Abstract
The dietary polyphenol resveratrol (RES) exists as cis- and trans-isomers with known stereospecific and stereoselective glucuronidation at the 3 and 4′ positions by distinct UGT1A isoforms. We examined cis-RES glucuronidation in various protein sources. UGT1A6 or UGT1A1 genotype-dependent cis-or trans-RES glucuronidation, respectively, was further determined. cis-RES exhibited partial substrate inhibition in UGT1A6 Supersomes and human embryonic kidney 293 cells overexpressing genetically variant UGT1A6 alleles. Cells expressing UGT1A6*4 had the highest activity with a Vmax of 612 ± 27.36 nmol/min/mg, followed by UGT1A6*3. The *2 allozyme had a higher Vmax (1.6-fold) and Km (1.9-fold) than *1. In 51 human liver samples genotyped for UGT1A6, four alleles (frequencies) were identified as *1 (0.58), *2 (0.36), *3 (0.01), and *4 (0.05), leading to assignment of the following genotypes (frequencies): *1/*1 (0.29), *1/*2 (0.45), *1/*3 (0.02), *1/*4 (0.10), and *2/*2 (0.14). Up to 5-fold variability in trans-RES glucuronidation was observed in individual liver samples. In livers stratified by UGT1A6 genotype, a significant difference in cis-RES glucuronidation activity (p < 0.05) was seen between the *2 variants compared with homozygous *1 livers. The trans-RES glucuronidation was quantitated in a human liver bank genotyped for the UGT1A1 TATA box repeat polymorphism. There was no significant difference for formation of trans-RES 3-O-glucuronide. We were surprised to find that trans-RES 4′-O-glucuronide formation was higher in livers with the 7/7 genotype compared with 6/6 and 6/7 (p < 0.05). In conclusion, cis-RES glucuronidation exhibited atypical partial substrate inhibition kinetics in vitro. Whereas cis-RES glucuronidation varied with UGT1A6 genotypes, the UGT1A1*28 polymorphism did not explain variability in trans-RES glucuronidation.
Resveratrol (3,4′,5-trihydroxystilbene, RES) is a polyphenol of plant origin found in grapes, peanuts, and red wine as the cis- and trans-isomers. RES is known to be both glucuronidated and sulfated. RES glucuronidation in humans is both regioselective and stereospecific (Aumont et al., 2001) and is mediated by members of the UDP glucuronosyltransferase (UGT) superfamily of enzymes. Several UGT isoforms exist, and their distinct but sometimes overlapping substrate specificity has been shown (Tukey and Strassburg, 2000). Their organ- and tissue-specific expression and variant polymorphic nature have also been documented.
The major UGT isoforms involved in RES metabolism are UGT1A1, UGT1A9, and UGT1A6 with minor contributions by UGT1A10, UGT1A7, UGT1A8, and UGT1A3. UGT1A1 catalyzes the glucuronidation of trans-RES preferentially at the 3 position, yielding the major 3-O-glucuronide (R3G) metabolite, and less so at the 4′ position to give the minor metabolite [trans-RES 4′-O-glucuronide (R4′G)]. UGT1A6 selectively glucuronidates the cis-isomer of RES at the 3 position to give the major metabolite [cis-RES 3-O-glucuronide (cis-R3G)], whereas UGT1A9 has been shown to glucuronidate both isomers (Aumont et al., 2001; Sabolovic et al., 2006). The relative contribution of UGT1A9 and UGT1A6 to cis-RES glucuronidation is not clearly established. Reports from assays conducted in recombinant Supersomes were equivocal and did not account for relative tissue expression of the two proteins.
Genetic polymorphisms have been reported in almost all the members of the UGT family, ranging from single nucleotide (exonic and intronic) to promoter polymorphisms (Guillemette, 2003; Argikar et al., 2008). The functional significance and genotype-phenotype correlation of these UGT polymorphisms is an ongoing area of research. Distinct polymorphisms in the two unique major isoforms involved in metabolism of RES are of interest in this study. The UGT1A1 polymorphism (UGT1A1*28) involves an additional (TA) dinucleotide repeat in the “TATA” box [(TA)6 > (TA)7] of the UGT1A1 promoter. This polymorphism has been linked to decreased expression of UGT1A1 and consequently reduced hepatic clearance of bilirubin and other xenobiotics (Iyer et al., 1999, 2002; Peters et al., 2003; Fang and Lazarus, 2004). UGT1A6 polymorphisms include three nonsynonymous coding single nucleotide polymorphisms (SNPs) found in exon 1—19T>G, 541A>G, and 552A>C—which encode the amino acid changes S7A, T181A, and R184S and yield four UGT1A6 alleles. The functional impact of these three polymorphisms on various UGT1A6 substrates has been studied with varying results, possibly because of differences among the studies in defining the alleles (Ciotti et al., 1997; Nagar et al., 2004b; Krishnaswamy et al., 2005).
