and 336%, respectively, but did not increase bile acid secretion. The ATP-dependent uptake of estradiol 17-β-D-glucuronide (E217βG; by 265%), leukotriene C4 (LTC4; by 161%), taurolithocholate-3-sulfate (TLC-3S; by 266%), and methotrexate (MTX; by 234%) was significantly stimulated in the CMVs from the liver. These effects were not observed in Mrp2-deficient rats. Under these conditions, genipin treatment increased the protein mass of Mrp2 in the CMVs but not the mRNA level. In immunoelectron microscopic studies, a marked increase in Mrp2 density in the canalicular membrane (CM) and microvilli was observed in the genipin-treated liver tissue sections when compared with the vehicle-treated liver tissue sections. In conclusion, genipin may enhance the bile acid-independent secretory capacity of hepatocytes, mainly by stimulation of exocytosis and insertion of Mrp2 in the bile canaliculi. ICKT may be a potent therapeutic agent for a number of cholestatic liver diseases. (HEPATOLOGY 2004;39:167–178.)

Multidrug resistance-associated protein 2 (Mrp2; Abcc2) is a multispecific organic anion transporter of the hepatocellular canalicular membrane (CM)1–5 and mediates the efflux of a variety of organic anions, including bilirubin, glucuronides, glutathione (GSH), glutathione conjugates, and sulfated and glucuronidated bile acids.3, 4, 6 Mrp2-dependent secretion of these solutes largely contributes to the bile acid-independent fraction of bile flow,7, 8 which accounts for approximately one-half of the bile flow in rats.9, 10 Impairment of Mrp2 function may lead to a reduction in the bile acid-independent canalicular bile flow. This may occur because biliary excretion of reduced GSH, an important determinant of bile acid-independent bile flow,11 is diminished in transport mutant rats such as transport-deficient (TR-/GY) and Eisai hyperbilirubinemic rats (EHBR). These rats have a congenital absence of the Mrp2 protein.2, 12

Experimental cholestasis has been associated with impaired Mrp2-mediated transport13, 14 and with the down-regulation and altered localization of Mrp2.8 These findings provide the molecular basis for impairment of biliary secretion of a broad range of anionic conjugates and the development of jaundice in human subjects with cholestasis.8 In cholestatic patients, reduced expression of Mrp2 occurs at the messenger RNA (mRNA) level to a lesser extent.15, 16 Importantly, intact Mrp2 mRNA levels, despite reduced canalicular Mrp2 immunostaining in severely cholestatic and jaundiced patients,16 suggest that the potential mechanisms for the regulation of Mrp2 expression in cholestatic liver diseases in humans include not only changes in the rate of transcription but also posttranscriptional/posttranslational changes.16–18 Stimulation and restoration of defective expression and function of Mrp2 in cholestasis may be an important target for specific pharmacotherapeutic interventions.19

Plants contain abundant bioactive materials. Kampo (a Chinese/Japanese herb) medicines are now manufactured in Japan using modern scientific quality controls. Using ethical guidelines, these medicines have been administered to patients with a variety of diseases for approximately 20 years. Inchin-ko-to (IKCT), one such medicine, has been recognized as a “magic bullet” for jaundice and has long been used in Japan and China as a choleretic20 and hepatoprotective21–23 agent for various types of liver diseases. No well-controlled, clinical human studies have been conducted, but clinical and experimental evidence shows the effectiveness of ICKT and its constituents in the treatment of jaundice.20, 24–27 Various ingredients with choleretic activity have been identified in ICKT. Genipin, an intestinal bacterial metabolite of geniposide, the major ingredient in ICKT (Fig. 1),28 shows remarkable choleretic activity.20, 29 Considerable evidence29 suggests that genipin may exert its choleretic activity in a bile acid-independent manner and modulate GSH contents in the liver. Although all aspects of the choleretic activity of genipin have not been clarified, we hypothesize that genipin stimulates Mrp2 function.

Figure 1.

Structures of geniposide (A), genipin (B), and genipin 1-O-β-D-glucuronide (C). A major ingredient of ICKT, geniposide, is converted to an active metabolite, genipin, by intestinal bacteria. Genipin is transported to the liver via portal circulation and subject to conjugation (mainly with glucuronide). Using thin-layer and/or high-performance liquid chromatographic analyses, approximately 20% of infused genipin was secreted into the bile as the main metabolite 1-O-β-D-glucuronide (unpublished observation). Unconjugated genipin was detected, but the amount was approximately 1/50 of that of genipin-1-O-β-D-glucuronide. The other unidentified metabolites detected in the bile also occurred in small amounts.

In the current study, we investigated whether genipin-induced choleretic activity is mediated by its effect on Mrp2, the transporter that plays the major role in generation of bile acid-independent bile flow. Biliary secretion of Mrp2-specific substrates and the protein mass, subcellular localization, and steady-state mRNA level of Mrp2 were assessed in rat liver tissue specimens after an infusion of genipin. Mrp2 function was assessed by the adenosine triphosphate (ATP)-dependent uptake of Mrp2-specific substrates using canalicular membrane vesicles (CMVs) from the liver. An in vivo comparative study on the effects of genipin in Sprague Dawley rats (SDRs) and EHBRs was also performed.

