Abstract
Mycophenolic acid (MPA) is the active immunosuppressive metabolite of the anti-organ rejection drug mycophenolate mofetil (MMF) and is implicated in the gastrointestinal toxicity associated with MMF therapy. Intestinal UDP-glucuronosyltransferases (UGT) have been proposed to provide intrinsic resistance against MMF-induced gastrointestinal toxicity by converting MPA to the inactive MPA 7-O-glucuronide. Using an optimized intestinal microsome preparation method that stabilized the intestinal MPA UGT activity, the MPA UGT activity of male Sprague-Dawley rat intestinal microsomes was characterized. A longitudinal gradient similar to that described for other phenolic compounds was observed, with the activity decreasing from the duodenum to the distal small intestine and colon. The catalytic efficiency of MPA glucuronidation decreased from the proximal to distal intestine as a result of decreasing Vmax and increasing Km. The finding that homozygous Gunn rats lack detectable intestinal MPA UGT activity indicates exclusive roles of UGT1A1, UGT1A6, and/or UGT1A7. Quantitative immunoblotting revealed a parallel between the MPA UGT activity and the content of UGT1A7-like immunoreactivity (18.7 and 7.3 μg/mg for duodenum and colon, respectively). In contrast, the lesser MPA-metabolizing UGT, UGT1A1 and UGT1A6, were lower in abundance (1.6–2.1 and 1.7–2.9 μg/mg, respectively), and their patterns of longitudinal distribution were distinct from the MPA UGT activity. These data suggest a dominant role of a UGT1A7-like enzyme, presumably UGT1A7 itself, in the catalysis of rat intestinal MPA glucuronidation. Studies are ongoing to investigate the relationship between intestinal UGT1A enzymes and susceptibility to MMF-induced gastrointestinal toxicity.
Conjugation with glucuronic acid represents one of the most important types of phase II drug biotransformation reactions (Dutton, 1980). The enzymes catalyzing this process are the UDP-glucuronosyltransferases (UGT), which consist of two multimembered families of UGT, UGT1 and UGT2 (Ritter, 2000). Whereas the liver and kidney are commonly viewed as the major sites of glucuronidation in the body, the intestine has also received considerable attention, with microsomal UGT activities approaching or exceeding the activity values for liver (for recent reviews, see Fisher et al., 2001; Tukey and Strassburg, 2001; Gregory et al., 2004). Intestinal UGT activity is likely to contribute to the exclusionary function of the gastrointestinal barrier, allowing the passage of nutrients while excluding other xenobiotics present in foods. For certain drugs taken p.o., intestinal UGT are also proposed to play a role in presystemic, “first-pass” drug elimination. Because glucuronide conjugates are usually less active toxicologically, they may also serve to protect the gastrointestinal mucosa from cytotoxic effects of ingested drugs and food substances.
Mycophenolate mofetil (MMF) (CellCept) is the prodrug form of mycophenolic acid [MPA, (1,3-dihydro-4-hydroxy-6-methoxy-7-methyl-3-oxo-5-isobenzylfuranyl)-4-methyl-4-hexenoate], an immunosuppressive agent widely used in transplant patients to prevent graft rejection. Use of the standard initial dose of MMF in transplant patients (1 g/b.i.d.) is associated with a high incidence of gastrointestinal toxicity (constipation and diarrhea with nausea and vomiting). The occurrence of the latter often necessitates dosage reduction, a consequence of which is reduced immunosuppressive efficacy (Busca et al., 2001). MPA is excreted primarily in the urine as its inactive 7-O-phenolic glucuronide (MPAG) (Bullingham et al., 1998). MPAG also undergoes biliary excretion, resulting in substantial enterohepatic cycling. The observation by multiple laboratories that intestinal microsome preparations are able to efficiently glucuronidate MPA (Bowalgaha and Miners, 2001; Shipkova et al., 2001; Bernard and Guillemette, 2004; Picard et al., 2005) raises the question of whether interindividual differences in intestinal MPA glucuronidation influence the onset or severity of MPA-induced gastrointestinal toxicity.