The abundant expression of these two UGT1A isoforms (UGT1A1 and UGT1A6) in the human liver makes genotype-phenotype correlation studies with regard to RES possible using genotyped liver microsomes. RES has been shown to be highly glucuronidated in the human liver, a process that has been suggested to limit its oral bioavailability (Walle et al., 2004) and pharmacologic effects. Most pharmacologic studies on RES have been on the trans-isomer possibly because of its relative abundance and commercial availability (Campos-Toimil et al., 2007). Studies that have compared the effects of both isomers have found minor quantitative differences in the magnitude of their effects (Leiro et al., 2004; Orallo, 2006). Despite its beneficial effects in vitro, the very low bioavailability of RES compared with its high absorption profile from the intestine (Walle et al., 2001; Wenzel et al., 2005) raises questions on where activity resides. It is possible that activity may reside with the parent compound(s) or their metabolites. If the latter is the case, then studies on the functional impact of genetic polymorphisms in enzymes catalyzing the formation of these metabolites are warranted. Such pharmacogenetic studies become relevant given that genetic variability may be one of the contributing factors to differences in exposure and therefore effectiveness of dietary phytochemicals in populations.
We previously characterized trans-RES glucuronidation in relevant enzyme sources across a wide concentration range as higher dose requirements had been suggested to account for its low bioavailability (Goldberg et al., 2003; Iwuchukwu and Nagar, 2008). However, the effects of other factors such as UGT pharmacogenetics on RES metabolism cannot be ruled out. To our knowledge, UGT genotype-dependent RES glucuronidation has not been evaluated to date. In the light of the above, we undertook the genotype-phenotype correlation of two of the major isoforms involved in the glucuronidation of the RES isomers: UGT1A1 for trans-RES and UGT1A6 for cis-RES. The goal of these studies is to add to the body of work regarding RES by serving as a foundation for future studies on effects of genetic polymorphisms on phytochemical metabolism.
Materials and Methods
Chemicals and Reagents. RES (trans-RES, purity ≥99%), the cofactor UDP-glucuronic acid, dimethyl sulfoxide (DMSO), alamethicin, acetaminophen, and chlorzoxazone were purchased from Sigma-Aldrich (St. Louis, MO). cis-RES (purity ≥98%) was purchased from Cayman Chemical (Ann Arbor, MI). All the chemicals used to prepare buffers and other reagents were of analytical grade. Pooled human liver microsomes and Supersomes expressing the recombinant human UGT1A1 and UGT1A6 enzymes were obtained from BD Gentest (BD Biosciences, San Jose, CA).
NuSieve agarose and Gel Star Nucleic Acid Stain were obtained from Lonza Group Ltd. (Basel, Switzerland). Restriction endonucleases and buffers were purchased from New England Biolabs (Ipswich, MA). UGT1A6 antibody was obtained from BD Biosciences (San Jose, CA). Cell culture reagents Dulbecco's modified Eagle's medium, penicillin/streptomycin, and fetal bovine serum were obtained from Mediatech (Manassas, VA). The UGT1A6-specific oligonucleotide polymerase chain reaction (PCR) primers were synthesized by Operon Biotechnologies (Huntsville, AL). DNA step ladder and PCR Master Mix (solution containing Taq DNA polymerase, dNTPs, MgCl2, and reaction buffers) were ordered from Promega (Madison, WI). Tris-EDTA, nuclease-free water, and other molecular biology grade reagents and solvents were purchased from Thermo Fisher Scientific Inc. (Waltham, MA).