Abbreviations

Mrp2, multidrug resistance-associated protein 2; CM, canalicular membrane; GSH, glutathione; EHBR, Eisai hyperbilirubinemic rat; mRNA, messenger RNA; ICKT, Inchin-ko-to; ATP, adenosine triphosphate; CMV, canalicular membrane vesicle; SDR, Sprague Dawley rat; UDC, ursodeoxycholate; E217βG, estradiol 17-β-D-glucuronide; LTC4, leukotriene C4; TLC-3S, taurolithocholate-3-sulfate; MTX, methotrexate; Bsep, bile salt export pump; pAb, polyclonal antibody; dppIV, depeptidyl peptidase IV; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; PCR, polymerase chain reaction.

Materials and Methods

Animals.

Male SDRs and EHBRs weighing 180 to 220 g were purchased from Nihon SLC (Shizuoka, Japan). All experiments were conducted according to the institution's guidelines for the care and use of laboratory animals in research.

Experimental Design.

General anesthesia was used in the animal experiments. The animals were anesthetized with urethane (1 g/kg body weight intraperitoneally) and the common bile duct was cannulated with a polyethylene tube (outer diameter, 0.6 mm; SP10, Natsume, Tokyo, Japan). Immediately after the cannulation, bile specimens were collected at 30-minute intervals over a period of 90 minutes. Genipin (Wako, Osaka, Japan) or ursodeoxycholate (UDC, sodium salt, Tokyo Kasei, Tokyo, Japan) was dissolved in saline (vehicle) and continuously infused at a rate of 1 μmol/minute per 100 g body weight into a jugular vein 30 minutes after the cannulation. In some experiments, genipin, geniposide, or ICKT (Tsumura, Tokyo, Japan) was suspended in saline and administered orally.

Biliary Analysis.

Bile specimens were collected at 30-minute intervals into preweighed tubes and bile flow was calculated by the weight of each specimen. GSH concentration in bile diluents or liver homogenates was determined using a GSH assay kit (GSH-400, Oxis International Inc., Portland, OR). Total bilirubin concentration in the bile specimen was measured by the Michaëlsson method and total bile acid concentration was measured by an enzymatic technique using 3α-hydroxysteroid dehydroxygenase.30

CM Preparations.

CMVs from liver homogenates were prepared using Percoll (Amersham Biosciences Corp., Piscataway, NJ) and sucrose gradient centrifugation as described by Müller et al.31

Transport Studies With CMVs.

Transport studies were performed using the rapid filtration technique described previously.32 Briefly, a 16-μL transport medium (10 mmol/L Tris, 250 mmol/L sucrose, 10 mmol/L MgCl2, pH 7.4) containing radiolabeled compounds ([3H] estradiol 17-β-D-glucuronide [E217βG], [3H] leukotriene C4 [LTC4], [3H] taurolithocholate-3-sulfate [TLC-3S], or [3H] methotrexate [MTX]), with or without an unlabeled substrate, was preincubated at 37°C for 3 minutes and rapidly mixed with 4 μL of CMV suspension (8-10 μg protein). The reaction mixture contained 5 mmol/L ATP or 5 mmol/L adenosine monophosphate and an ATP-regenerating medium. The ATP-dependent uptake of radiolabeled ligands was calculated by subtracting the ligand uptake in the absence of ATP from that in the presence of ATP.

Determination of Kinetic Parameters.

The CMVs isolated from the vehicle, genipin, or UDC-treated liver tissue specimens were incubated at 37°C for 10 minutes in a medium containing 3H-labeled (100 nmol/L) and unlabeled E217βG. The kinetic parameters for ligand uptake were estimated from the following equation: V0/S = Vmax/(Km + S), where V0 is the rate of initial uptake of the ligand by CMV (pmol/min−1/mg protein−1), S is the ligand concentration in the medium (μM), Km is the Michaelis constant (μmol/L), and Vmax is the maximum uptake rate (pmol/min−1/mg protein−1). The equation was fitted to the uptake data by an iterative nonlinear least-squares method using a MULTI program33 to obtain estimates of the kinetic parameters.

Immunoblot Analysis of Mrp2 and Bile Salt Export Pump in the Liver.

Immunoblotting of Mrp2 and bile salt export pump (Bsep) was performed using 100 μg of crude plasma membrane proteins and 2 μg of CMV proteins isolated from liver homogenate as described previously.34 The membrane was first probed with the polyclonal antibody (pAb) raised against the carboxy terminal of the rat Mrp235 or the pAb raised against the carboxy terminal of the rat Bsep.36 The membrane was reprobed with a monoclonal antibody raised against rat dipeptidyl peptidase IV (dppIV/CD26, a reference marker of CM protein; PharMingen International, Tokyo, Japan). The immunoreactive bands were visualized, quantified, and normalized to the amounts of dppIV present in each specimen and then averaged (five rats in each group).

Immunohistochemical Localization of Mrp2 and Bsep in the Liver.

