UDP-glucuronosyltransferase (UGT) expressed in various tissues including liver and intestine catalyzes phase II metabolic biotransformation. UGTs conjugate lipophilic compounds with glucuronic acid from UDP-glucuronic acid (UDPGA), thereby increasing hydrophilicity and enhancing excretion through bile and urine (Dutton, 1980). In both humans and rodents, two families of UGT, UGT1 and UGT2, are known. The human UGT1 gene contains 13 individual promoter/first exons and shares exons 2 to 5 (Mackenzie et al., 2005). As with the human genes, rat and mouse Ugt1 genes also share exons 2 to 5 and have 10 and 14 first exons, respectively (Mackenzie et al., 2005). Among species, the numbers of the first exons and pseudogenes differ. There are four, two, and five pseudogenes in human, rat, and mouse UGT enzymes, respectively. For example, human UGT1A4 is functional but rat UGT1A4 and mouse Ugt1a4 are pseudogenes. Human UGT1A9 and mouse Ugt1a9 are functional, but rat UGT1A9 is a pseudogene. In the case of UGT1A6, mice have two functional copies of Ugt1a6, Ugt1a6a and Ugt1a6b, whereas humans and rats have one copy of UGT1A6. On the other hand, the UGT2 gene in humans, rats, and mice consists of six exons, except for UGT2A1 and UGT2A2 genes (seven exons). The UGT2 family contains three enzymes of the UGT2A subfamily and seven enzymes of the UGT2B subfamily among humans, rats, and mice. The UGT2 subfamily shares more than 70% sequence homology, thus orthologs across species are hard to elucidate (Mackenzie et al., 2005). These species differences in UGT genes could result in species differences in UGT activities.

Liver and intestine are the important tissues for drug metabolism including glucuronidation. The expression of UGT mRNAs has been reported in various tissues in humans (Tukey and Strassburg, 2000), rats (Shelby et al., 2003), and mice (Buckley and Klaassen, 2007). Comparison of expression levels among UGT enzymes is difficult. Because UGT antibodies are not available for the specific quantification of each UGT enzyme, the expression ratio of each UGT enzyme remains unclear in the liver and intestine. Which UGT enzyme in mice and rats corresponds to human UGT enzyme is controversial.

The UGT substrates include many endogenous and xenobiotic compounds. Typical endogenous substrates are bilirubin and estradiol. In particular, hyperbilirubinemia caused by a UGT defect is well known (Burchell et al., 2000). Small xenobiotic planar phenols such as 4-methylumbelliferone (4-MU) and 4-nitrophenol (4-NP) are often used for measuring the glucuronidation (Hanioka et al., 2006). Propofol, widely used as an intravenous anesthetic for the induction and maintenance of anesthesia, is metabolized mainly to its glucuronide in humans (Sneyd et al., 1994). A prodrug of mycophenolic acid (MPA), mycophenolate mofetil (MMF), exhibited severe gastrointestinal toxicity and a relationship between its toxicity and MPA glucuronidation is suspected in rats (Stern et al., 2007). Morphine, an analgesic drug used for the treatment of acute and chronic pain syndromes in cancer patients, is glucuronidated mainly by UGT2B7 in a stereoselective manner to morphine 3-O- and 6-O-glucuronide in humans (Coffman et al., 1998), whereas the main metabolite in humans and rodents is 3-O-glucuronide. Trifluoperazine (TFP), which is one of the antischizophrenic agents, is metabolized as N-glucuronide by human UGT1A4 (Uchaipichat et al., 2006).

In drug development, experimental animals are frequently used for pharmacokinetic studies. The investigation of species differences in drug metabolism is essential for understanding the results of in vivo animal studies. However, there is insufficient information on species differences in glucuronidation. The purpose of the present study was to clarify the species differences in mouse and rat UGT activities in intestine and liver using seven typical substrates (estradiol, 4-MU, 4-NP, MPA, propofol, morphine, and TFP).