DNA Isolation and UGT1A6 Genotyping Assay. Human liver samples (67) were purchased from the Fox Chase Cancer Center Tumor Bank Facility (Philadelphia, PA) and consisted of surgically excised tissue from nonmalignant patients and patients with colorectal cancer that had metastasized to the liver. All the samples were deidentified with respect to age, ethnicity, and gender. DNA was extracted from human liver tissue (30 mg) using the Promega Wizard SV Genomic DNA Purification System by incubating overnight in 275 μl of digestion solution, after which samples were spun down and supernatant was transferred to fresh DNase/RNase-free tubes for purification. Purification was carried out by adding 250 μl of Nuclei Lysis solution, transferring to a minicolumn, spinning at 13,000 relative centrifugal force for 3 min, and discarding waste from collection tubes. The lysate was washed four times with 650 μl of ethanol wash buffer and spun at 13,000 relative centrifugal force for 1 min each wash. After the last wash, DNA was eluted using nuclease-free water heated to 65°C. Concentrations were determined on a spectrophotometer using the SoftMax software (Molecular Devices, Sunnyvale, CA). Purified DNA samples were resuspended in sterile water and stored at 4°C for further use.
A modification of the PCR-restriction fragment length polymorphism assay described by Nagar et al. (2004b) was used to detect the 19T>G, 541A>G, and 552A>C cSNPs in the UGT1A6 gene. In brief, 200 to 400 ng of genomic DNA was mixed with 25 μl of Promega Master Mix, and 2 μl each of the forward and reverse primers was used to characterize the 19T>G SNP [F(–53) and R184 yielding amplicon 1] and the 541A>G and 552A>C SNPs [F414 and R628 yielding amplicon 2]. Nuclease-free water was used to make up to a final volume of 50 μl. PCR amplification was carried out for 45 s at 94°C followed by 34 cycles of 94°C for 30 s, 56°C for 30 s, 72°C for 50 s, and finally 3 min at 72°C. PCR products were digested as follows: HhaI for amplicon 1 and NsiI and BbvI separately for amplicon 2. Digested products were then subjected to electrophoresis on 2% NuSieve agarose gels at 100 V for 1 h. The primer sequences used were as follows: F(–53), 5′-GAT TTG GAG AGT GAA AAC TCT TT-3′ and R184, 5′-CAG GCA CCA CCA CTA CAA TCT C-3′ for amplicon 1, and F414, 5′-CTT TAA GGA GAG CAA GTT TGA TG-3′ and R628, 5′-CCA CTC GTT GGG AAA AAG TC-3′ for amplicon 2. Of the 67 samples, 51 samples were assigned clear UGT1A6 genotypes. Genotyping for the UGT1A1 promoter polymorphisms was carried out at the Fox Chase Cancer Center Genotyping Facility with a high-throughput GeneScan assay (Carlini et al., 2005). All 67 samples were successfully assigned UGT1A1 genotypes.
Preparation of Enzyme Fractions.Human liver microsomes. Microsomes were prepared from human liver tissue samples (30 mg) previously stored at –80°C by modification of a standard method (Franklin and Estabrook, 1971). Samples were homogenized in 1 ml of microsome isolation buffer using a Potter-Elvehjem homogenizer, and low-speed centrifugation (20 min at 9000g) was used to clear cellular debris. Ultracentrifugation of the 9000g supernatants using a type 50TI rotor in an L5-40 Beckman Ultracentrifuge (Beckman Coulter, Fullerton, CA) for 60 min at 100,000g yielded microsomal pellets that were then resuspended in isolation buffer (50 mM Tris-HCl, pH 7.5, 10 mM KCl). Aliquots were then kept at –80°C for further use.
Culture and generation of cell homogenates. Human embryonic kidney carcinoma cells (HEK293) were previously engineered to stably express UGT1A6*1, *2, *3, or *4 allozymes (Nagar et al., 2004b). Different clones were generated for each allozyme, and individual clones were selected based on comparable UGT1A6 expression levels. Cells were thawed and cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum albumin, 1% l-glutamate, and 5% penicillin/streptomycin (3.5 units/3.5 μg/ml) in 10-cm plates. Cells were allowed to grow to 80% confluency. Zeocin (0.6 mg/ml; Invitrogen, Carlsbad, CA) was added to the culture medium in each plate at every other passage. Trypsinized cells were harvested between passages 8 and 10. Harvested cells were homogenized in 10 mM Tris buffer, pH 7.4, containing 0.25 M sucrose by sonicating for five 20-s bursts with 1-min resting on ice between each pulse/burst and then stored at –80°C. Protein content for both microsomes and cell homogenates was determined with the Bradford assay (Bio-Rad, Hercules, CA) with bovine serum albumin as a standard.