For studies on light microscopic localization of Mrp2 and Bsep, liver tissue specimens (each approximately 50 mg) taken from rats 30 minutes after intravenous administration of vehicle, UDC, or genipin were frozen in Freon/liquid nitrogen, embedded in OCT compound (Miles, Elkhart, IN), cut into 6-μm–thick sections, and mounted on slides. The sections were immunostained with the pAb of Mrp2 or the pAb of Bsep as described previously.34

Immunoelectron Microscopic Localization of Mrp2 in the Liver.

For studies on electron microscopic localization of Mrp2, liver tissue specimens (each approximately 50 mg) were fixed for 6 hours in 4% paraformaldehyde dissolved in 0.01 mol/L phosphate-buffered saline (PBS, pH 7.4). The tissue specimens were immersed in graded concentrations of sucrose in PBS (10% for 1 hour, 15% for 2 hours, and 20% for 4 hours) and embedded in OCT compound (Tissue-Tek; Miles). Six-micrometer–thick frozen sections were mounted on slides. After pretreatment with normal goat serum, the sections were immunostained with the pAb of Mrp2 diluted 1:300 in bovine serum albumin-PBS at 4°C overnight. A peroxidase-labeled antibody against rabbit immunoglobulin G (Dako, Glostrup, Denmark) was applied as the secondary antibody for 3 hours. After being dipped in freshly prepared 3,3′-diaminobenzidine dissolved in 0.05 mol/L Tris-HCl (pH 7.6), the sections were developed with H2O2 in diaminobenzidine solution. After osmification and dehydration with ethanol, the resulting tissue sections were embedded in Quetol-812 (Nisshin EM, Tokyo, Japan), which was polymerized at 60°C for 72 hours. Ultrathin sections were made and inspected using electron microscopy.

Determination of Steady-State Mrp2 and Bsep mRNA Levels in the Liver.

Steady-state mRNA levels of Mrp2, Bsep, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in rat liver tissue specimens were determined by real-time quantitative polymerase chain reaction (PCR) using the GeneAmp 5700 sequence detection system (Applied Biosystems, Foster City, CA). PCR was performed in triplicate. Primers and probes of Mrp2 and Bsep were designed from complementary DNA sequences for rat Mrp235 and Bsep36 using Primer Express (Applied Biosystems). For GAPDH, the primers and probe in predeveloped TaqMan assay reagents (Applied Biosystems) were used. The TaqMan EZ reverse-transcription PCR kit (Applied Biosystems) or reverse transcriptase qPCR Master Mix (Eurogentec SA, Seraing, Belgium) was used in the assay. Data were normalized to the amounts of GAPDH mRNA present in each specimen and then averaged (five rats in each group).

Statistics.

Values are given as means ± SEM. The means of two groups were compared with the unpaired Student's t test or one-way ANOVA with the post-hoc Scheffe's F test. P < .05 was defined as a statistically significant value.

Results

Effect of Genipin on Bile Flow and Biliary Secretion of Constituents.

To characterize the choleretic effect of genipin, genipin was intravenously infused into SDRs and EHBRs. In SDRs, the bile flow was relatively constant in the vehicle-treated group (Fig. 2A). A continuous intravenous infusion of genipin at a rate of 1 μmol/min/100 g body weight rapidly increased the bile flow (Fig. 2A). The bile flow reached a maximum plateau 60 minutes after the start of the infusion and remained at that level until the end of the experiment. An infusion of UDC at the same rate resulted in a similar increase in bile flow. The basal bile flow was significantly lower in EHBRs than in SDRs (Fig. 2B; 3.91 ± 0.09 vs. 5.39 ± 0.44 μL/min/100 g of body weight; P < .01). EHBRs treated with genipin and EHBRs treated with UDC showed clear differences in the choleretic effect. The bile flow increased significantly after UDC treatment in both strains, whereas the choleresis stimulated by genipin treatment in SDRs was not observed in EHBRs (Fig. 2B).

Figure 2.

Bile flow (A,B) and rate of biliary secretion of GSH (C,D) after bolus infusion of genipin. The bile specimen was collected over a 30-minute period before infusion in SDRs (A and C) and Mrp2-defective EHBRs (B and D). The initial values of bile flow and rate of GSH secretion were calculated. The rats were continuously infused with a vehicle (○), genipin (•), or UDC (□). The choleretic activity and rate of GSH secretion were measured at 30 minute-intervals over a 90-minute period. Data are given as means ± SEM (n = 6). **P < .01, significantly different from the vehicle-treated group.

The different effects of genipin in the two strains was also observed in the biliary excretion rate of GSH, the level of which affects the bile acid-independent choleresis. In SDRs, treatment with genipin increased the biliary GSH excretion rate to approximately 300% to 350%, as much as that in the vehicle-treated group (Fig. 2C), whereas the capability of GSH output stimulated by genipin treatment was defective in EHBRs (Fig. 2D). The GSH concentration in the liver tissue specimens was determined because strong enhancement of GSH secretion stimulated by genipin may affect hepatic GSH levels. The basal hepatic GSH concentration was significantly higher in EHBRs than in SDRs (P < .01; Table 1). In SDRs, genipin treatment for 90 minutes reduced the hepatic GSH concentration to apprximately 15% of that in the vehicle-treated group, whereas there was no depletion of hepatic GSH in EHBRs treated with genipin. UDC did not affect hepatic GSH concentrations in either strain, despite a remarkable choleretic action. Furthermore, high-performance liquid chromatography analysis of GSH and glutathione disulfide (GSSG) concentrations in the bile specimens collected from vehicle and genipin-treated SDRs revealed that genipin had no effect on the GSSG concentration (data not shown), suggesting the compound enhanced biliary secretion of GSH without affecting the intracellular oxidative status of hepatocytes.