Materials and Methods

Materials. UDPGA, alamethicin, aprotinin, bestatin, leupeptin, trypsin inhibitor (type II-S: soybean), estradiol, estradiol 3-O-glucuronide, 4-MU, 4-MU O-glucuronide, and 4-NP O-glucuronide were purchased from Sigma-Aldrich (St. Louis, MO). (p-Amidinophenyl)methanesulfonyl fluoride, MPA, 4-NP, and TFP were obtained from Wako Pure Chemicals (Osaka, Japan). Morphine hydrochloride was purchased from Takeda Pharmaceutical Company (Osaka, Japan). Morphine 3-O-glucuronide was kindly provided by Dr. Kazuta Oguri (Kyusyu University, Fukuoka, Japan). MPA O-glucuronide and carboxybutoxy ether of mycophenolic acid were generous gifts from Roche Bioscience (Palo Alto, CA). All other chemicals and solvents were of analytical grade or the highest grade commercially available.

Preparation of Intestinal and Hepatic Microsomes. C57BL/6J mice (7-week-old male, 21–26 g, and female, 17–20 g), BALB/c, C3H/HeJ, DBA/2, ddY, and ICR mice (7-week-old male, 21–34 g), and Sprague-Dawley rats (7-week-old male, 220–240 g, and female, 140–160 g) were obtained from SLC Japan (Hamamatsu, Japan). Animals were housed in the institutional animal facility in a controlled environment (temperature 25 ± 1°C and 12-h light/dark cycle) with access to food and water ad libitum. Animals were acclimatized for a week before use. Animal were maintained in accordance with the National Institutes of Health Guide for Animal Welfare of Japan, as approved by the Institutional Animal Care and Use Committee of Kanazawa University.

Pooled duodenal, jejunal, iliac, and colonic microsomes from five mice or rats were prepared according to the method of Emoto et al. (2000) with slight modifications. Briefly, duodenum, jejunum, ileum, and colon were divided, cut longitudinally, and then washed in ice-cold 1.15% KCl by gentle swirling. The intestine was suspended in 3 vol of ice-cold buffer A [50 mM Tris-HCl buffer (pH 7.4) containing 150 mM KCl, 20% (v/v) glycerol, 1 mM EDTA, 1 mM (p-amidinophenyl)methanesulfonyl fluoride, 1 mg/ml trypsin inhibitor, 10 μM leupeptin, 0.04 U/ml aprotinin, and 1 μM bestatin] and homogenized using a motor-driven Teflon-tipped pestle. The homogenate was centrifuged at 9000g at 4°C for 20 min, and then the supernatant was centrifuged at 105,000g at 4°C for 60 min. The microsomal pellets were resuspended in ice-cold buffer A.

Pooled hepatic microsomes from five mice or rats were prepared according to the method described by Emoto et al. (2000). Protein concentrations were determined according to the method of Lowry et al. (1951) using bovine serum albumin as the standard.

Enzyme Assays. A typical incubation mixture contained 50 mM Tris-HCl buffer (pH 7.4), 5 mM MgCl₂ (except propofol and TFP, 10 mM), 25 μg/ml alamethicin, UDPGA, microsomes, and a substrate. In the preliminary study, the concentration of UDPGA was confirmed to reach a plateau level for each UGT activity. Microsomes with alamethicin were placed on ice for 15 min. In the preliminary study, a change of preincubation time did not affect the UGT activities. The final concentration of methanol (estradiol and propofol) or ethanol (MPA) in the reaction mixture was <1.5% (v/v). As described by Uchaipichat et al. (2004), because methanol (>1%) decreased more than 20% of the UGT1A6 activity, the final concentrations of methanol for 4-MU and 4-NP O-glucuronidation were <0.75% and <0.5% (v/v), respectively. A portion of the sample was subjected to high-performance liquid chromatography. The flow rate was 1.0 ml/min.

Estradiol 3-O-glucuronidation was determined according to the method of Yoon et al. (2003) with slight modifications. In the preliminary study, the rate of this activity was linear with respect to the microsomal protein concentrations (<0.1 mg/ml in mice and <0.05 mg/ml in rats) and incubation time (<15 min in mice and <30 min in rats). Therefore, in both mice and rats, the concentrations of microsomal protein were 0.05 mg/ml, and the reaction mixture was incubated for 15 min. The concentrations of UDPGA were 3 (mice) and 7 mM (rats). The analytical column was a TSKgel ODS-80Ts (4.6 × 150 mm, 5 μm; TOSOH, Tokyo, Japan), and the mobile phase was acetonitrile-1 mM perchloric acid (25:75, v/v).