Establishment of a laboratory animal model for MPA intestinal toxicity would be a valuable step taken toward testing this hypothesis. The potential utility of the rodent model is supported by the observation that male Sprague-Dawley rats treated subchronically with MMF (doses >75 mg/kg/day for 3–6 days) develop diarrhea and pathologic signs of gastrointestinal toxicity (S. Stern and P. Smith, unpublished data). Although there are no published reports to our knowledge showing activity of rat intestine in MPA glucuronidation, the rat intestine has been reported to be highly active in the glucuronidation of a number of phenolic substrates (Hietanen et al., 1972; Hietanen and Vainio, 1973; Koster et al., 1985; Vargas and Franklin, 1997). A recent study by our laboratory revealed significant activities (>1 nmol/mg/min) of three rat recombinant UGT1A isoforms, UGT1A1, UGT1A6, and UGT1A7, toward MPA, and each of these is expressed at significant levels in rat intestine (Kobayashi et al., 1998; Grams et al., 2000; Shelby et al., 2003; Webb et al., 2005).
The major objective of this study was to show and characterize the microsomal MPA glucuronidating activities of rat small intestine and colon and to relate these activities to the levels of UGT1A1, UGT1A6, and UGT1A7-like protein found in these tissues. A secondary objective was to determine the absolute contents of these UGT in liver, kidney, and intestine microsomes through quantitative immunoblotting. Our study revealed a gradient of MPA UGT activity along the longitudinal axis of the rat intestine along with enzyme-specific patterns of expression. The marked proximal to distal gradient in MPA UGT activity was strongly correlated with the content of UGT1A7-like immunoreactivity.
Materials and Methods
Chemicals and Reagents. MPA, chymostatin (catalog no. C-7268), leupeptin (l-2023), aprotinin (A-1153), antipain (A-6191), pepstatin A (P-5318), phenylmethylsulfonyl fluoride (PMSF), and soybean trypsin inhibitor (T-2011) were from Sigma Chemical (St. Louis, MO). MPAG was from a preparation previously synthesized and described (Wiwattanawongsa et al., 2001). Solvents used for high-performance liquid chromatography (HPLC) analysis were purchased from Fisher Scientific (New Lawn, NJ). All the other chemicals used were purchased from Sigma and were the highest grade available. Generation and amplification of recombinant adenoviruses coding for rat UGT1A1, UGT1A6, or UGT1A7 under the control of cytomegalovirus promoter sequences were described previously (Miles et al., 2005).
Animals. Young adult male rats (6–8 weeks, 180–240 g body weight) were used for our studies. Sprague-Dawley rats were from Harlan Sprague-Dawley (Frederick, MD). Homozygous j/j rats or heterozygous j/+ Gunn rats were from a breeding colony established at Virginia Commonwealth University. Animals were housed two per cage and maintained under controlled conditions (standard 12-h light/dark cycle, 22°C, and 21% humidity). They were allowed free access to standard rodent chow (Harlan Teklad Lab Diet LM-485) and water. All the experiments using animals were conducted according to protocols approved by the Virginia Commonwealth University Institutional Animal Care and Use Committee using the National Institutes of Health Guidelines for Animal Care and Use.