Table 1. GSH Concentration in Rat Liver After Bolus Infusion of Genipin
GroupStrains
SDREHBR
  • NOTE. Rats were killed after infusion of a vehicle, genipin, or UDC. Livers were removed and homogenized. GSH concentration (μmol/g) was determined as described in the Materials and Methods. Data are means ± SEM (n = 10).

  • *

    P < .01, significantly different from the vehicle-treated group.

  • P < .01, significantly different from the corresponding group of SDR.

  • Abbreviations: GSH, glutathione; SDR, Sprague Dawley rat; UDC, ursodeoxycholate; EHBR: Eisai hyperbilirubinemic rat.

Vehicle4.0 ± 0.48.1 ± 0.7
Genipin0.8 ± 0.3*8.1 ± 0.7
UDC3.8 ± 0.79.6 ± 0.4

To determine whether the observed choleretic activity of genipin is associated with choleresis by a bile acid-dependent mechanism or stimulation of inorganic ion transport, biliary secretion rates of total bile acids and electrolyte compositions were determined. Table 2 shows biliary secretion rates of total bile acids, biliary concentrations of the major electrolytes, and total bilirubin conjugates in SDRs. The secretion of total bilirubin conjugates, Mrp2 substrates, was increased only by genipin treatment. The secretion rates of total bile acids in the genipin and vehicle-treated groups were similar (Table 2). In contrast, UDC treatment resulted in a marked increase in the biliary secretion rate of total bile acids, indicating that the choleretic action of UDC is bile acid dependent. No significant differences were found between the compositions of major electrolytes in the bile specimens in the genipin and vehicle-treated groups. Bicarbonate and chloride concentrations in the bile specimens were significantly higher and lower in the UDC-treated group than in the vehicle-treated group, respectively.

Table 2. Biliary Bile Acid Secretion Rates and Concentrations of Electrolytes and Total Bilirubin After Bolus Infusion of Genipin
Time0 min30 min60 min90 min
  • NOTE. The biliary bile acid secretion rate was calculated from the concentration of bile acids. Bile acid concentration in bile was measured by an enzymatic technique using 3α-hydroxysteroid dehydroxygenase. Sodium and potassium concentrations were measured using a photometer. Chloride ions were determined by voltametric determination using a silver electrode. Bicarbonate concentration was measured in a microgasometer. Total bilirubin was analyzed by Michaëlsson's method. Data are means ± SEM (n = 10).

  • Abbreviations: BW, body weight; UDC, ursodeoxycholate.

  • *

    P < .05, significantly different from the vehicle-treated group.

  • P < .01, significantly different from the vehicle-treated group.

Bile acid secretion rate (nmol/min/100 g/BW)
 Vehicle262.8 ± 42.2166.8 ± 13.6144.2 ± 7.5137.1 ± 6.5
 Genipin318.2 ± 28.7166.8 ± 8.5120.3 ± 6.6116.8 ± 8.8
 UDC265.2 ± 18.3461.0 ± 41.7591.7 ± 34.4566.8 ± 19.2
Electrolytes
 Na (mmol/L)
  Vehicle183.5 ± 2.3175.7 ± 4.5177.0 ± 1.9169.7 ± 3.9
  Genipin228.0 ± 5.9183.2 ± 7.8177.8 ± 2.3179.7 ± 2.2
  UDC181.8 ± 2.1180.0 ± 1.3175.8 ± 3.9177.7 ± 2.2
 K (mmol/L)
  Vehicle4.8 ± 0.14.8 ± 0.15.1 ± 0.15.0 ± 0.2
  Genipin6.1 ± 0.25.3 ± 0.25.4 ± 0.25.5 ± 0.2
  UDC4.6 ± 0.14.7 ± 0.15.1 ± 0.15.8 ± 0.2*
 Cl (mmol/L)
  Vehicle83.0 ± 4.989.5 ± 2.092.7 ± 0.791.8 ± 1.9
  Genipin105.0 ± 3.389.8 ± 3.286.7 ± 1.489.0 ± 1.2
  UDC84.7 ± 1.278.5 ± 1.4*74.5 ± 1.3*75.2 ± 1.2
 HCO (mmol/L)
  Vehicle28.5 ± 1.232.2 ± 1.531.4 ± 0.432.4 ± 1.0
  Genipin28.2 ± 1.431.8 ± 0.630.2 ± 0.631.2 ± 0.7
  UDC27.1 ± 1.244.4 ± 2.049.9 ± 2.552.4 ± 2.3
 Total bilirubin (mg/dL)
  Vehicle2.4 ± 0.21.8 ± 0.21.7 ± 0.21.7 ± 0.2
  Genipin3.2 ± 0.66.3 ± 2.09.1 ± 3.412.3 ± 3.2
  UDC2.8 ± 0.41.8 ± 0.11.0 ± 0.11.0 ± 0.1

Effect of Genipin on Mrp2-Mediated Transport In Vitro.