4-MU O-glucuronidation was determined according to the method of Uchaipichat et al. (2004) with slight modifications. In the preliminary study, the rate of this activity was linear with respect to the microsomal protein concentrations (<0.2 mg/ml in mice and <0.1 mg/ml in rats) and incubation time (<30 min in mice and <15 min in rats). In the reaction mixture, the concentrations of microsomal protein were 0.1 (mice) and 0.05 mg/ml (rats). In both mice and rats, the concentration of UDPGA was 3 mM, and the reaction mixture was incubated for 15 min. The analytical column was a CAPCEL PAK C₁₈ UG120 (4.6 × 150 mm, 5 μm; Shiseido, Tokyo, Japan), and the mobile phase was methanol-50 mM potassium phosphate buffer, pH 4.5 (20:80, v/v).

4-NP O-glucuronidation was determined according to the method of Hanioka et al. (2001) with slight modifications. In the preliminary study, the rate of this activity was linear with respect to the microsomal protein concentrations (<0.2 mg/ml in both mice and rats) and incubation time (<30 min in mice and <15 min in rats). In the reaction mixture, the concentrations of microsomal protein were 0.1 (mice) and 0.05 mg/ml (rats). In both mice and rats, the concentration of UDPGA was 3 mM, and the reaction mixture was incubated for 15 min. The analytical column was a Mightysil RP-18 (4.6 × 150 mm, 5 μm; Kanto Chemical, Tokyo, Japan), and the mobile phase was methanol-50 mM potassium phosphate buffer, pH 6.5 (6:94, v/v).

Propofol O-glucuronidation was determined according to the method of Fujiwara et al. (2007) with slight modifications. In the preliminary study, the rate of this activity was linear with respect to the microsomal protein concentrations (<0.25 mg/ml in mice and <1.0 mg/ml in rats) and incubation time (<30 min in mice and <45 min in rats). In the reaction mixture, the concentrations of microsomal protein were 0.25 (mice) and 0.5 mg/ml (rats). In both mice and rats, the concentration of UDPGA was 5 mM, and the reaction mixture was incubated for 30 min (mice) and 45 min (rats). The analytical column was a Mightysil RP-18 (4.6 × 150 mm, 5 μm), and the mobile phase was acetonitrile-0.1% acetic acid (40:60, v/v).

MPA O-glucuronidation was determined according to the method of Picard et al. (2005). In the preliminary study, the rate of this activity was linear with respect to the microsomal protein concentrations (<0.2 mg/ml in mice and <0.5 mg/ml in rats) and incubation time (<45 min in mice and <60 min in rats). In the reaction mixture, the concentrations of microsomal protein were 0.1 (mice) and 0.2 mg/ml (rats) and the concentration of UDPGA was 7 (mice) and 3 mM (rats). The reaction mixture was incubated for 30 min. Carboxybutoxy ether of mycophenolic acid (4.6 nmol) was added as an internal standard. The analytical column was an Inertsil ODS-3 (4.6 × 250 mm, 5 μm; GL Sciences, Tokyo, Japan), and the mobile phase was acetonitrile-0.1% phosphoric acid (30:70, v/v).

Morphine 3-O-glucuronidation was determined according to the method of Katoh et al. (2005) with slight modifications. In the preliminary study, the rate of this activity was linear with respect to the microsomal protein concentrations (<1.0 mg/ml in mice and <0.2 mg/ml in rats) and incubation time (<105 min in mice and <30 min in rats). In both mice and rats, the concentrations of UDPGA and microsomal protein were 10 mM and 0.2 mg/ml, respectively. The reaction mixture was incubated for 90 min (mice) and 30 min (rats). The analytical column was a Develosil C30-UG-5 (4.6 × 150 mm, 5 μm; Nomura Chemical, Aichi, Japan) and the mobile phase was 50 mM sodium dihydrogen phosphate.