Western Blot Analyses. UGT1A6 Supersomes and cell homogenates expressing the various UGT1A6 allozymes were subjected to Western blot analysis for determination of relative levels of expressed UGT protein. Cell homogenates (50 μg) were electrophoresed on 8% SDS-polyacrylamide gels at 150 V for 1 h and transferred onto nitrocellulose membranes. The membrane was blocked in 2% milk in Tris-buffered saline with 0.05% Tween 20 (TBST) for 1 h. This was followed by incubation with 1:500 anti-UGT1A6 primary antibody overnight at room temperature. Membranes were washed for 30 min (3 × 10 min) with TBST, followed by incubation with 1:1000 horseradish peroxidase-conjugated anti-rabbit secondary antibody for 1 h. Membranes were washed with TBST and incubated for 1 min in SuperSignal West Pico chemiluminescent substrate (Thermo Fisher Scientific). On exposure and film development, images were analyzed with the software ImageJ 1.38× (National Institutes of Health, Bethesda, MD) to obtain densitometric measurements. Blots were stripped and reprobed for β-actin, and UGT1A6 levels in cell homogenates were normalized against β-actin.
cis- and trans-RES Glucuronidation Assays. Glucuronidation activity was determined in individual and pooled human liver microsomes (HLMs) stratified by UGT genotypes, commercially available HLM, and recombinant UGT1A6 and UGT1A1 isoforms, as well as cultured cell homogenates from HEK293 cells. Conditions for linearity with respect to time and protein concentration were optimized in preliminary studies, and all the assays were linear with respect to the selected time points and protein concentrations. Incubations with trans-RES were as described previously (Iwuchukwu and Nagar, 2008). The cis- and trans-RES were dissolved in DMSO to yield a final concentration of 5% v/v DMSO in incubation mixtures. All the solutions of either cis- or trans-RES used in glucuronidation assays were prepared fresh from stock solutions in DMSO, adequately protected from light, and stored at 4°C. For all the cis-RES incubations (as previously described for trans-RES), the incubation mixture consisted of enzyme fractions (final concentration, 1 mg/ml protein except HEK293 cell homogenates, which were 0.2 mg/ml), the substrate dissolved in DMSO (final DMSO concentration 5% v/v), alamethicin (10 μg/ml), MgCl2 (5 mM final concentration), and made up to final incubation volume with Tris-HCl buffer (100 mM, pH 7.4 at 37°C). cis-RES concentrations ranged from 0.05 to 2 mM, and the internal standard was acetaminophen (0.1 mM) in methanol. The tubes containing the reaction mixtures in a total volume of 50 μl were incubated in a shaking water bath at 37°C, with varying times: 30 min for the HEK293 cell homogenates and 60 min for UGT1A6 Supersomes and HLMs. The reactions were quenched with equal volumes of an ice-cold solution of the internal standard acetaminophen (100 μM) in methanol. Samples were centrifuged at 12,000 rpm for 6 min, and the supernatant was directly injected onto the high-performance liquid chromatography (HPLC) column for analysis. For each enzyme fraction, appropriate negative control experiments were performed under the same conditions but without either substrate or UDP-glucuronic acid. For cell homogenates expressing UGT1A6 allozymes, mock-transfected HEK293 cells were used as the negative control. For both the UGT1A1 and UGT1A6 phenotyping assays in individual human livers, commercially available pooled HLMs were used as a positive control. Because of the extremely small sample size of livers from nonmalignant patients, 25 liver samples were phenotyped for UGT1A6 (cis-RES) activity and 21 for UGT1A1 (trans-RES) activity. All the incubations were carried out in triplicate.