The uptake of Mrp2 substrates ([3H]E217βG, [3H]LTC4, [3H]TLC-3S, and [3H]MTX) by CMVs isolated from liver tissue specimens was studied in the genipin-treated rats. The results were compared with those for the vehicle and UDC-treated rats (Fig. 3A). By treatment with genipin, the ATP-dependent uptake in vitro of [3H]E217βG, [3H]LTC4, [3H]TLC-3S, and [3H]MTX by the CMVs was significantly increased to 265%, 161%, 266%, and 234% of the corresponding values of the CMVs from the vehicle-treated liver tissue specimens, respectively. By treatment with UDC, only the uptake of[3H]MTX by the CMVs was increased to 195%.

Figure 3.

(A) ATP-dependent uptake of Mrp2 substrates into CMVs in SDRs. Livers were harvested 30 minutes after intravenous administration of vehicle, genipin, or UDC. CMVs from vehicle, UDC, or genipin-treated SDR livers were incubated at 37°C for 2 minutes in a medium containing 100 nmol/L [3H]E217βG, 100 nmol/L [3H]LTC4, 100 nmol/L [3H]TLC-3S, or 100 nmol/L [3H]MTX in the presence and absence of ATP. ATP-dependent uptake of the ligands was calculated by subtracting the ligand uptake in the absence of ATP from that in the presence of ATP. Data are expressed as percentages of the values obtained in the vehicle-treated rats. Each column and vertical bar represents the mean ± SEM of triplicate determinations. †P < .05, ‡P < .01, significantly different from the two groups. (B) ATP-dependent uptake of [3H]E217βG into CMVs in EHBRs. Livers were harvested 30 minutes after intravenous administration of vehicle or genipin. CMVs from vehicle or genipin-treated EHBR liver tissue sections were incubated at 37°C for 2 minutes in a medium containing 100 nmol/L [3H]E217βG in the presence and absence of ATP. (C) Concentration dependence of [3H]E217βG uptake by CMVs. Livers were harvested 30 minutes after intravenous administration of vehicle, UDC, or genipin. The CMVs isolated from vehicle (○), UDC (□), and genipin-treated liver tissue sections (•) were incubated at 37°C for 10 minutes in a medium containing 3H-labeled (100 nmol/L) and unlabeled E217βG. The open circles, closed circles, and open squares represent the ATP-dependent uptake by the CMVs. Each point and vertical/horizontal bar represents the mean ± SEM of triplicate determinations. Eadie-Hofstee plots were generated and the kinetic parameters (Km and Vmax) were calculated.

In contrast to SDRs, the ATP-dependent uptake of [3H]E217βG was not stimulated in EHBRs by treatment with genipin (Fig. 3B).

Transport Kinetics of [3H]E217βG.

The ATP-dependent uptake of [3H]E217βG in CMVs was saturable (Fig. 3C). Nonlinear regression analysis of the uptake by the CMVs isolated from liver tissue specimens revealed the following: the saturable uptake was described by a Km value of 12.3 ± 3.2 μmol/L (mean ± SEM) and a Vmax value of 95.8 ± 19.7 pmol/min/mg protein in the vehicle-treated rats; a Km value of 14.3 ± 4.5 and a Vmax value of 252.0 ± 63.4 in the genipin-treated rats; and a Km value of 9.5 ± 1.4 and a Vmax value of 90.0 ± 10.0 in the UDC-treated rats (Fig. 3C). With respect to the transport characteristics of Mrp2 in the genipin-treated liver tissue specimens, a significant increase in the Vmax value but not in the Km value was found. This suggested that genipin increases the protein mass of Mrp2 in the CMVs, whereas genipin does not alter the affinity of Mrp2 with its specific substrate.

Immunoblot Analysis of Mrp2 and Bsep in the Liver.

To determine whether an increase in the protein mass of Mrp2 is involved in the enhancement of choleretic activity by treatment with genipin, immunoblotting of Mrp2 was performed using crude plasma membrane fractions and CM fractions isolated from vehicle, genipin, and UDC-treated liver tissue specimens of SDRs 30 minutes after infusion. The analysis revealed Mrp2 protein as the major band with a molecular weight of approximately 180 kd (Fig. 4A). Although no significant increase was observed in the crude plasma membrane fractions, a significant increase in the Mrp2 protein levels was observed in the CMVs of genipin-treated liver tissue specimens compared with the levels of vehicle-treated liver specimens (Fig. 4A). Using densitometric analysis, genipin treatment significantly increased Mrp2 protein levels in the CMVs to 157 ± 14% (mean ± SEM) of vehicle-treated rats (P < .05), whereas UDC treatment had no significant change in the levels, 71 ±15% of vehicle-treated rats (Fig. 4A). The results suggest a redistribution of Mrp2 protein to CM fractions in genipin-treated livers.