TFP N-glucuronidation was determined according to the method of Uchaipichat et al. (2006) with slight modifications. The concentrations of UDPGA and microsomal protein were 2.5 mM and 0.25 mg/ml, respectively. The reaction mixture was incubated for 30 min. The analytical column was an Inertsil ODS-3 (4.6 × 150 mm), and the mobile phase was acetonitrile-0.1% trifluoroacetic acid (30:70, v/v).

Kinetic Analyses. The kinetic studies were performed using pooled liver microsomes of C57BL/6J mouse, pooled duodenal microsomes of C57BL/6J mouse, pooled rat liver microsomes, and pooled rat duodenal microsomes. When the kinetic parameters were determined, the concentrations of estradiol, 4-MU, 4-NP, propofol, MPA, and morphine ranged from 5 to 150, 10 to 640, 10 to 1000 (10 to 1500 in mouse duodenum), 5 to 2000, 10 to 1000, and 20 to 6000 μM, respectively. The kinetic parameters and S.E.s were estimated from the fitted curves using a computer program (KaleidaGraph; Synergy Software, Reading, PA) designed for nonlinear regression analysis. The following equations were used: Michaelis-Menten equation, V = V_max · [S]/(K_m + [S]); Hill equation, V = V_max · [S]ⁿ/(S₅₀ⁿ + [S]ⁿ); and substrate inhibition equation, V = V_max · [S]/(K_m + [S] + [S]²/K_si), where V is the velocity of the reaction, S is the substrate concentration, K_m is the Michaelis-Menten constant, V_max is the maximum velocity, S₅₀ is the substrate concentration showing the half-V_max, n is the Hill coefficient, and K_si is the substrate inhibition constant. Intrinsic clearance (CL_int) was calculated as V_max/K_m for Michaelis-Menten kinetics. For sigmoidal kinetics, maximum clearance (CL_max) was calculated as V_max · (n – 1)/(S₅₀ · n(n – 1)^1/n) to estimate the highest clearance (Houston and Kenworthy, 2000). In the present study, if the Hill coefficient was more than 1.2, the kinetic data were fit to the Hill equation.

UGT Activities in Intestine. The UGT activities were determined using pooled duodenal, jejunal, ileal, and colonic microsomes from C57BL/6J mice or rats. The concentrations of estradiol, 4-MU, 4-NP, propofol, MPA, and morphine were 20, 100, 300, 60, 200, and 200 μM, respectively, which were the concentrations below apparent K_m values in duodenal microsomes from mice or rats. Other experimental conditions were the same as described above. For investigation of strain differences in mice, the UGT activities were determined in pooled duodenal and liver microsomes from six strains using six substrates.

Results

Kinetic Analyses of UGT Activities in Mouse and Rat Liver Microsomes Using Seven UGT Substrates. To investigate species and sex differences, kinetic analyses of estradiol 3-O-, 4-MU O-, 4-NP O-, propofol O-, MPA O-, morphine 3-O-, and TFP N-glucuronidation were determined in liver microsomes from male mice, female mice, male rats, and female rats. When kinetic parameters were determined, the concentrations of estradiol, 4-MU, 4-NP, propofol, MPA, and morphine ranged from 5 to 150, 10 to 640, 10 to 1000, 5 to 2000, 10 to 1000, and 20 to 6000 μM, respectively. Kinetic parameters are shown in Table 1. In all kinetic analyses, the r values for fitting to the kinetic model were more than 0.97. The estradiol 3-O-glucuronidation in all liver microsomes was fitted to the Hill equation. The Hill coefficient in both male and female mice (2.3) was higher than that in both male and female rats (1.8). The 4-MU O-glucuronidation in all liver microsomes was fitted to the Hill equation. Kinetic parameters were different between mice and rats. The 4-NP O-glucuronidation in all liver microsomes was fitted to the Michaelis-Menten kinetics. The propofol O-glucuronidation in male and female mouse liver microsomes was fitted to the substrate inhibition kinetics, whereas those in male and female rat liver microsomes were fitted to the Michaelis-Menten kinetics. The MPA O-glucuronidation in all liver microsomes was fitted to the Michaelis-Menten kinetics. In females, species differences in the apparent K_m and V_max values were observed as in males. The morphine 3-O-glucuronidation in all liver microsomes was fitted to the Michaelis-Menten kinetics. The TFP N-glucuronidation in male mouse and male rat liver microsomes was measured. However, this activity was not detected in either mice or rats.

