Prevalence of Polymorphisms in the Human UDP-Glucuronosyltransferase 2B Family: UGT2B4(D⁴⁵⁸E), UGT2B7(H²⁶⁸Y), and UGT2B15(D⁸⁵Y)¹

In humans, two families of UGTs³catalyze the transfer of the glucuronyl group from uridine 5′-disphosphoglucuronic acid to endogenous molecules, such as bile acids, steroids, and thyroid hormones, and a wide range of exogenous substrates. Generally, the UGT1 family catalyzes the glucuronidation of bilirubin and xenobiotic phenols as well as some steroids (1). The UGT2 family includes the subfamilies 2A and 2B;the 2B enzymes glucuronidate primarily steroids (2). UGTs play an important role in detoxification and chemoprotection (3), as well as drug metabolism and regulation of steroid hormone levels (2, 4). Polymorphisms in the UGTs are postulated to contribute to interindividual variation in drug disposition (5) and certain UGT1A variants are known to be associated with altered bilirubin excretion (6). It is highly likely that certain UGT polymorphisms may also influence steroid hormone metabolism and cancer susceptibility; however, the relationship between genetic variation in glucuronidation and cancer risk has yet to be explored.

Polymorphisms have been identified in at least three of the human UGT2B genes: UGT2B4, UGT2B7, and UGT2B15 (7, 8, 9). Reported nucleotide sequence differences in UGT2B4 lead to amino acid changes at positions 109, 396, and 458. The variant UGT2B4(L^109,396D⁴⁵⁸),isolated from human liver, has leucine (L), instead of phenylalanine(F) residues at amino acids 109 and 396 (7, 10). A difference of two nucleotides also leads to another single amino acid change of aspartate (D) to glutamate (E) at position 458 (7). UGT2B4 is primarily involved in the conjugation of catechol estrogens, bile acids (specifically hyodeoxycholic acid), and certain exogenous phytochemicals (7). Both UGT2B4(D⁴⁵⁸) and (E⁴⁵⁸) are expressed in liver and a wide range of extrahepatic tissues, including those that are targets for steroid hormones; however, there is no apparent difference in substrate specificity between the two forms (7). Lévesque et al. (7)reported that in a sample of 26 Caucasian individuals, the frequency of the variant E⁴⁵⁸ allele was 0.385. The authors did not detect the presence of the UGT2B4(L^109,396D⁴⁵⁸) variant in any of the tissues or genomic DNA sampled; however, they suggested that this may be a rare allele not found in their sample population.

In UGT2B7, there is an amino acid change from histidine (H) to tyrosine(Y) at amino acid 268 (8), the proposed location of the substrate-binding site (11). The prevalence of the two alleles and a corresponding phenotypic difference have not been determined. UGT2B7 glucuronidates 4-hydroxycatechol estrogens (12) and mono- and dihydroxylated androgens with a hydroxyl group in the 3α, 6α, and 17β positions; however, UGT2B7 has the highest activity toward steroids containing a 3α-hydroxy moiety (13, 14). UGT2B7 expression has been detected in liver, kidney, pancreas, brain, and the gastrointestinal tract (15, 16, 17).

In UGT2B15, a thymine in place of a guanine leads to an amino acid change at position 85 from aspartic acid (D) to tyrosine(Y; Ref. 9). UGT2B15 is expressed in numerous human tissues (e.g., liver, kidney, testis, mammary gland,prostate, and lung; Ref. 9). This isozyme catalyzes the glucuronidation of a wide range of substrates, including simple phenolic compounds, drugs, and C₁₉ steroids, such as 5α-androstane-3α,17β-diol and dihydrotestosterone, at the 17β position (18, 19). Coumarins, flavonoids, and anthraquinones, phytochemicals that are present in high amounts in certain plant foods, also are glucuronidated by UGT2B15 (18). To date, only one study has examined the prevalence of the UGT2B15 polymorphic allele; Lévesque et al. (9) determined that of a sample of 27 Caucasians, 6 (19%) were homozygous for Y⁸⁵(Y⁸⁵/Y⁸⁵), 5(22%) were homozygous for D⁸⁵(D⁸⁵/D⁸⁵),and 16 (73%) had both alleles(D⁸⁵/Y⁸⁵).