Figure 4.

(A) Immunoblot analysis of Mrp2 in crude plasma membrane (crude) fractions and CM fractions and immunoblot analysis of Bsep in CM fractions isolated from livers of vehicle, genipin, and UDC-treated SDRs. Membrane-enriched fractions were prepared from rat livers 30 minutes after intravenous administration of vehicle, UDC, or genipin. Immunoblotting was performed with 100 μg of crude plasma membrane fractions or with 2 μg of CM fractions. The blots were probed with the pAb raised against Mrp2 or Bsep. P, positive control. The CMVs from LLC-PK1 cells transfected with rat Mrp2 complementary DNA were used as a positive control for Mrp2. Densitometric analysis of protein levels of Mrp2 and Bsep in the CM fractions. Immunoreactive bands of Mrp2 and Bsep in five independent CM fractions were densitometrically quantified and normalized to the amounts of dppIV present in each specimen and then averaged. Data are given as means ± SEM. *P < .05, significantly different from vehicle-treated rats. (B) Real-time PCR of Mrp2 and Bsep mRNA levels in the livers of vehicle, genipin, and UDC-treated SDRs. Livers were harvested 30 and 60 minutes after intravenous administration of vehicle, genipin, or UDC. The abundance of GAPDH mRNA was determined as an internal standard. The data were expressed as averaged percentages of the amounts of GAPDH mRNA present in the specimens. Primers and probes of Mrp2 and Bsep were designed as follows: Mrp2, forward 5′-TTCTGGATCCTCTCGGTCTTATG-3, ′ reverse 5′-ATCTGGAAACCGT AGGAGACGAA-3′, probe FAM5′-CGTATTCCAGTTTCAGACTCTGATACGAGCACTC-3′TAM; Bsep: forward 5′-CATCATTGCGGCCTTGCT-3, ′ reverse 5′-GCGAATCCCGTCAACATTTT-3, ′ probe FAM5′-TTATAACGATCTTCTTCCCCTTTCTGGCTTTATCG-3′TAM.

The Bsep protein levels were the major band with a molecular weight of approximately 160 kd (Fig. 4A). However, the protein levels of Bsep in the CMVs were not increased in the genipin-treated liver tissue specimens, 121 ± 6% of vehicle-treated rats, nor in the UDC-treated rats, 80 ± 16% of vehicle-treated rats (Fig. 4A).

Effect of Genipin on Immunohistochemical Localizations of Mrp2 and Bsep in the Liver.

The immunohistochemical localizations of Mrp2 and Bsep in liver tissue sections were studied. The results are shown in Figures 5 and 6. In light microscopic studies, Mrp2 was generally expressed in the bile canaliculi of the liver tissue sections. Consistent with the increased protein mass of Mrp2 in the CMVs revealed by immunoblot analysis (Fig. 4A), the immunostaining of Mrp2 was more intensive and more diffuse in the bile canaliculi of genipin-treated than in those of vehicle and UDC-treated liver tissue sections (Fig. 5A). The image analysis of light microscopic localization of Mrp2 (Fig. 5B) showed a significant increase in the protein level of Mrp2 in the CM of genipin-treated liver tissue sections (198 ± 18% of vehicle-treated liver tissue sections; P < .05) but no significant increase in UDC-treated liver tissue sections (103 ± 6% of vehicle-treated liver tissue sections). However, the immunostaining of Bsep in the bile canaliculi of genipin-treated liver tissue sections did not differ significantly compared with that of vehicle-treated liver tissue sections (Fig. 5C).

Figure 5.

(A) Immunohistochemical localizations of Mrp2 and Bsep in the livers (light microscopic view). Liver tissue sections were prepared from SDR livers 30 minutes after intravenous administration of vehicle, UDC, or genipin. Mrp2 immunostaining shows a diffuse and linear pattern outlining the CM domain of each liver section from vehicle, UDC, and genipin-treated rats. The immunostaining of Mrp2 was more intense and more diffuse in the bile canaliculi in genipin-treated liver tissue sections compared with vehicle and UDC-treated liver tissue sections. (Bars = 100 μm.) (B) Quantification of the data of light microscopic localization of Mrp2 in the liver tissue sections was performed by image analysis using MetaMorph (Universal Imaging, West Chester, PA). A liver tissue section was prepared from each of the five rats in the vehicle, UDC, and genipin-treated groups. Five photographs of low power (800-1,000 hepatocyte nuclei in each visual field) were taken of each section and then subjected to the image analysis. The bile canalicular area showing immunoreactivity of Mrp2 was extracted in a photograph and the integrated area was calculated. The integrated area calculated for each of the five photographs in a liver tissue section was averaged. The integrated areas of the vehicle, UDC, and genipin-treated groups (five rats in each group) were compared. ‡P < .01, significantly different from the two groups. (C) Bsep immunostaining also shows a diffuse and linear pattern outlining the CM domain of each liver tissue section from vehicle and genipin-treated rats. The immunostaining of Bsep in the bile canaliculi was not significantly different between vehicle and genipin-treated liver tissue sections. (Bars = 100 μm.)

Figure 6.