Microsomal UGT Activities in Duodenum, Jejunum, Ileum, and Colon from Mice and Rats. UGT activities using six different substrates were measured in microsomes prepared from male and female mouse and rat duodenum, jejunum, ileum, and colon (Fig. 1). In male mice, 4-NP O- and morphine 3-O-glucuronidation in colon were higher than those in other parts of the intestine, whereas estradiol 3-O-, 4-MU O-, and propofol O-glucuronidation in duodenum were higher than those in other parts of the intestine. In male rats, the duodenum exhibited higher activities of 4-NP, propofol, MPA, and morphine glucuronidation than those in other parts of the intestine. In female rats, all UGT activities except morphine were decreased distally through the intestine. In all parts of the intestine, 4-MU O-, 4-NP O-, and morphine 3-O-glucuronidation in mice were lower than those in rats.

Kinetic Analyses of UGT Activities in Duodenal Microsomes from Male Mice and Rats Using Seven UGT Substrates. Kinetic analyses of estradiol 3-O-, 4-MU O-, 4-NP O-, propofol O-, MPA O-, morphine 3-O-, and TFP N-glucuronidation in microsomes from male mice and rats were determined. Kinetic parameters are shown in Table 2. The estradiol 3-O-glucuronidation in duodenal microsomes from mice was fitted to the Hill equation. In comparison with liver, the CL_max values in duodenum were 6.5-fold lower. The Hill coefficient in duodenum was also lower than that in liver. On the other hand, this activity in rat duodenal microsomes was fitted to the Michaelis-Menten equation rather than to the Hill equation. When the kinetic data from rat duodenal microsomes were fit to the Hill equation, the S₅₀ value, the V_max value, and the Hill coefficient were 29 μM, 1.2 nmol/min/mg protein, and 1.1, respectively. The 4-MU O-glucuronidation in duodenal microsomes from male mice and rats was fitted to the Michaelis-Menten equation. The 4-NP O-glucuronidation in duodenal microsomes from male mice did not reach a plateau level up to 1500 μM. In rat duodenal microsomes, this activity was fitted to the Michaelis-Menten kinetics with lower K_m values than in mice. The propofol O-glucuronidation in duodenal microsomes showed substrate inhibition at substrate concentrations >400 μM in male mice and >500 μM in female rats, but the plot did not fit to either the substrate inhibition kinetics or the two-site model used by Houston and Kenworthy (2000). Therefore, we did not calculate the kinetic parameters for propofol O-glucuronidation in duodenal microsomes. The MPA O-glucuronidation in duodenal microsomes from both male mice and rats was fitted to the Michaelis-Menten kinetics. The CL_int value in mouse duodenum was 11-fold lower than that in liver. However, in rats, that value in duodenum (20 μl/min/mg protein) was similar to the value in liver (17 μl/min/mg protein). The morphine 3-O-glucuronidation in duodenal microsomes from both male mice and rats was fitted to the Michaelis-Menten kinetics. The CL_int value in mouse duodenum was lower than that in liver. The TFP N-glucuronidation in mouse and rat duodenal microsomes was determined. As in liver, these activities in both mice and rats were not detected.

Strain Differences of UGT Activities in Mouse Duodenal Microsomes. To investigate the strain differences in mice, the UGT activities using six UGT substrates were determined in duodenal microsomes from C57BL/6J, BALB/c, C3H/HeJ, DBA/2, ddY, and ICR mice (Fig. 2). For all UGT activities except 4-MU, C3H/HeJ mice showed the highest values among the six strains. The UGT activities in BALB/c and C3H/HeJ mice were relatively high compared with those in other mice. The UGT activities in the six strains showed a similar tendency between 4-NP O- and MPA O-glucuronidation or between estradiol 3-O- and propofol O-glucuronidation. The strain differences in UGT activities varied according to the substrates. Conversely, in liver, there was not much difference in UGT activities among the six mouse strains (Table 3).