HPLC Analysis. Individual and pooled HLMs, purified recombinant UGT1A1 and UGT1A6 Supersomes, and cell homogenates generated for each of the four UGT1A6 allozymes were characterized for catalytic activity toward the substrates cis- or trans-RES by a sensitive reverse-phase HPLC assay. Conditions for the assay were as described previously (Iwuchukwu and Nagar, 2008). Another λmax was set up at 286 nm to detect cis-RES and its metabolites. Retention times for the two cis-RES metabolites were 7.3 and 8.8 min for cis-RES 4′-O-glucuronide (cis-R4′G) and cis-R3G, respectively. The areas of the glucuronide peaks formed were normalized to that of the internal standards chlorzoxazone or acetaminophen. As pure standards of the RES glucuronides were not commercially available, quantification of any RES glucuronide(s) was performed by comparing the normalized peak areas with that of a standard curve consisting of the parent RES. Preliminary diode array detector scans confirmed maxima for all the metabolites at the same wavelength as the parent. The standard curve correlation coefficients were ≥0.99, and the validated assay exhibited intraday and interday coefficients of variation and bias of less than 10% for the concentration range studied.
Data Analysis for Enzyme Kinetics. Before nonlinear regression analysis, all the data were transformed and Eadie-Hofstee (E-H) curves were plotted. The following equation was used to fit the data exhibiting nonlinear E-H plots indicative of partial substrate inhibition profile (Zhang et al., 1998): where v is the rate of the reaction, V1 and V2 are maximum velocity estimates, [S] is the substrate concentration, Km is the Michaelis-Menten constant, and Ki is the substrate inhibition constant.
Nonlinear regression was performed with GraphPad Prism for Windows (version 4.03; GraphPad Software Inc., San Diego, CA). Statistical comparison of the parameter estimates (Table 1) was performed with a two-sided t test assuming normal distribution, for which a p value <0.01 was considered significant (Nagar et al., 2004a). For comparison of glucuronidation velocities among genotyped liver samples, GraphPad Instat was used to conduct one-way analysis of variance (ANOVA), followed by post hoc Tukey's multiple comparison test. A p value <0.05 was considered significant.
Results
cis-RES glucuronidation was studied in UGT1A6 recombinant Supersomes and other protein sources to determine and characterize kinetic profile(s) for the formation of the two known metabolites—cis-R3G and cis-R4′G (Aumont et al., 2001; Sabolovic et al., 2006). Only cis-R3G was formed at the concentration range studied (0–2000 μM). Figure 1 depicts the kinetic profile obtained in UGT1A6 Supersomes, and the relevant kinetic parameters are listed in Table 1. cis-R3G exhibited a partial substrate inhibition (PSI) profile that was determined by nonlinear regression curve fitting and direct visualization of the E-H plot (Fig. 1, inset). The Vmax obtained in UGT1A6 Supersomes (27.2 ± 1.2 nmol/min/mg UGT1A6 protein) was 3- to 30-fold lower than that obtained in allozymes overexpressed in HEK293 cells. There was no significant difference between the apparent Km and Ki values in UGT1A6 Supersomes (Table 1).
To conduct pharmacogenetic studies on cis-RES, glucuronidation assays were carried out in HEK293 cells engineered to stably express allozymes representing the UGT1A6 alleles (*1–*4) of interest. The expression level of UGT1A6 protein in the four allozymes was determined via immunoblotting techniques, and a representative Western blot is shown in Fig. 2A. Glucuronidation velocities were normalized with UGT1A6 protein levels. cis-R3G formation in all the allozymes exhibited a PSI profile (Fig. 2B). Cell homogenates expressing the *4 allozyme had the highest Vmax (612 ± 27.36 nmol/min/mg UGT1A6 protein) followed by the *3 and *2 allozymes (Table 1). The *1 allozyme had the lowest Vmax of 86.1 ± 5.5 nmol/min/mg UGT1A6 protein. In the HEK293 cells, there was a significant difference (p < 0.01) between the Km (402.1 ± 39.2 and 376 ± 28.8 μM) and Ki (931.7 ± 101.2 and 763.9 ± 63 μM) values for *1 and *3, respectively (Table 1). No significant difference between Km and Ki was seen in the *2 and *4 allozymes.