(A) Immunohistochemical localizations of Mrp2 in SDR liver tissue sections (electron microscopic view). Liver tissue sections were prepared from rat livers 30 minutes after intravenous administration of vehicle or genipin. The Mrp2 protein was localized mostly in canalicular microvilli in vehicle-treated livers, whereas the localization was predominant in both microvilli and CM in genipin-treated livers. The proportion of Mrp2-containing microvilli (▴) to total microvilli (including Mrp2-negative microvilli [▵] was larger in genipin-treated than in vehicle-treated liver tissue sections and the membrane density of Mrp2 protein around the base of microvilli in genipin-treated liver tissue sections (*) was markedly increased in genipin-treated liver tissue sections. Bars = 1.0 μm. (B) The proportion of Mrp2-containing microvilli (▴) to total microvilli (including Mrp2-negative microvilli [▵]) was calculated in the liver tissue sections of vehicle and genipin-treated rats. For 300 microvilli (corresponding to approximately 10-15 bile canaliculi) in each liver tissue section of five rats in each group, the number of Mrp2-containing microvilli was counted. The proportion of Mrp2-containing microvilli to total microvilli was calculated for each liver tissue section and then averaged. ‡P < .01, significantly different from vehicle-treated rats.

In electron microscopic studies, morphometric analysis showed that neither genipin nor UDC treatment caused any significant changes in the pericanalicular zone (defined as a rim of 1 μm around the canaliculus), CM (without microvilli), and microvilli compared with vehicle treatment (data not shown). The localization of Mrp2 protein in bile canaliculi was further studied for pericanalicular zone, CM, and microvilli. Mrp2 protein was found mostly in microvilli in vehicle-treated liver tissue sections, whereas the localization was predominant in both microvilli and CM in genipin-treated liver tissue sections (Fig. 6A). Moreover, in the bile canaliculi, the proportion of Mrp2-containing microvilli to total microvilli (Fig. 6B) was significantly larger in genipin-treated (88 ± 3%; P < .01) than in vehicle-treated liver tissue sections (36 ± 2%) and the density of Mrp2 in the CM was markedly increased in genipin-treated compared with vehicle-treated liver tissue sections (Fig. 6A). The results suggest that with genipin treatment, the localization of Mrp2 shifts from the cytoplasm to the CM or microvilli and that the largest proportion of Mrp2 localizes in these sites.

Effects of Genipin on Expression Levels of Mrp2 and Bsep mRNAs in the Liver.

In contrast to the protein levels, the analysis of real-time PCR revealed that there was no significant difference in the steady-state Mrp2 and Bsep mRNA levels between the genipin-treated and vehicle or UDC-treated liver tissue sections (Fig. 4B). The observation suggests posttranscriptional effects of genipin on the expression levels of Mrp2.

Effects of Oral Administration of ICKT, Geniposide, and Genipin on Choleretic Activity In Vivo.

Although genipin is not currently available as a therapeutic drug, the effect of genipin is believed to be expressed in ICKT, which contains an extremely high content of geniposide (7.0%, in the ICKT used in the study).37 Effects of oral administration of ICKT, geniposide, and genipin on Mrp2-mediated choleretic activity in SDRs were investigated. ICKT (2 g/kg per day), geniposide (150 mg/kg per day), or genipin (100 mg/kg per day) was administered orally for 7 days and choleretic activity was assessed by measuring basal bile flow, rate of biliary secretion of GSH, hepatic GSH concentration, and serum GSH concentration. ICKT, geniposide, or genipin treatment resulted in significant increases in the basal bile flow and rate of biliary secretion of GSH (Table 3). Significant increases in hepatic GSH concentrations were observed in ICKT, geniposide, and genipin-treated groups, whereas the serum GSH concentrations were not affected by the treatments.

Table 3. Comparison of the Choleretic Activities of ICKT, Geniposide, and Genipin In Vivo
GroupBileLiverSerum
Bile Flow (μL/min/100 g BW)GSH Secretion Rate (nmol/min/100 g BW)GSH Concentration (μmol/g)GSH Concentration (μmol/mL)
  • NOTE. Sprague Dawley rats received orally ICKT, geniposide, or genipin for 1 week. Bile specimens were collected over a 30-minute period to estimate the choleretic activity and GSH secretion rate. Blood samples were collected from an abdominal artery and livers were removed. The GSH levels of bile, serum, and liver were determined as described in the Materials and Methods. Each value represents the mean ± SEM (n = 10).

  • Abbreviations: ICKT, Inchin-ko-to; BW, body weight; GSH, glutathione.

  • *

    P < .05, significantly different from the vehicle-treated group.

  • P < .01, significantly different from the vehicle-treated group.

H2O5.4 ± 0.212.6 ± 0.95.5 ± 0.14.1 ± 0.0
ICKT (2 g/kg)7.4 ± 0.325.1 ± 1.47.1 ± 0.14.0 ± 0.1
Geniposide (150 mg/kg)6.8 ± 0.3*23.3 ± 1.46.8 ± 0.24.0 ± 0.1
Genipin (100 mg/kg)7.6 ± 0.427.5 ± 2.96.9 ± 0.33.9 ± 0.1

Effects of a Microfilament Stabilizer Phalloidin on Basal Biliary Secretion and Genipin-Induced Choleresis.