The purpose of our work was to determine the prevalence of the described genetic polymorphisms in UGT2B4, UGT2B7, and UGT2B15 in a sample of individuals from the Seattle, Washington area.

Two hundred and forty-five nonsmokers, 20–40 years of age, from the Seattle area were screened for participation in a cross-sectional study of diet and biotransformation enzymes. Exclusion criteria included history of gastrointestinal disorders, weight loss or gain>4.5 kg within the past year, major changes in eating habits within the past year (e.g., adoption of a different dietary pattern), exercise regimens requiring significant short-term dietary changes, antibiotic use within the past 3 months, body weight >150%of ideal, current drug therapy for a diagnosed disease, chronic nonsteroidal anti-inflammatory drug use, alcohol intake greater than two drinks per day (two drinks being equivalent to 720 ml of beer, 240 ml of wine, or 90 ml of hard liquor), occupational exposure to smoke or organic solvents, chronic passive exposure to tobacco smoke, intake of pharmacological doses of dietary supplements, and serum alanine aminotransferase concentrations above the normal range.

All participants had venous blood samples drawn for genomic DNA collection. DNA was extracted from lymphocytes according to standard methods using a proteinase K digestion followed by phenol-chloroform extractions (20). A demographic and health questionnaire was completed. Information requested on the questionnaire included usual diet (e.g., vegetarian, diabetic, low-fat), food preferences, alcohol intake, medical history, vitamin/mineral and herbal supplement information, usual activity, medications,environmental exposures, and ethnic/racial background. The study design was approved by the Institutional Review Board: Human Subjects Committee at the Fred Hutchinson Cancer Research Center. Informed written consent was obtained from all participants prior to the start of the study.

The UGT2B4 F¹⁰⁹L, F³⁹⁶L, and D⁴⁵⁸E; UGT2B7; and UGT2B15 genotypes were determined using an oligonucleotide ligation assay (21, 22). Amplification of the correct PCR fragments was confirmed by sequencing.

To distinguish the UGT2B4(D⁴⁵⁸E) (GenBank accession nos. AJ005162 and AF064200, respectively) and UGT2B4(L^109,396,D⁴⁵⁸)(GenBank accession no. AF081793) alleles, the polymorphisms at amino acids 109, 396, and 458 were determined. Three fragments were amplified by PCR using the following primers: for F¹⁰⁹L, FP1 (5′-GAGGATATTATCAAGCAGCT-3′)and RP1 (5′-CAGCATCTGCAAGAACAACA-3′); for F³⁹⁶L,FP2 (5′-ACCTCATGGTGGAGCCAATG-3′) and RP2 (5′-ACATTGTGTGGAAGTCCAAA-3′);and for D⁴⁵⁸E, FP3 (5′-TTCATCATGATCAACCAGTGA-3′)and RP3 (5′-CTTCCAGCCTCAGACGTAAT-3′). The PCR reactions contained 10 mm Tris-HCl (pH 8.3), 50 mm KCl, 1.5 mmMgCl₂, 0.001% gelatin, 50 ng/μl BSA, 200μ m deoxynucleotide triphosphates (100μ m for F¹⁰⁹L and D⁴⁵⁸E), 150 nmamplification primers, 100 ng of genomic DNA, and 0.5 units of AmpliTaq DNA polymerase (Perkin-Elmer, Foster City, CA). The cycling conditions on a PTC-100 thermal cycler (MJ Research, Inc., Watertown, MA) were as follows: (a) for F¹⁰⁹L and F³⁹⁶L, 94°C for 5 min, followed by 30 cycles of 94°C for 30 s, 58°C for 45 s, and 72°C for 60 s and 1 cycle of 72°C for 5 min; and (b) for D⁴⁵⁸E, 94°C for 5 min, followed by 30 cycles of 94°C for 30 s, 60°C for 45 s, and 72°C for 60 s and 1 cycle of 72°C for 5 min.

