Volume 98, Issue 4 p. 872-880
Original Article
Free Access

Detection of UGT1A10 polymorphisms and their association with orolaryngeal carcinoma risk

Abul Elahi Ph.D.

Abul Elahi Ph.D.

Cancer Epidemiology and Prevention Program, H. Lee Moffitt Cancer Center, Department of Interdisciplinary Oncology, University of South Florida, Tampa, Florida

Dr. Elahi and Mr. Bendaly contributed equally to this work.

Search for more papers by this author
Jean Bendaly M.S.

Jean Bendaly M.S.

Cancer Epidemiology and Prevention Program, H. Lee Moffitt Cancer Center, Department of Interdisciplinary Oncology, University of South Florida, Tampa, Florida

Dr. Elahi and Mr. Bendaly contributed equally to this work.

Search for more papers by this author
Zhong Zheng M.D., Ph.D.

Zhong Zheng M.D., Ph.D.

Cancer Epidemiology and Prevention Program, H. Lee Moffitt Cancer Center, Department of Interdisciplinary Oncology, University of South Florida, Tampa, Florida

Search for more papers by this author
Joshua E. Muscat Ph.D., M.P.H.

Joshua E. Muscat Ph.D., M.P.H.

Division of Epidemiology and Cancer Susceptibility, American Health Foundation, Valhalla, New York

Search for more papers by this author
John P. Richie Jr. Ph.D.

John P. Richie Jr. Ph.D.

Division of Epidemiology and Cancer Susceptibility, American Health Foundation, Valhalla, New York

Search for more papers by this author
Stimson P. Schantz M.D.

Stimson P. Schantz M.D.

Department of Otolaryngology, The New York Eye and Ear Infirmary, New York, New York

Search for more papers by this author
Philip Lazarus Ph.D.

Corresponding Author

Philip Lazarus Ph.D.

Cancer Epidemiology and Prevention Program, H. Lee Moffitt Cancer Center, Department of Interdisciplinary Oncology, University of South Florida, Tampa, Florida

Department of Biochemistry and Molecular Biology, University of South Florida, Tampa, Florida

Department of Pharmacology and Therapeutics, University of South Florida, Tampa, Florida

Fax: (813) 632-1328

H. Lee Moffitt Cancer Center, University of South Florida, 12902 Magnolia Drive, MRC-2E, Tampa, FL 33612===Search for more papers by this author
First published: 01 July 2003
Citations: 46

Abstract

BACKGROUND

UGT1A10 exhibits glucuronidating activity against metabolites of the tobacco smoke carcinogen, benzo(a)pyrene, and is expressed highly in numerous target tissues for tobacco-related cancers including the upper aerodigestive tract. The current study was conducted to determine the prevalence of genetic polymorphisms in the UGT1A10-specific region of the UDP-glucuronosyltransferase family 1A locus and their relationship with risk for orolaryngeal carcinoma.

METHODS

The authors analyzed UGT1A10-specific sequences in a population of black, white, and Asian individuals. Ten UGT1A10 alleles were identified by direct sequencing of UGT1A10 sequences amplified by polymerase chain reaction (PCR) using DNA purified from buccal cell swabs that were taken from individual subjects.

RESULTS

In addition to three silent polymorphisms, three missense polymorphisms were found at codons 139 (Glu > Lys), 240 (Thr > Met), and 244 (Leu > Ile). Using PCR-restriction fragment length polymorphism analysis of buccal cell DNA, the prevalence of the UGT1A10240Met variant was less than 0.01% in whites and blacks. Similarly, the prevalence of both the UGT1A10139Lys and UGT1A10244Ile variants was less than 0.01% in whites but it was significantly higher (0.04 and 0.05, respectively, P < 0.01) in blacks. None of the missense UGT1A10 variants were found in any of the Asian individuals examined. In a case–control study of black individuals, a significant association with orolaryngeal carcinoma risk was found in persons with at least 1 UGT1A10139Lys allele (crude odds ratio, 0.29 [95% confidence interval, 0.10–0.81]; adjusted odds ratio, 0.20 [95% confidence interval, 0.05–0.87]). No association was observed for the codon 244 (Leu > Ile) polymorphism.

CONCLUSIONS

The data from the current study show that the UGT1A10 gene has several low-frequency missense polymorphisms and that the codon 139 polymorphism is an independent risk factor for orolaryngeal carcinoma in blacks. Cancer 2003;98:872–80. © 2003 American Cancer Society.