Appropriate vesicular targeting of transporters, including Mrp2, into the CM requires the integrity of an actin network in hepatocytes.38 Therefore, the effects of phalloidin, which stabilizes actin microfilament and inhibits the vesicular targeting of canalicular transporters including Mrp2,38–40 on basal biliary secretion and genipin-induced choleresis were evaluated. Pretreatment of phalloidin (1 mg/kg body weight of phalloidin was injected via the tail vein) markedly decreased the basal bile flow. Continuous genipin infusion at a rate of 1 μmol/min/100 g body weight into a jugular vein 1.5 hours after phalloidin treatment did not restore the phalloidin-induced decrease in bile flow (data not shown). The results suggest that the microfilament-dependent translocation of Mrp2 into the CM is critically involved in the Mrp2-mediated choleretic activity induced by genipin.

Discussion

Various hormones and drugs, including glucocorticoids, pregnane-X receptor ligands, phenobarbital, and UDC, up-regulate Mrp2 mRNA and stimulate Mrp2-dependent organic anion transport.41 The effect of genipin, however, could not be attributed to the regulation of Mrp2 transcription because genipin stimulated Mrp2-mediated transport within 30 minutes after its intravenous infusion with little effect on steady-state mRNA levels of Mrp2 (Fig. 4B). Several Mrp2 substrates, such as sulfinpyrazone, penicillin G, indomethacin,42 and 4-methylumbelliferone,43 enhance Mrp2-mediated transport, presumably by substrate-stimulated ATPase activity. Although Mrp2-mediated transport activity of CMVs isolated from genipin-treated liver tissue sections has been clearly enhanced (Fig. 3A), direct addition of genipin to native CMVs enhanced the transport only marginally (data not shown). Accordingly, the enhancement of Mrp2-mediated transport in the CMVs from genipin-treated liver tissue sections was due to the alteration of Vmax values but not the Km values (Fig. 3C), suggesting that genipin increases Mrp2 protein mass in the CMVs but does not alter the affinity of Mrp2 with its substrates. Therefore, the rapid choleresis induced by genipin may be due primarily to a redistribution of transporter molecules in the genipin-treated liver tissue sections.

The redistribution of Mrp2 in genipin-treated liver tissue sections is most likely due to an increased reinsertion of Mrp2 into the CM or to a decreased retrieval into the subapical compartment, assuming that the endocytosis/exocytosis equilibrium is shifted toward the CM. The increased Mrp2 density in the CM of genipin-treated liver tissue sections is strongly supported by the results of immunohistochemical Mrp2 localizations in the genipin-treated sections (Figs. 5 and 6). Because only the short-term effects of genipin treatment on liver tissue sections were investigated in the current study, it is reasonable to assume that changes in the membrane density of Mrp2 protein during this short period of time would mainly be observed at the site of first contact of carrier protein-transporting vesicles with their target membrane.44 Fusion of vesicles with the apical membrane of hepatocytes is assumed to occur mainly along the circumference, and early alterations in the membrane density of Mrp2 should occur at this site.

Although the mechanism by which genipin enhances the redistribution of Mrp2 protein into bile canaliculi remains to be clarified, we may exclude several possibilities of mechanistic factors. First, genipin 1-O-β-D-glucuronide, the main metabolite of genipin,29 was detected in the bile of SDRs but not in that of EHBRs (data not shown). This observation suggests that genipin is secreted into the bile via Mrp2 in glucuronide forms. Unconjugated genipin and GSH-conjugated genipin are detected only in small amounts, if any, in SDRs. However, the rapid redistribution of Mrp2 protein into the bile canaliculi induced by genipin could not be explained by the suggestion that genipin 1-O-β-D-glucuronide is a Mrp2 substrate, because it has been reported that the intravenous administration of Mrp2 substrates does not enhance Mrp2 function in perfused rat liver.45–47 Second, cyclic AMP regulates transport activity at the CM by stimulating exocytosis and/or fusion of vesicles containing Mrp2.48 However, we did not find any significant increase in the hepatic cyclic AMP concentrations after genipin treatment in rats (data not shown). Third, UDC was suggested to stimulate biliary secretion, due in part to stimulation of the insertion of Mrp2 protein into the CM via Ca2+ and α-protein kinase C (PKC)-dependent mechanisms.44 However, the immunoblot analysis of the PKC isozymes (data not shown) did not reveal any significant induction of the translocation of the PKC isozymes to a particulate membrane fraction, a key step for the activation of PKC.

In summary, the results of the current study suggest that a choleretic effect of the administration of genipin, a major active ingredient of a herbal medicine, ICKT, may be due to a bile acid-independent mechanism through selective stimulation of Mrp2-mediated bile formation and secretion. The mechanism induced by genipin for the stimulatory effect on Mrp2-mediated transport is different from that induced by UDC. This observation suggests the possibility of development of new pharmacotherapeutic strategies aimed at stimulation and restoration of defective Mrp2 expression and function in various types of cholestatic liver diseases, many of which are currently untreatable.

Ancillary