Preparation of Intestinal Microsomes. After isoflurane anesthesia, rats were guillotined. The entire intestine from the pylorus to the distal colon was rapidly removed and placed on a bed of ice. All the subsequent steps in tissue handling and microsome preparation were performed at 4°C. The duodenum was defined as the first 10 cm of intestine distal to the stomach. The remainder of the small intestine extending to the cecum was divided into four equal segments, which were designated as S-1, S-2, S-3, and S-4, with S-1 being most proximal and S-4 most distal. The colon corresponded to the 10-cm segment immediately distal to the cecum. All the segments were flushed with ice-cold isotonic phosphate-buffered saline containing 0.25 mM PMSF to remove intestinal contents. Mucosal epithelial cells were obtained by slitting the tissue longitudinally, scraping the epithelial, luminal side with a glass microscope slide, and collecting the scrapings in 30 ml of ice-cold 0.25 M sucrose containing protease inhibitors as indicated in Table 1. In two cases (protocol C and protocol F), the scrape step was omitted, and the intestinal segment samples were minced with scissors before homogenization. The following protease inhibitor stocks were prepared and stored at –20°C between use: PMSF (0.1 M in ethanol), EDTA (0.5 M in water), leupeptin (5 mg/ml in water), and 250× protease inhibitor mixture (25 μg/ml pepstatin A, 25 μg/ml chymostatin, 25 μg/ml antipain, 25 μg/ml leupeptin, and 10 μg/ml aprotinin in 1% dimethyl sulfoxide). Soybean trypsin inhibitor (50 mg/ml in phosphate-buffered saline) was prepared fresh for each experiment. Samples were homogenized in Potter-Elvehjem glass tubes using a drill-driven, serrated Teflon pestle (four passes) and then centrifuged at 18,800g for 15 min at 4°C. The supernatants were filtered through a loose cotton plug into a 30-ml ultracentrifuge tube on ice and recentrifuged at 100,000g for 30 min at 4°C. The microsome pellets were resuspended in 0.1 M potassium phosphate buffer (pH 7.5) containing 1 mM EDTA, 20% glycerol, and the indicated protease inhibitors (Table 1). Microsomes from rat liver and kidney were prepared as described previously (Kessler and Ritter, 1997). Microsomal protein concentration was estimated by the bicinchoninic acid method using a commercially available kit (Pierce Biochemical, Rockford, IL).
In Vitro MPA Glucuronidation Assays. MPA UGT reactions (50 μl) contained 75 mM Tris-Cl, pH 7.45, 10 mM MgCl2, 6 mM saccharic acid 1,4-lactone, 50 μg alamethicin/mg protein, 1 mM MPA, and 1 mg/ml microsomal protein. Reactions were preincubated at 37°C for 3 min and were started by the addition of UDP-glucuronic acid (3 mM final concentration). After 30 min at 37°C, the reactions were stopped by addition of 70% perchloric acid (5.6% final concentration). Tubes were placed on ice for 5 min and then centrifuged at 16,000g for 5 min. A 20-μl aliquot was then analyzed by reverse-phase HPLC as described previously (Wiwattanawongsa et al., 2001). The HPLC system consisted of a Hewlett Packard 1050 HPLC with a Partisil 10 ODS-2 C18 column (4.6 mm × 25 cm) (Whatman Inc., Clifton, NJ) and a Waters Resolve C18 Guard Pak precolumn cartridge (Millipore Corporation, Billerica, MA). The mobile phase was 55% MeOH/45% aqueous trifluoroacetic acid (0.1%). The flow rate was 1.5 ml/min. Absorbance was monitored with a Hewlett Packard 1050 diode array detector set at 250 nm. Under our conditions, the retention times for MPAG, the suprofen internal standard, and MPA were 3.3, 8.8, and 11.8 min, respectively. MPAG was quantitated using standard curves prepared using blank microsome reactions spiked with known amounts of authentic MPAG standard. Care was taken to perform all the reactions under conditions when MPAG formation was linearly proportional to time and protein concentration. The limit of detectable activity of our assay was ∼0.010 nmol/mg/min.
Kinetic Analysis of MPAKm andVmax. Michaelis-Menten enzyme kinetic parameters were determined by assessing the effect of MPA concentration (varied between 0.01 and 1 mM) on the rate of MPAG formation. Concentrations >1 mM were found to inhibit the MPAG reaction. The kinetic constants Km and Vmax were estimated by nonlinear regression using WinNonlin pharmacokinetic software v. 2.0 (Pharsight Corporation, Palo Alto, CA).
Preparations of Recombinant Expressed Rat UGT. HepG2 cells were grown to 90 to 95% confluence in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, at which time individual plates were infected with adenovirus lysates encoding the UGT indicated. Cells were harvested by gentle pipetting to release cells 72 h after infection and centrifuging at 5000 rpm in a clinical centrifuge at 4°C. Supernatants were discarded, and the resulting cell pellets were washed by resuspension in ice-cold 1× phosphate-buffered saline and recentrifugation. The washed cell pellets were resuspended in 0.1 M potassium phosphate buffer, pH 7.5, 20% glycerol, and 1 mM EDTA. After three consecutive freeze/thaw cycles and a brief sonication to disperse aggregates, the cells were centrifuged at 5000g for 10 min. The supernatants were collected and transferred to microfuge tubes. Protein concentration was determined as described above. Membrane preparations were aliquoted and stored at –80°C until assay.