Subsequently, 51 human livers were genotyped for three UGT1A6 SNPs using a PCR-restriction fragment length polymorphism assay (Nagar et al., 2004b). All four UGT1A6 alleles were detected, and the corresponding alleles and their frequencies are listed in Table 2. Five UGT1A6 genotypes were detected based on restriction patterns obtained from restriction enzyme digests of the relevant PCR amplicons. The restriction patterns obtained on DNA electrophoreses are depicted in Fig. 3A. The genotypes assigned and their respective frequencies are also listed in Table 2. All the frequencies were in Hardy-Weinberg equilibrium. Pooled genotyped human livers bearing the relevant polymorphisms were then used to conduct cis-RES glucuronidation assays, and Fig. 3B depicts the results obtained with UGT1A6-genotyped livers. cis-R3G was formed to a greater extent than cis-R4′G, as has been previously reported (Aumont et al., 2001; Sabolovic et al., 2006). The velocities in UGT1A6 *1/*2 and *2/*2 livers were significantly different (p < 0.05) from that in UGT1A6 *1/*1 livers for the formation of cis-R3G. No difference was observed between the *1/*4 livers and homozygous wild-type livers. Paucity of livers bearing the *1/*3 genotype precluded its use in this study. For the minor cis-R4′G metabolite, no difference was seen across all the genotypes.
For studies on UGT1A1 and trans-RES, 67 livers were genotyped, and Table 2 lists all the genotype and allele frequencies calculated for UGT1A1. As seen with UGT1A6, all the frequencies also adhered to the Hardy-Weinberg equilibrium.
trans-RES glucuronidation assays were conducted in individual human livers and pooled livers bearing the relevant UGT1A1 genotypes. There was some variability (5-fold) observed for the formation of trans-RES metabolites in the individual livers (Fig. 4A). Figure 4B shows the formation of trans-RES metabolites in pooled human livers stratified by UGT1A1 genotype. R3G was the major metabolite, but R4′G was also formed to a limited extent. No significant difference in the formation of R3G was seen in all the genotypes studied. However, for the formation of R4′G, we observed a statistically significant difference (p < 0.05) in the 7/7 livers when compared with the 6/6 or 6/7 livers (Fig. 4B).
Discussion
RES attracted interest as a chemopreventive agent after its roles in multiple stages of carcinogenesis were elucidated (Jang et al., 1997). The doses required to elicit RES's in vitro activities are up to 100 times greater than peak concentrations seen in humans (Gescher and Steward, 2003). Although it has been shown to be well absorbed, the low bioavailability observed on oral administration of low doses (25 mg) was attributed to extensive metabolism including sulfation and glucuronidation (Walle et al., 2004). Clinical trials involving high-dose RES (up to 5 g) yielded 2-fold lower peak concentrations than those required to elicit its effects in vivo (Boocock et al., 2007). Phase I trials in patients with resectable colorectal cancer are currently underway. Boocock et al. (2007) showed circulating levels of RES metabolites to be 3- to 8-fold higher than that of the parent aglycone and suggested that their chemopreventive effects warranted further investigation.
In nature, RES extracts comprise both its cis- and trans-isomers. Most studies have been conducted on only the trans-isomer, possibly because of commercial unavailability of the cis-isomer until recently. Activities attributed to the cis-isomer include modulation of inflammatory genes associated with endothelial dysfunction (Leiro et al., 2004). Comparison of its antioxidant and other activities with that of the trans-isomer showed only minor quantitative differences (Orallo, 2006; Campos-Toimil et al., 2007). Consumption of high-dose RES would probably saturate the sulfation pathway, making glucuronidation the major route of metabolism. Therefore, we set out to characterize 1) metabolic profiles obtained on high-dose consumption of total RES using in vitro systems and 2) the effects of common polymorphisms present in two major UGTs responsible for RES glucuronidation.
To the best of our knowledge, this is the first report on kinetics of cis-RES glucuronidation. The kinetic profile obtained for the formation of the major cis-RES metabolite cis-R3G by UGT1A6 Supersomes followed a partial substrate inhibition profile, confirmed by the E-H plot exhibiting a hook in the upper quadrant (Hutzler and Tracy, 2002). This profile is similar to the one-enzyme, two-site model proposed for UGT1A1-catalyzed formation of R3G (Iwuchukwu and Nagar, 2008). The profiles for glucuronidation at the 3-OH position in both isomers of RES seem to be similar irrespective of which UGT isoform is involved. The in vitro inhibition profiles obtained for RES at these higher concentrations are of interest because the presence of such kinetics may affect how data are scaled to the in vivo situation (Hutzler and Tracy, 2002). A study on effect of metabolism on RES transport across monolayers using Caco-2 cells found formation and transport of RES sulfates to be inhibited at concentrations higher than 50 μM, whereas that of the glucuronides followed classic hyperbolic kinetics up to the highest concentration used (Maier-Salamon et al., 2006). The concentrations used in that study were from 0 to 200 μM, a range at which we have also established Michaelis-Menten kinetics for both RES isomers. At much higher concentrations (>500 μM), we found that the kinetics of RES glucuronidation in most UGT isoforms becomes inhibitory (Iwuchukwu and Nagar, 2008). Future studies will be conducted to determine whether atypical kinetics observed for formation of RES glucuronides in vitro have any pharmacological implications on absorption and transport in vivo.