DOI 10.1002/cncr.11587

The UDP-glucuronosyltransferase (UGT) superfamily of enzymes catalyze the glucuronidation of a variety of endogenous compounds such as bilirubin and steroid hormones and xenobiotics such as drugs and environmental carcinogens.1-4 Based on structural as well as sequence homology, UGTs are classified into several families and subfamilies.5 UGT family 2B members are derived from independent genes, whereas the entire UGT1A family is derived from a single locus in chromosome 2. This locus codes for nine functional proteins that differ only in their amino-terminus as a result of alternate splicing of independent exon 1 regions to a shared carboxy-terminus encoded by exons 2–5.6 Several family 1A UGTs have been implicated in the conjugation and detoxification of tobacco carcinogen metabolites, including the tobacco-specific nitrosamine, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK),4 and polycyclic aromatic hydrocarbons like benzo(a)pyrene (BaP).7-11 Although most family 1A UGTs are expressed in the liver,10, 12-14 UGTs 1A7, 1A8, and 1A10 are extrahepatic and are expressed in several target areas for tobacco-induced malignancies, including the aerodigestive tract.15

Many genes that encode enzymes involved in the metabolic activation or detoxification of carcinogens are polymorphic, and many also are associated with an increased risk of aerodigestive tract carcinomas.16, 17 Polymorphisms have been identified in several UGT genes, including UGT1A1, UGT1A6, UGT1A7, UGT2B4, UGT2B7, and UGT2B15.11, 18-21 In the family of 2B polymorphic variants, UGT2B1585Tyr was associated with an increase in UGT2B15 function19 and an increased risk of prostate carcinoma.22 Few high-prevalence polymorphisms (> 1%) have been identified in exons 2–5 of the common region of the family UGT1A locus.23 The “TATA” box polymorphism in the promoter region of UGT1A1, commonly associated with Gilbert syndrome, is associated with reduced function in the UGT1A1 transcriptional promoter24 and has been implicated in the increased risk for breast carcinoma.25 Results from other studies suggest that UGT1A7-specific genetic variants are associated with reduced UGT1A7 metabolic function11 and are linked strongly to the increased risk for orolaryngeal26 and hepatocellular carcinoma.27

Because UGT1A10 is expressed in the aerodigestive tract and exhibits high activity against several BaP metabolites,28 it was hypothesized that polymorphisms in the UGT1A10 gene could be associated with the risk for aerodigestive tract carcinomas. The goals of the current study were to determine whether genetic polymorphisms exist in the UGT1A10-specific region of the UGT family 1A locus, to determine the prevalence of these polymorphisms in multiple racial groups, and to assess their association with the risk of orolaryngeal carcinoma.

MATERIALS AND METHODS

Study Population

For the identification of UGT1A10 polymorphisms and determination of prevalence in different racial groups, our population included 162 whites and 110 blacks from New York City and 200 whites, 79 blacks, and 69 Asians (35 of Indian descent and 34 of East Asian descent) from Tampa, FL. These individuals were participants in previous studies of genetic polymorphisms and other risk factors for aerodigestive tract carcinoma.29-31 The allele and genotype frequencies for polymorphisms in the CYP1A1, CYP2E1, GSTM1, GSTT1, and GSTP1 genes in this population were similar to those observed in a pooled analysis of more than 15,000 persons in other studies, despite differences in the type of control, age, gender, and other characteristics.32

For the case–control study, the importance of UGT1A10 polymorphisms in the risk for orolaryngeal carcinoma was determined for 115 black case–control pairs recruited between 1996 and 2000 from Temple University Hospital (Philadelphia, PA), the New York Eye and Ear Infirmary (New York, NY), and the State University of New York at Brooklyn (New York, NY). All cases were newly diagnosed patients (i.e., they were diagnosed within 1 year before study entry) who had histologically confirmed primary oral (n = 63; including cancers of the tongue [n = 21], tonsil [n = 11], pharynx or hypopharynx [n = 9], oral cavity [n = 19], or mixed sites [n = 3]) or laryngeal (n = 52) squamous cell carcinoma. Controls were outpatients without cancer treated at the ear, nose, and throat or dental clinics of participating institutions. Controls were individually matched to cases based on age (within 5 years) and month of case interview (± 2 months). To control for biases in demographics or other factors inherent in the recruitment of participants from institutions in different locations, controls also were matched to case patients based on the institution of the case patient. Depending on the institute, eligible cases were identified either from admission rosters, surgical operating schedules, or cancer care listings. Ninety-five percent of controls and 98% of case patients who were asked to participate in the study consented.