Immunoblot Analysis of UGT1A1, UGT1A6, UGT1A7-Like, and Total UGT1A Protein. Relative and absolute amounts of UGT1A proteins were determined using an immunoblotting procedure described previously (Ritter et al., 1999; Kessler et al., 2002) with some modifications. Microsomal protein (30 μg) was electrophoresed in parallel with serial dilutions of the desired recombinant expressed rat UGT on a 7.5% SDS-polyacrylamide gel and transferred onto a nitrocellulose membrane. The blots were probed using antisera raised in mice against bacterial-expressed UGT fusion proteins. The latter represent amino acid residues 29 to 162 (UGT1A1), 77 to 183 (UGT1A6), 21 to 149 (UGT1A7), and 360 to 455 (rat UGT1A common region), respectively. The specificities of these antisera have been established (Kessler et al., 2002; Webb et al., 2005, 2006). Whereas the UGT1A1 and UGT1A6 antisera exhibit high specificity for their respective enzymes, the UGT1A7 antiserum cross-reacts weakly with UGT1A8 (Kessler et al., 2002), and the immunoreactivity with this antiserum is therefore designated as being UGT1A7-like. The rat UGT1A common region antibody shows complete specificity for UGT1A versus UGT2 enzymes (Webb et al., 2006). For quantitative immunoblotting, a previously described procedure was used (Kessler et al., 2002). Proteins were detected using enhanced chemiluminescence (Amersham Biosciences, Piscataway, NJ). Images were captured on film and quantified using Image J (freeware available at http://rsb.info.nih.gov/ij/). All the calculations were based on recombinant rat UGT standards, which were used to generate standard curves.
Data Analysis. Data were analyzed by analysis of variance testing using Sigma Stat v. 2.03 software (SPSS Inc., Chicago, IL). When differences between groups were found, a Student-Newman Keuls post hoc test was used to identify specific groups showing differences. A p value < 0.05 was considered to be statistically significant. Correlation analyses were performed using Spearman's rank order correlation test.
Results
Evaluation of Protease Inhibitors Used in the Preparation of Intestinal Microsomes. Comparison of intestinal microsome glucuronidation data between different laboratories is complicated by the lability of intestinal UGT activities, a problem stemming from the contamination of intestinal preparations with proteases. Although the inclusion of protease inhibitors has been standard in protocols for preparing intestinal microsomes, both the types and amounts of protease inhibitors used have varied among laboratories. Therefore, a systematic study was performed to compare the efficacies of different protease inhibitors and protease inhibitor combinations, including two of the standard protease inhibitors described in the literature, PMSF and soybean trypsin inhibitor (Table 1). Protocols B and C, which included soybean trypsin inhibitor at 5 mg/g wet weight, yielded higher MPAG formation rates (2.4–3.1 nmol/min/mg) compared with protocols lacking soybean trypsin inhibitor (Fig. 1A). Reduced proteolytic digestion of UGT (detected either by the loss of intact UGT protein or the appearance of immunoreactive degradation products) was also evident in immunoblot analyses of the samples prepared using soybean trypsin inhibitor (lanes B and C of Fig. 1B). These data suggested that the methods using the combination of soybean trypsin inhibitor with PMSF (protocols B and C) were superior to those relying on PMSF (protocols D and E).
However, even in the samples prepared using the combination of soybean trypsin inhibitor and PMSF, a problem with stability of the MPA UGT activity was evident. Loss of MPA UGT activity was observed on freezing and thawing of the samples or by storing the microsomes on ice overnight. Representative results are shown in Fig. 2 (top) for duodenum and S-1 microsomes prepared using protocol C. The duodenal microsomes exhibited a 10% decrease in the activity after overnight incubation on ice, whereas >20% was lost by a single freeze/thaw cycle. S-1 microsomes showed a more dramatic effect with an 80% reduction in MPA UGT activity after either overnight incubation on ice or a single freeze/thaw cycle. The observed instability of MPA UGT activity of S-1 microsomes was also evident over the time course of a 30-min reaction at 37°C (Fig. 2, bottom). The susceptibility to inactivation by freezing and thawing and incubations on ice for extended periods varied between segments. It appeared greatest for microsomes prepared from upper small intestine segments, especially S-1 and S-2. Duodenum and S-3 showed an intermediate degree of instability, whereas S-4 and colon appeared refractory (data not shown). These data overall suggested that microsomes prepared from the small intestine of rats have residual protease contamination and that the more proximal segments have the greatest contamination.