Aumont et al. (2001) reported higher activity of UGT1A6 toward cis-RES, which contrasts with results published by Sabolovic et al. (2006), who found UGT1A9 to be the higher activity isoform. We also observed higher activity of UGT1A6 toward cis-RES compared with UGT1A9 (O. F. Iwuchukwu and S. Nagar, unpublished data). However, a major drawback in all these studies is the use of recombinant Supersomes, which do not account for the relative expression levels of UGT isoenzymes in vivo. UGT1A6 is genetically polymorphic, with three nonsynonymous cSNPs (19T>G, 541A>G, and 552A>C) that occur in the variable first exon and encode amino acid changes S7A, T181A, and R184S (Ciotti et al., 1997; Nagar et al., 2004b; Krishnaswamy et al., 2005). The four alleles generated from these SNPs were used for glucuronidation assays.
All the UGT1A6 allozymes showed PSI profiles for cis-R3G formation with differences seen in their pattern of substrate inhibition. Thus, whereas all the allozyme data were described by a one-enzyme, two-binding site PSI model, the kinetics in *1 and *3 allozymes with significantly different Km and Ki values (p < 0.01) suggests sequential substrate binding to two sites. The *2 and *4 with no difference in Km and Ki exhibit simultaneous substrate binding (Korzekwa et al., 1998; Zhang et al., 1998). It is interesting to note that UGT1A6 *3 exhibited a Km similar to *1 but a significantly greater Vmax. The only difference between these allozymes is the S7A change, which has been suggested to alter the insertion of UGT1A6 in the endoplasmic reticulum membrane (Nagar et al., 2004b). This effect may account for variable orientation of identical mature proteins in the membrane, resulting in similar Km but varying Vmax values. The higher activities (velocities) seen with the UGT1A6 allelic variants have also been reported previously toward other UGT1A6 substrates (Nagar et al., 2004b; Chen et al., 2007). We cannot compare our data directly with any previous reports on effects of UGT1A6 polymorphisms (Krishnaswamy et al., 2005) as the atypical kinetics observed with cis-RES precludes the use of intrinsic clearance values obtained by the Vmax/Km ratio.
Human livers were genotyped to confirm phenotypes obtained in cellular studies. All four UGT1A6 alleles of interest were identified in our liver bank, and the five genotypes assigned were in Hardy-Weinberg equilibrium. In human livers stratified by UGT1A6 genotype, there was a correlation with phenotype for formation of cis-R3G (Table 2; Fig. 3). The UGT1A6*2 variants exhibited significantly higher glucuronidation activity (p < 0.05) compared with the homozygous *1 livers. The homozygous *2 livers had 1.6-fold greater activity than the *1 homozygote. As expected, UGT1A6 genotypes had no effect on cis-R4′G formation, a metabolite formed via UGT1A1 in the liver (Aumont et al., 2001). This genotype dependence may have implications on the bioavailability and chemopreventive effects of RES. Formation of cis-R3G has been reported to occur at a much faster rate than its trans counterpart (Aumont et al., 2001; Sabolovic et al., 2006), and results from liver samples used in this study also showed higher rates for cis-RES glucuronidation (Figs. 3 and 4).