A structured questionnaire that contained items on demographics, life-long smoking habits, and other habits was administered by trained interviewers as previously described.30 Tobacco use was categorized into pack-years (1 pack-year = 1 pack of cigarettes per day for 1 year, 4 cigars per day for 1 year, or as 5 pipes per day for 1 year).33 Alcohol consumption was calculated as shots per day. One shot was defined as 12.9 g of 43% alcohol, which is roughly equivalent to 1 oz of 86-proof hard liquor, a 3.6 oz glass of wine, or a 12 oz can of beer. Study participants were defined as drinkers of alcohol if they reported drinking a minimum of 1 shot per week for a minimum of 10 years. Participants were classified as never-drinkers if they consumed 1 or fewer shots per week, light drinkers if they consumed from 1 to less than 7 shots per week, moderate drinkers if they consumed from 7 to less than 28 shots per week, and heavy drinkers if they consumed 28 or more shots per week.

Buccal cell samples were collected from all participants and used for the analysis of polymorphic UGT1A10 genotypes. Protocols involving the analysis of buccal cell specimens and the administering of questionnaires were approved by the institutional review board at the H. Lee Moffitt Cancer Center (Tampa, FL) and collaborating institutions and were performed in accordance with assurances filed with and approved by the U.S. Department of Health and Human Services. Informed consent was obtained from all participants.

UGT1A10 Polymerase Chain Reaction Amplification Sequencing and Genotyping Analysis

DNA was isolated from exfoliated buccal cell specimens by incubating cell pellets with proteinase K (0.1 mg/mL) in 1% sodium dodecyl sulfate overnight at 50 °C, extracting with phenol:chloroform, and precipitating with ethanol as previously described.30 Care was taken during DNA purification and isolation to prevent contamination and cross-contamination between samples during polymerase chain reaction (PCR). The purification of DNA samples was performed in a location distant from the workstation where PCR amplifications were performed. All equipment used for tissue blending and homogenization was washed in a bath of concentrated chromic acid/sulfuric acid, rinsed 3 times in autoclaved double-distilled water and once in 70% ethanol, air-dried, and autoclaved after each tissue sample was processed as described above.

The family 1A locus comprises divergent and individually regulated exon 1 sequences that transcribe for mRNAs that are spliced alternatively onto the 5′-end of the sequence encoded by the common UGT exon 2–5 region. Therefore, UGT mRNAs consist of a unique region encoded by exon 1 and a region encoded by exons 2–5 that is common for all family 1A UGTs. To evaluate sequences that were UGT1A10 specific and that spanned the entire UGT1A10 exon 1 region, a dual PCR strategy was employed specifically to amplify and sequence 853 of the 855 base pairs (bp) of the reported UGT1A10 exon 1 sequence (GenBank34 accession number HSU8950). The 5′ end of UGT1A10 exon 1 (fragment “1”, size = 657 bp) was amplified by PCR using a sense primer (1A10S1; 5′-TCCGCCTACTGTATCATAGCA-3′) corresponding to nucleotides −61 through −41 relative to the translation start site in UGT1A10 exon 1 (GenBank34 accession numbers AF297093 and HSU89508) and a UGT1A10 exon 1-specific antisense primer (1A10AS1; 5′-TCTGAGAACCCTAAGAGATCA-3′') corresponding to nucleotides 576–596 of the UGT1A10 cDNA (GenBank34 accession number HSU89508). The 3′ end of UGT1A10 exon 1 (fragment “2”, size = 416 bp) was amplified by PCR using a UGT1A10 exon 1-specific sense primer (1A10S2; 5′-CTCTTTCCTATGTCCCCAATG-3′) corresponding to nucleotides 557–577 of the UGT1A10 cDNA and an antisense primer (1A10AS2; 5′-CTGGAAAGAAATCTGAAATGCAACAAAC-3′) corresponding to nucleotides 54047–54074 of the UGT family 1 loci, which corresponds to nucleotides 90–117 downstream of UGT1A10 exon 1 (GenBank34 accession number AF297093). To maximize specificity for PCR amplification of UGT1A10 sequences, the UGT1A10-specific nucleotides 576 and 577 of the UGT1A10 cDNA were present at the 3′-end of both the antisense primer used for PCR amplification of fragment 1 and the sense primer used for PCR amplification of fragment 2. A 1-bp mismatch was introduced into the 1A10AS2 primer (denoted by the underlined ‘C’) to match PCR primer annealing temperatures and increase PCR efficiency. PCR amplifications were performed routinely in a 50-μL reaction volume containing 50 ng of purified genomic DNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 0.2 mM of each of the deoxynucleotide trisphosphates, 20 pmol of both sense and antisense UGT1A10 primers, and 2.5 units (U) of Taq DNA polymerase (Boehringer Mannheim, Indianapolis, IN). The reaction mixtures for both fragments 1 and 2 underwent the following incubations in a GeneAmp 9700 thermocycler (Perkin-Elmer, Foster City, CA): 1 cycle of 94 °C for 2 minutes, 41 cycles of 94 °C for 30 seconds, 56 ° for 30 seconds, and 72 °C for 30 seconds, after which a final cycle of 7 minutes at 72 °C was performed. The PCR amplification integrity of all samples was confirmed by electrophoresis in 8% polyacrylamide or 1.5% agarose gels that were stained subsequently with ethidium bromide and evaluated under ultraviolet light using a computerized photoimaging system (AlphaImager 2000, Alpha Innotech, San Leandro, CA).