To address the lability issue, a combination protease method was tested, which included PMSF, soybean trypsin inhibitor, and a more concentrated protease inhibitor mixture containing aprotinin, pepstatin A, chymostatin, leupeptin, and antipain (see protocol F in Table 1). Representative results are presented in Fig. 2 (middle). Microsomal MPA UGT activities from duodenum, S-1, S-2, and S-3 were completely stabilized using this combination of protease inhibitors (data shown only for duodenum and S-1). No effect of either freezing/thawing (up to 5 cycles) or time of incubation on ice was evident in any of the samples. Moreover, the time course of MPAG formation catalyzed by S-1 microsomes prepared by this protocol was linear.
Distribution of MPA UGT Activity along the Small Intestine and Colon. Using the optimized intestinal microsome preparation method (protocol F), we examined the distribution of MPA UGT activity along the male Sprague-Dawley rat small intestine (Fig. 3). Chromatograms of intestinal microsome incubations showed only a single apparent MPA glucuronidation product, which coeluted with authentic MPAG (data not shown). MPAG formation rates showed a longitudinal gradient along the gastrointestinal tract, with the highest activities observed toward the upper end of the small intestine (including duodenum, S-1, and S-2) and progressively lower activities in the lower segments of small intestine and the colon. In comparison, kidney microsomes had intermediate MPA UGT activity, and liver microsomes had high activity, comparable with S-1 intestinal microsomes (8–12 nmol/mg/min rates).
The kinetic properties of MPA UGT activity along small intestine were characterized in experiments that determined the influence of MPA concentration (0.0156–1 mM) on the MPAG formation rate. All the microsomes exhibited concentration-dependent glucuronidation that fit the standard Michaelis-Menten model as assessed using WinNonlin enzyme kinetics modeling software. Average Vmax and Km values determined for small intestine segments and colon of three male rats are presented in Table 2. For comparison, the Vmax and Km for rat liver microsomal MPA UGT were 6.7 ± 1.1 nmol/mg/min and 0.20 ± 0.02 mM, respectively (Miles et al., 2005). Each parameter varied along the intestine, with Vmax decreasing from duodenum to colon in a manner paralleling the MPA UGT activities determined at 1 mM MPA (Fig. 3), and Km increasing significantly down the length of the small intestine. Because of the reciprocal nature of the changes, the Vmax/Km ratio (an index of intrinsic clearance) decreased markedly along the small intestine, from a high of 180 in the duodenum to a low of 16 in the S-3 segment. Colon also exhibited a low Vmax/Km ratio.
The MPA UGT activities of heterozygous and homozygous male Gunn rats (genotypes j/+ and j/j, respectively) were also evaluated to assess the role of UGT1A enzymes in the intestinal activity. No activity was detected with any of the intestine or colon samples prepared from homozygous Gunn rats (<0.03 nmol/mg/min). HPLC chromatograms illustrating this effect using duodenum microsomes are shown in Fig. 4. The corresponding rate data are presented in Fig. 5 (left). Samples from heterozygote (j/+) rats exhibited activities intermediate between Sprague-Dawley and homozygous Gunn rats. A Spearman rank order correlation test revealed a highly significant correlation between genotype and intestinal MPA UGT activity (Fig. 5, right). These data indicate that the MPA UGT activity in rat small intestine is constituted completely by UGT1A family enzymes with no UGT2B contribution. Because our previous study revealed MPAG-forming activity of only three rat UGT1A family enzymes, i.e., the UGT1A1, UGT1A6, and UGT1A7 forms (Miles et al., 2005), the results suggest that one or more of these three enzymes are responsible for the intestinal MPAG activity.