Glucuronidation of trans-RES in individual livers from our liver bank showed a 5-fold degree of variability. We hypothesized that the UGT1A1 promoter polymorphism may contribute to this variability. This promoter polymorphism (UGT1A1*28) was detected in a panel of human livers, and genotypes were assigned. The TA repeat polymorphism confers several genotypes based on the number of TA repeats in the UGT1A1 promoter TATA box. The 7 (or 8) TA repeat SNP has been associated with decreased promoter activity and consequently reduced protein expression and activity (Bosma et al., 1995; Raijmakers et al., 2000; Fang and Lazarus, 2004; Girard et al., 2005; Yoder Graber et al., 2007). Pooled HLMs were used to carry out genotype-phenotype correlation studies with trans-RES as the substrate. No differences in glucuronidation based on genotype (6/6, 6/7, or 7/7 TA repeats) were observed for R3G, the major metabolite formed by UGT1A1. The 7/7 livers were seen to have significantly higher activity (p < 0.05) toward the minor metabolite R4′G, when compared with the 6/6 or 6/7 livers. Whether this trend is real or a result of linkage with another polymorphic UGT will be ascertained in future work.
We have previously reported UGT1A9 to be a major isoform involved in the formation of both R3G and R4′G (Iwuchukwu and Nagar, 2008). The polymorphic nature of UGT1A9 is known, and genotype-phenotype association studies for some of the more common UGT1A9 polymorphisms have been detailed in a recent review (Argikar et al., 2008). We can only postulate that major UGT1A9 polymorphisms with functional significance might correlate with trans-RES glucuronidation and account for the observed variability. Effects of major UGT1A9 polymorphisms on trans-RES glucuronidation will be elucidated in future studies.
The conundrum observed with low plasma levels of dietary polyphenols such as RES and their pharmacologic activity is yet to be fully explained. This study sheds some light on the role of UGT pharmacogenetics (like those observed with UGT1A6 polymorphisms) in cancer chemoprevention. Interethnic differences in frequency of UGT polymorphisms are known to exist, confounded by populations with variable exposures to dietary phytochemicals known to be conjugated by various UGTs. With respect to chemoprevention, several epidemiologic studies have attempted to correlate colorectal cancer risk with various UGT polymorphisms and exposure to dietary carcinogens (Fang and Lazarus, 2004; Butler et al., 2005; Girard et al., 2005). However, these polymorphic UGTs are not limited to simply inactivating carcinogens but also play a role in inactivating compounds such as chemopreventive nonsteroidal anti-inflammatory drugs and phytochemicals (Bigler et al., 2001; Chan et al., 2005). Bearing this in mind, future studies should take a double-sided approach to UGT pharmacogenetics by studying them in the context of overall exposure to cancer-related dietary chemicals—both carcinogens and chemopreventives. This would enable better appreciation of the impact of polymorphisms in major UGT isoforms responsible for metabolism of compounds like RES.
In summary, we determined the kinetics of cis-RES glucuronidation and found a positive correlation with UGT1A6 genotypes. On the other hand, the UGT1A1 TA repeat promoter polymorphism did not explain variable trans-RES glucuronidation. The role of UGT pharmacogenetics in variable dietary exposure to cancer-related compounds such as RES is not fully understood and should be further studied.
Acknowledgments
We thank Valencia Williams for assistance with DNA electrophoresis.
Footnotes
-
Part of this work was accepted for poster presentation as follows: Iwuchukwu OF, Ajetunmobi J, Ung D, and Nagar S (2008) Characterizing the effects of common polymorphisms in UGT1A1 and UGT1A6 on the glucuronidation of cis and trans resveratrol. Annual Meeting of the American Association of Pharmaceutical Scientists; 2008 Nov 16–20; Atlanta, GA. American Association of Pharmaceutical Scientists, Arlington, VA.
-
J.A. and D.U. contributed equally to this work.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.109.027391.
-
ABBREVIATIONS: RES, resveratrol; UGT, UDP glucuronosyltransferase; R3G, trans-resveratrol 3-O-glucuronide; R4′G, trans-resveratrol 4′-O glucuronide; cis-R3G, cis-resveratrol 3-O-glucuronide; SNP, single nucleotide polymorphism; DMSO, dimethyl sulfoxide; PCR, polymerase chain reaction; HEK, human embryonic kidney; TBST, Tris-buffered saline/Tween 20; HLM, human liver microsome; HPLC, high-performance liquid chromatography; cis-R4′G, cis-resveratrol 4′-O-glucuronide; E-H, Eadie-Hofstee plots; ANOVA, analysis of variance; PSI, partial substrate inhibition.
- Accepted April 29, 2009.
- Received March 3, 2009.
- The American Society for Pharmacology and Experimental Therapeutics