Discussion

Information on species differences in UGT activities is insufficient. Intestine as well as liver plays an important role in xenobiotic metabolism. Numerous phase I and phase II drug-metabolizing enzymes are expressed in intestine. Recently, species differences in glucuronidation of the anti-human immunodeficiency virus drug bevirimat were reported (Wen et al., 2007). However, there have been no comprehensive analyses of UGT activities in intestine and liver from mice and rats. In the present study, kinetic analyses of UGT activities using seven typical substrates (estradiol, 4-MU, 4-NP, propofol, MPA, morphine, and TFP) in intestine and liver microsomes from mice and rats were investigated. In addition, strain differences in UGT activities in mouse duodenum were studied.

Estradiol 3-O-glucuronidation is catalyzed mainly by human UGT1A1. Rat UGT1A1 is responsible for this reaction (King et al., 1996). In human liver microsomes, estradiol 3-O-glucuronosyltransferase yielded an S₅₀ value of 17 μM, and a Hill coefficient of 1.8 (Fisher et al., 2000a). The S₅₀ value was almost the same among three species. In microsomes from humans, the estradiol 3-O-glucuronidation in small intestine was higher than that in liver as reported by Fisher et al. (2000b), which was contrary to the present results in mice and rats. In the present study, the CL_max value for estradiol 3-O-glucuronidation in female rats was higher than that in male rats. For glucuronidation of bilirubin, another UGT1A1 substrate, sex differences in Wistar rats have been clarified (Muraca and Fevery, 1984). However, the sex differences in UGT1A1 activity are still unclear.

UGT1A6 is a major enzyme catalyzing the glucuronidation of various simple phenolic compounds such as 4-MU and 4-NP in the liver. UGT1A6 is likely to be functionally orthologous among several species including humans, rats, mice, and rabbits (Iyanagi et al., 1986; Harding et al., 1988; Lamb et al., 1994). The 4-MU glucuronidation in human liver microsomes followed the Michaelis-Menten kinetics (Miners et al., 1988) and was catalyzed by several human UGTs (Uchaipichat et al., 2004). The different kinetic models between humans and rodents might be due to the different UGT enzymes. In duodenum, the apparent K_m value for 4-NP glucuronidation in mice was higher than that in rats, suggesting that the affinity of 4-NP to mouse Ugt may be lower than that to rat UGT. In contrast, in liver, the apparent K_m value in mice was lower than that in rats. In the case of other UGT1A6 substrates, the hepatic serotonin O-glucuronidation in Wistar rats was higher than that in CD-1 mice (Krishnaswamy et al., 2003), whereas the hepatic acetaminophen O-glucuronidation in Wistar rats was lower than that in CD-1 mice (Court, 2001).

Rat UGT1A9 is a pseudogene, but human UGT1A9 and mouse Ugt1a9 are not. The glucuronidation of propofol is catalyzed by human UGT1A8 (Mano et al., 2004) and by human UGT1A9 in the liver (Court, 2005) in a manner consistent with the substrate inhibition kinetics (Fujiwara et al., 2007). In the present study, the glucuronidation of propofol in liver microsomes from both male and female mice was fitted to the substrate inhibition kinetics, whereas those in other microsomes were not fit. These differences in the kinetic profile may be accounted for by the absence of UGT1A9 protein in rats. Propofol O-glucuronidation may be catalyzed by other UGTs, possibly UGT1A8, in rats.