To further evaluate this question, quantitative immunoblotting was conducted to assess differences in the absolute content of UGT1A1, UGT1A6, and UGT1A7-like enzymes in rat intestine, as well as kidney and liver microsomes. For quantitative Western blot analysis, our approach extended a method described previously by our laboratory for estimating the UGT1A7-like content of rat liver microsomes (Kessler et al., 2002). Application of the method allowed for the determination of UGT1A7 in a preparation of UGT1A7 in recombinant adenovirus-infected HepG2 cells (34 μg/mg, data not shown). This standardized preparation of HepG2-r1A7 was subsequently used to estimate the contents of UGT1A1 and UGT1A6 in two other preparations of HepG2 expressing rat UGT1A1 and UGT1A6. Each was normalized to the HepG2-r1A7 preparation using a rat UGT1A common region antiserum (r1CR) generated in this laboratory, which is assumed to recognize all the UGT1A enzymes with equal efficiency. The approach of using a UGT1A common region to assess relative levels of UGT1A enzyme expression has been widely used by investigators in the glucuronidation field (Gagne et al., 2002; Basu et al., 2004). The content of UGT1A1 and UGT1A6 in microsomes prepared from HepG2 cells infected for 72 h with adenoviruses expressing rat UGT1A1 and UGT1A6, respectively, was determined to be 36 and 21 μg/mg, respectively (data not shown).
Using the recombinant rat UGT-expressing HepG2 cell microsomes as standards in conjunction with the antisera to rat UGT1A1, UGT1A6, and UGT1A7, quantitative immunoblotting was applied toward analysis of microsomes prepared from different tissues of male Sprague-Dawley rats (n = 3). The cumulative quantitative data are presented in Fig. 6. Representative immunoblots for the intestine samples (excluding the recombinant UGT standards) are shown for each antiserum. The major UGT expressed in intestine was a UGT1A7-like enzyme. Its content declined sharply along the intestine, from a peak of 23.8 μg/mg in S-1 microsomes to a low value of 7.3 μg/mg in colon (p < 0.05). The UGT1A7-like content in kidney and liver microsomes was severalfold lower than colon. The UGT1A1 content was more even along the duodenum, S-1, S-2, and S-3 with a slight elevation apparent in S-4 and colon (not statistically significant). In contrast to the UGT1A7-like expression profile, the kidney and liver content of UGT1A1 were much higher than that of intestine. A third distinct pattern of expression was observed for the UGT1A6 enzyme. The UGT1A6 content was roughly equal across all the tissues. Although slightly higher levels were evident in the colon, liver, and kidney, the UGT1A6 content was not found to differ statistically.
Given the apparent correspondence between UGT1A7-like protein content and MPA UGT activity, the data were also analyzed by the Spearman rank order correlation test (Fig. 7). The MPA UGT activity of rat intestine was highly correlated with UGT1A7-like content (p = 0.001), whereas no correlation was evident with the content of either UGT1A1 (p = 0.86) or UGT1A6 (p = 0.84).
Discussion
This study characterized the MPA UGT activity of rat intestine, which may act to modulate the gastrointestinal toxicity associated the use of MMF. Several findings of interest were uncovered, including the demonstration of high catalytic activity of rat intestinal microsomes toward MPA, the presence of a distinct distribution pattern in the activity as well as in kinetic properties for MPA glucuronidation (Vmax, Km, and Vmax/Km ratio), the strong parallel observed between MPAG formation rates and the content of a UGT1A enzyme detected with a UGT1A7-selective antibody (presumably UGT1A7 itself), and the finding at the protein level that not all the UGT isoforms are distributed in the same pattern along the gastrointestinal tract.