The F¹⁰⁹L, F³⁹⁶L, and D⁴⁵⁸E polymorphisms were determined by oligonucleotide ligation assay using the primers listed in Table 1. The modification for the allele-specific primers was 5′-biotin and 5′-phosphate and 3′-digoxigenin for the common primer. For the ligation, the PCR reactions were diluted with 35 μl of 0.1% Triton X-100. The ligation reactions consisted of 10 μl of diluted PCR product, and 10 μl of 40 mm Tris-HCl (pH 8.0),20 mm MgCl₂, 25 mm KCl, 2 mm DTT, 2 mm NAD⁺, 0.1% Triton X-100, 16 fmol/μl biotinylated primer, 16 fmol/μl common primer,and 0.015 units of thermostable ligase (Epicentre Technologies,Madison, WI). The cycling conditions for the ligation were as follows:15 cycles of 93°C for 30 s and 58°C for 2 min. The reaction was stopped with 10 μl of a buffer containing 0.1 m EDTA (pH 8.0) and 0.1% Triton X-100. The ligation reactions were then transferred into streptavidin-coated 96-well plates. After incubation at room temperature for 60 min, the plates were washed twice with 10 mm NaOH containing 0.05% Tween 20, followed by two washes with 200 μl of 100 mm Tris-HCl (pH 7.5), 150 mm NaCl, and 0.05% Tween 20. The plates were then incubated with 40 μl of a 1000-fold dilution of antidigoxigenin Fab fragment-phosphatase conjugate (0.75 units/μl; Boehringer Mannheim, Indianapolis, IN) for 30 min at room temperature. After four washes with 100 mm Tris-HCl (pH 7.5), 150 mm NaCl, and 0.05% Tween 20, the Life Technologies amplification system was applied for the color reaction according to the manufacturer’s recommendations. The absorbance at 490 nm was recorded using a SpectraMax 250 plate reader (Molecular Devices,Sunnyvale, CA).

The UGT2B7(H²⁶⁸Y) polymorphism consists of a C-to-T point mutation, resulting in an amino acid change from histidine (H) to tyrosine (Y) at amino acid 268 (8). Primers 5′-AAGCTGACGTATGGCTTATT-3′ and 5′-CAAAATCAACATTTGGTAAGAG-3′ were used to amplify a 74-bp fragment of the UGT2B7 gene containing the mutation. The PCR reactions contained 10 mm Tris-HCl (pH 8.3), 50 mm KCl, 1.5 mmMgCl₂, 0.001% gelatin, 50 ng/μl BSA, 200μ m deoxynucleotide triphosphates, 200 nm amplification primers, 100 ng of genomic DNA,and 0.5 units of AmpliTaq DNA polymerase (Perkin-Elmer). The cycling conditions on a PTC-100 thermal cycler (MJ Research) were as follows:94°C for 5 min, followed by 35 cycles of 94°C for 30 s, 61°C for 45 s, 72°C for 45 s and 1 cycle of 72°C for 5 min.

The primers for the ligation reactions were as follows: H²⁶⁸ (5′-CCTGGAATTTTCAGTTTCCTC-3′); Y²⁶⁸ (5′-CCTGGAATTTTCAGTTTCCAT-3′); and common primer (5′-ATCCACTCTTACCAAATGTTG-3′). The modifications for the primers were 5′-biotin for H²⁶⁸- and Y²⁶⁸-specific primers and 5′-phosphate and 3′-digoxigenin for the common primer. The ligation reactions were as described for UGT2B4.