MPA is the active metabolite of MMF, which induces gastrointestinal toxicity. Stern et al. (2007) reported that female rats were more susceptible to MMF-induced gastrointestinal toxicity than male rats, because of the fact that female rats showed lower intestinal MPA O-glucuronidation than male rats. In the present study, in duodenum the CL_int value for the MPA O-glucuronide formation in male mice was lower than that in male rats. This result suggests that male mice may be more susceptibility to MMF-induced gastrointestinal toxicity than male rats. Human UGT1A9 is mainly involved in hepatic MPA O-glucuronidation (Bernard and Guillemette, 2004). Picard et al. (2005) reported that in microsomes from humans the CL_int value in liver (28.7 μl/min/mg protein) was higher than that in intestine (0.7 μl/min/mg protein). In the present study in mice, the CL_int value in liver was higher than that in duodenum, but mouse Ugt catalyzing MPA O-glucuronidation has not been determined yet. Rat UGT1A7 is mainly involved in MPA O-glucuronidation (Miles et al., 2005, 2006). In the present study, the CL_int value in rats was similar in liver and duodenum and may be catalyzed by UGT1A7.

In humans, approximately 55% of morphine is metabolized into morphine 3-O-glucuronide and approximately 15% into morphine 6-O-glucuronide (Milne et al., 1996). The ratios of morphine 3-O-glucuronide to morphine 6-O-glucuronide in liver microsomes of mice, rats, guinea pigs, and rabbits were 300:1, 90:1, 4:1, and 40:1, respectively (Kuo et al., 1991). The formation of morphine 3-O-glucuronide is catalyzed by human UGT2B7 (Turgeon et al., 2001) and rat UGT2B1 (King et al., 2000). However, there is no information on the mouse Ugt enzyme that catalyzes morphine 3-O-glucuronidation. Rat UGT2B1 and mouse Ugt2b1 are predominantly expressed in the liver (Shelby et al., 2003; Buckley and Klaassen, 2007). In the present study, the morphine 3-O-glucuronidation was higher in the liver, compared with intestine, in both mice and rats as well as humans (Fisher et al., 2000b). The indication is that the glucuronidation of morphine in mice may occur mainly in liver as in rats and humans. In addition, in the present study, there were no sex differences in morphine 3-O-glucuronidation in either mice or rats, consistent with a report on the sex differences in morphine glucuronidation in vivo and in vitro (Rush et al., 1983). However, in the case of bisphenol A glucuronidation catalyzed by rat UGT2B1, the ratio of bisphenol A glucuronide to total bisphenol A in liver microsomes was significantly higher (P = 0.015) in female than in male Wistar-Imamichi rats, and the relative hepatic expression level of UGT2B1 mRNA was significantly higher (P < 0.001) in female than in male rats (Takeuchi et al., 2004). Therefore, when morphine 3-O-glucuronidation in rat liver is evaluated, the involvement of other UGTs as well as UGT2B1 may need to be considered.

TFP is a specific probe substrate for human UGT1A4 (Uchaipichat et al., 2006). However, rat UGT1A4 and mouse Ugt1a4 are known to be pseudogenes. In the present study, TFP N-glucuronidation in mice and rats could not be detected. Therefore, we should be careful in studies of drugs catalyzed by UGT1A4.

In the present study, the UGT activities in mouse and rat duodenum, jejunum, ileum, and colon were determined. The UGT activities in rats tended to decrease from duodenum to ileum, whereas this tendency in mice differed according to the substrates. These phenomena may be due to the expression levels of UGT or to the function of the UGT, but further study is needed. Strassburg et al. (2000) reported that the UGT activities in human intestine were higher in the jejunum than in duodenum or ileum. The present study clarified that this tendency in intestine may be different among mice, rats, and humans.

There are few reports concerning strain differences in UGT activities in mice. The UGT activities of five different substrates, except for 4-MU, showed the highest activities in C3H/HeJ mice. In contrast, the glucuronidation in DBA/2 mice was relatively low compared with that in the other strains. The strain differences in estradiol 3-O-, 4-NP O-, propofol O-, and MPA O-glucuronidation were similar. Why such strain differences were observed is not known exactly.

In conclusion, the present study clarified the fact that species differences exist between rats and mice in terms of their duodenal and hepatic UGT activities. The species, strain, and sex differences may depend on the substrate or UGT enzyme. The present study will provide useful information for the selection of species for in vivo UGT studies. Furthermore, experimental animals are useful tools for the development of new drugs, and thus studies on species differences are of great value for extrapolating results from animals to humans. The present study will also provide useful information for predicting drug metabolism catalyzed by UGTs.