The demonstration of high catalytic activity toward MPA, a phenolic substrate, is consistent with data from early studies indicating that rat intestinal microsomes are quite active (>1 nmol/mg/min) in the glucuronidation of substrates with phenolic -OH groups, such as o-aminophenol, 4-nitrophenol, 1-naphthol and 2-naphthol, 4-methyl-umbelliferone, and morphine (Hietanen et al., 1972; Hietanen and Vainio, 1973; Koster et al., 1985; Vargas and Franklin, 1997). Moreover, several of these studies examined the distribution of the phenol glucuronidating activities along the gastrointestinal tract. Although direct comparisons are difficult because of differences in the intestinal microsome preparation methods and the way in which the intestine was divided into segments, a common feature of these reports is the observation of a gradual decline in UGT activity from the duodenum to the large bowel. This decline in activity was also apparent in our study for MPA.
The characterization of kinetic parameters for MPA glucuronidation along the intestine was a unique aspect of this study. The kinetic analysis indicates that the decline in MPA UGT activity observed under standard MPA UGT reaction conditions (including 1 mM MPA and 3 mM UDP-glucuronic acid, see Fig. 3) was attributable to changes in both Vmax and Km. The progressive decrease in Vmax observed between the duodenum, S-2, and colon (Table 2) suggests that the content of an MPA UGT (or several) is reduced as one proceeds along the intestine. The association of the effect with an increase in Km (decreased affinity) of the MPA UGT(s) for MPA may be explained by multiple UGT contributing to MPA catalysis. These UGT would have different Vmax and Km for MPA glucuronidation, and the different ratios of Vmax/Km would be a consequence of differential changes in their expression along the length of the intestine.
Because Gunn (j/j) rat intestinal microsomes exhibited no detectable MPA UGT activity, the data show that UGT1A family enzymes catalyze all the MPA UGT activity and that UGT2B enzymes have an insignificant role in naive rat intestine. This result contrasts with the MPA UGT activity of rat liver, which has a small but significant UGT2B component (estimated to be between 10 and 20% of the total activity measured under standard reaction conditions). During screens of recombinant rat UGT1A and UGT2B enzymes for catalytic activity toward MPA (Miles et al., 2005), three active enzymes were identified: UGT1A7, a high affinity-high capacity MPA-metabolizing enzyme (Vmax = 10.3 nmol/mg/min and Km = 0.034 mM); UGT1A1, an intermediate affinity and capacity enzyme (Vmax = 1.7 nmol/mg/min and Km = 0.21 mM); and UGT1A6, a low affinity/intermediate capacity enzyme (Vmax = 1.3 nmol/mg/min and Km = 1 mM). Results from UGT1A mRNA analyses show detectable expression of each of these enzymes in rat intestine, raising the question about their relative quantitative roles in the intestinal catalysis of MPA glucuronidation.
It seems likely that UGT1A7 is the major enzyme mediating rat intestinal MPA glucuronidation, particularly in the upper small intestine if not also the more distal segments and the colon. One line of evidence is the correlation observed between MPAG formation rate and the UGT immunoreactivity detected with a UGT1A7-selective polyclonal antiserum. Although we cannot directly exclude the possibility, it does not appear that rat UGT1A8 and UGT1A10 are significant contributors to the UGT1A7-like immunoreactivity of intestine based on several recent studies showing high abundance of the UGT1A7 transcript in rat intestine (Kobayashi et al., 1998; Grams et al., 2000; Shelby et al., 2003; Webb et al., 2005). In contrast, Shelby et al. (2003) reported detectable expression of the UGT1A8 mRNA in liver, kidney, and colon but not small intestine. The same pattern of low UGT1A8 mRNA expression in liver and kidney and undetectable expression in intestine was confirmed in a recent study by our laboratory using an alternative approach (RNase protection analysis) (Webb et al., 2005). A parallel analysis of UGT1A10 mRNA in the same study (data not shown) revealed a profile similar to UGT1A8, with low levels in liver and kidney and undetectable levels in intestine.
The current report is the first to our knowledge to provide quantitative estimates of the UGT content of rat liver, kidney, and intestinal microsomes, as well as to describe the longitudinal distribution of specific UGT enzymes along the rat gastrointestinal tract. Of the three enzymes studied, UGT1A7 was quantitatively the most abundant UGT in the intestine with much lower levels in the liver and kidney, and it was the only one of the three enzymes active toward MPA to show a marked longitudinal gradient in expression. Two different studies have evaluated the distribution of UGT1A7 mRNA along the rat gastrointestinal tract and its tissue specificity (Grams et al., 2000; Shelby et al., 2003). Both studies observed declines in UGT1A7 expression along the length of the intestine, which is consistent with our protein level data. In addition, both studies noted more abundant expression of UGT1A7 in the intestine versus the kidney and liver. Taken together, the data suggest that the mechanism underlying the proximal to distal gradient in UGT1A7 activity is transcriptional in origin.