The UGT2B15(D⁸⁵Y) polymorphism consists of a G-to-T point mutation, resulting in an amino acid change from aspartic acid (D) to tyrosine (Y) at amino acid 85 (9). Primers 5′-GTTACTTTAGCTCTGGAAGC-3′ and 5′-AGAGCTTGTTACTGTAGTCAT-3′ were used to amplify a 333-bp fragment of the UGT2B15 gene containing the mutation. The PCR reactions contained 10 mm Tris-HCl (pH 8.3), 50 mm KCl, 1.5 mmMgCl₂, 0.001% gelatin, 50 ng/μl BSA, 100μ m deoxynucleotide triphosphates, 100 nm amplification primers, 100 ng of genomic DNA,and 0.5 units of AmpliTaq DNA polymerase (Perkin-Elmer). The cycling conditions on a PTC-100 thermal cycler (MJ Research) were as follows:94°C for 5 min, followed by 35 cycles of 94°C for 30 s, 60°C for 45 s, 72°C for 60 s and 1 cycle of 72°C for 5 min.

The ligation was performed as described above for UGT2B7. The ligation primers for UGT2B15 were D⁸⁵ (5′-TCCTACATCTTTAACTAAAAATG-3′); Y⁸⁵ (5′-TCCTACATCTTTAACTAAAAATT-3′); and common primer (5′-ATTTGGAAGATTCTCTTCTGAA-3′). The modifications for the primers were 5′-biotin for D⁸⁵- and Y⁸⁵-specific primers and 5′-phosphate and 3′-digoxigenin for the common primer.

Statistical analyses were conducted using SAS software (Version 6.12 for Windows; SAS Institute Inc., Cary, NC). We measured differences between observed and expected UGT2B4(D⁴⁵⁸E),UGT2B7(H²⁶⁸Y), and UGT2B15(D⁸⁵Y) genotype frequencies using the χ² test and compared genotype frequencies between individuals of Asian and Caucasian background using Fisher’s exact test.

The 245 individuals (146 women and 99 men) who were genotyped for this project described their ethnic backgrounds as follows: 202 Caucasian, 32 Asian (including individuals of Indian origin), 4 Hispanic, 2 Pacific Islander, 2 African, 1 Alaskan Native, 1 Asian/Caucasian, and 1 Native American/Caucasian. Data are presented only for Caucasians and Asians because the small numbers of individuals in the other categories preclude statistical comparison among these ethnic groups.

The UGT2B4(D⁴⁵⁸E), UGT2B7(H²⁶⁸Y), and UGT2B15(D⁸⁵Y) allele and genotype frequencies are presented in Table 2. We did not detect the presence of the UGT2B4(L^109,396,D⁴⁵⁸)allelic variant in any of the samples. The allele frequencies were significantly different (P < 0.02) between individuals of Caucasian and Asian descent for all three polymorphisms. Genotype frequencies for the UGT2B4(D⁴⁵⁸E), UGT2B7(H²⁶⁸Y), and UGT2B15(D⁸⁵Y) polymorphisms also differed significantly by ethnic group (P < 0.0001, P = 0.0002, and P = 0.02,respectively). All of the Asians were homozygous wild type(D⁴⁵⁸/D⁴⁵⁸) for UGT2B4. For UGT2B7, almost 60% of individuals of Asian background were H²⁶⁸/H²⁶⁸ and only 9% were Y²⁶⁸/Y²⁶⁸,whereas among the Caucasians only 22% were H²⁶⁸/H²⁶⁸ and 29% were Y²⁶⁸/Y²⁶⁸. Similarly, for UGT2B15, 47% of Asians were D⁸⁵/D⁸⁵ and 19% were Y⁸⁵/Y⁸⁵,whereas 22% of Caucasians were D⁸⁵/D⁸⁵ and 32% were Y⁸⁵/Y⁸⁵. Within the Asian and Caucasian samples, the genotype frequencies were in Hardy-Weinberg equilibrium.