This study establishes for the first time the presence of UGT1A1 and UGT1A6 protein in the rat gastrointestinal tract. However, their levels were much lower than that of UGT1A7-like enzyme, and each showed lower variation along the intestine. A strong contrast was also evident in their expression across liver, kidney, and intestine. UGT1A1 exhibited a much higher level in liver compared with kidney and intestine, whereas UGT1A6 was more evenly expressed across all three tissues. Our data suggesting the absence of a significant gradient of UGT1A1 and UGT1A6 protein along the gastrointestinal tract is in general agreement with the results of mRNA analysis (Grams et al., 2000; Shelby et al., 2003), with neither study noting any evidence for any longitudinal variation in their expression. However, our finding of much higher UGT1A1 abundance in liver versus intestine contrasts with the mRNA analysis data (Shelby et al., 2003), which indicated fairly even expression of UGT1A1 mRNA across these tissues (2-fold range in variation). The reason for this difference remains unclear but may be explained by tissue-specific differences in the efficiencies of UGT1A1 mRNA translation or protein degradation.
Overall, our results suggest the following model for MPA glucuronidation catalyzed by rat intestine. MPA glucuronidation is highest in the duodenum and other proximal segments of rat intestine, an effect related to the high expression in these tissues of UGT1A7, a high affinity/high MPA-metabolizing enzyme. UGT1A1 and UGT1A6 are also predicted to contribute to the total activity in the upper intestine, but quantitatively their influence is minor. In the more distal small intestine, as UGT1A7 enzyme levels decline, the efficiency of MPA catalysis decreases. The observed increase in Km of the lower rat small intestine and colon for MPA UGT may be caused by an increased relative contribution of UGT1A1 and UGT1A6, low affinity MPA-metabolizing UGT. This interpretation is supported by the upward shift in the Km of rat liver microsomes for MPA (0.2 mM). In liver, the UGT1A1 and UGT1A6 forms are more abundant than UGT1A7 (Fig. 6).
It is of interest to consider the relationship of our rat intestinal UGT expression data to human. Although intestinal microsomes from both species catalyze MPA glucuronidation at high rates (Bowalgaha and Miners, 2001; Shipkova et al., 2001; Bernard and Guillemette, 2004; Picard et al., 2005), differences are likely to exist in the UGT responsible for MPA catalysis in rat versus human intestine. With human intestine, MPA UGT activity is unlikely to be dominated by UGT1A7, which exhibits low expression in human intestine and colon (Tukey and Strassburg, 2001). However, it does appear that the intestinal MPA UGT activity in the two species shares the characteristic of being dominated by UGT1A enzymes in the “bulky phenol” UGT1A7 through UGT1A10 subfamily. Human UGT1A8 and UGT1A10 are expressed in human intestine and are active in MPA catalysis (Mojarrabi and Mackenzie, 1997; Mackenzie, 2000).
In summary, the current study enhances our understanding of intestinal MPA glucuronidation in rats. This knowledge will be useful for future studies using the rat model to assess the influence of intestinal UGT expression on MPA-induced gastrointestinal toxicity.
Footnotes
-
This work was supported in part by National Institutes of Health Grants 1R01-GM61188 and 5R01-ES07762 from the National Institutes of General Medical Sciences and Environmental Health Sciences, respectively.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.106.010140.
-
ABBREVIATIONS: UGT, UDP-glucuronosyltransferase(s); MMF, mycophenolate mofetil; MPA, mycophenolic acid; MPAG, mycophenolic acid 7-O-glucuronide; PMSF, phenylmethylsulfonyl fluoride; HPLC, high-performance liquid chromatography.
- Received March 9, 2006.
- Accepted June 20, 2006.
- The American Society for Pharmacology and Experimental Therapeutics