We have demonstrated that UGT2B4(D⁴⁵⁸E), UGT2B7(H²⁶⁸Y), and UGT2B15(D⁸⁵Y) are prevalent polymorphisms and that the genotype distributions appear to vary by ethnic group. All of the individuals of Asian descent in our study,compared with only 57.9% of Caucasians, were homozygous wild type for UGT2B4(D⁴⁵⁸). Although our sample of Asians is small, we would have expected to find the variant allele in∼13 individuals if the prevalence rate was similar to that in Caucasians. The variant UGT2B7Y²⁶⁸ and UGT2B15Y⁸⁵ alleles also were significantly more common among Caucasians compared with Asians. The prevalence of the UGT2B7 Y²⁶⁸/Y²⁶⁸ genotype was 3-fold higher in Caucasians than Asians. Similarly, the prevalence of the UGT2B15 Y⁸⁵/Y⁸⁵ genotype was 1.7-fold higher in Caucasians than in Asians and also may be higher in our sample of Caucasians (32%) than in the sample of 27 Caucasians genotyped by Lévesque et al. (9), of which 19% were homozygous Y⁸⁵/Y⁸⁵.

The UGT2B4(L^109,396,D⁴⁵⁸)variant allele appears to be very rare. Lévesque et al. (7) did not find expression of the UGT2B4(L^109,396,D⁴⁵⁸) transcript in any tissue samples tested, nor did they find the allele in 26 samples of genomic DNA. Similarly, in our 10-fold larger sample, we also did not find any individuals with this allele.

The extent to which the 2B polymorphisms may influence xenobiotic and endogenous steroid conjugation remains to be determined. In UGT2B4,substitution of E for D at 458 probably is a relatively minor amino acid modification, and stably expressed UGT2B4(D⁴⁵⁸) and UGT2B4(E⁴⁵⁸) in HK293 cells show similar substrate specificity (7). In the case of UGT2B7(H²⁶⁸Y), H-to-Y constitutes a nonconservative amino acid change in the region of the substrate binding site. Consequently, Jin et al. (8)postulated that the polymorphism would probably affect substrate specificity. In addition, they reported that UGT2B7(Y²⁶⁸), but not(H²⁶⁸), conjugated menthol and androsterone. More recently, Cheng et al. (12) and Coffman et al. (23) reported that in the case of catechol estrogens, androsterone, opioids, and several xenobiotics,substrate glucuronidation efficiency did not differ between cells designed to express UGT2B7(H²⁶⁸) or(Y²⁶⁸).

The polymorphism in UGT2B15(D⁸⁵Y) does not appear to alter specificities among the substrates tested to date, but it does increase the V_max by 2-fold and may contribute to individual variability in glucuronidation observed with some drugs and other compounds (24). Specifically,UGT2B15(Y⁸⁵) has a higher V_max than UGTB15(D⁸⁵) for 5α-androstane-3α,17β-diol and dihydrotestosterone (9). Given the high prevalence of the variant allele(Y⁸⁵; 50%) and the small differences in kinetics, Lévesque et al. (9)suggested that differences in steroid hormone metabolism are unlikely to be observed in vivo, but that this polymorphism may contribute to individual variability in xenobiotic glucuronidation. On that basis, the implications of carrying the Y⁸⁵ homozygous genotype are unclear;however, if the V_max is higher for the numerous phytochemicals that are metabolized by UGT2B15, one might postulate that these compounds would be cleared more rapidly and their potential chemopreventive effects would be reduced.

In conclusion, we have detected a high prevalence of the polymorphisms in UGT2B4(D⁴⁵⁸E), UGT2B7(H²⁶⁸Y), and UGT2B15(D⁸⁵Y) in a small,convenient sample of healthy individuals. In addition, we observed significant ethnic differences between Asians and Caucasians in the distribution of these polymorphisms. The relationship between UGT2B polymorphisms and cancer risk has not been explored;however, the role of UGT2Bs in the conjugation and excretion of steroid hormones and the international differences in risk of hormone-dependent cancers suggest that these enzymes may be important candidates for additional study. Whether the genotypic differences explain, in part,observed ethnic differences in steroid hormone profiles and drug metabolism remains to be determined. Our results suggest that further investigation of these polymorphisms in large, multiethnic cohorts is warranted.