For dideoxy sequencing, PCR products were purified after electrophoresis in 1.5% agarose using the Qiaex II gel extraction kit (Qiagen, Valencia, CA). Dideoxy sequencing was performed at the Molecular Biology Core Facility at the H. Lee Moffitt Cancer Center using the same sense and antisense primers that were used for PCR amplification.

UGT1A10 genotypes for missense polymorphisms were assessed by restriction fragment length polymorphism (RFLP) analysis 1) to confirm polymorphic sequences in sequenced samples, 2) to screen buccal cell DNA specimens not initially analyzed by DNA sequencing for overall polymorphic prevalence, and 3) to screen buccal cells from orolaryngeal carcinoma cases and matched controls. UGT1A10 exon 1 sequences were PCR-amplified as described above, and RFLP analysis was performed at 37 °C for 2 hours using 10–15 μL of PCR product and 5 U of the appropriate restriction enzyme (EarI for UGT1A10 codon 139, BceA I for UGT1A10 codon 240, and EcoR V for UGT1A10 codon 244; all enzymes were purchased from New England Biolabs, Beverly, MA). For identification of UGT1A10 alleles in specimens with more than one polymorphism, PCR-amplified UGT1A10 exon 1 sequences from polymorphic subjects were cloned into the pCRII plasmid (Invitrogen, Carlsbad, CA). Colonies were selected and appropriately cloned UGT1A10 exon 1-containing plasmids were sequenced at the Molecular Biology Core Facility at the Moffitt Cancer Center using a T7-homologous plasmid-specific primer.

Statistical Analyses

Bivariate analysis included chi-square tests for differences in genotype frequencies and the Student t test for continuous variables such as cigarette consumption. The risk of orolaryngeal carcinoma in relation to UGT1A10 genotypes was determined by conditional logistic regression to calculate odds ratios (OR) and 95% confidence intervals (CI). For all analyses, the regression models included gender, age (continuous), pack-years of smoking (continuous), and alcohol consumption (categorical). The statistical computer software SPSS (version 10.1) was used to perform all statistical analyses. All statistical tests were two sided.

RESULTS

Screening for UGT1A10 Polymorphisms

Although single-nucleotide polymorphism database searches are useful for the detection of polymorphisms in unique or nonhomologous genes, database searches for the identification of polymorphisms in UGT family 1A exon 1 regions are less useful as a result of high nucleotide homology with other UGT family 1A members.6 Because UGT1A10 exon 1 exhibits high homology with other family 1A UGT genes, including greater than 95% homology with UGT1A8 exon 1, potential UGT1A10 exon 1 polymorphisms were identified by exploiting the UGT1A10-specific T:A/G:C doublet at positions 576 and 577 of the UGT1A10 gene (GenBank34 accession number AF297093). This region is one of the few sequences of two or more consecutive UGT1A10-specific nucleotides (compared with other family 1A UGT exon 1 sequences) present in UGT1A10 exon 1. PCR primers were designed with nucleotides 576 and 577 of the UGT1A10 cDNA present on both the antisense primer used for PCR amplification of the 5′ end of UGT1A10 exon 1 (fragment 1) and the sense primer used for PCR amplification of the 3′ end of UGT1A10 exon 1 (fragment 2). A total of 102 individuals (53 whites and 49 blacks) were screened by dideoxy sequencing of UGT1A10 exon 1 fragment 1 and fragment 2 sequences. Informative sequencing information was obtained for all specimens evaluated in the current analysis.

Described in GenBank are both the UGT1A10 cDNA (GenBank34 accession number HSU89508) and the UGT1A10 exon 1 sequence within the family 1A UGT gene (GenBank34 accession number AF297093; see reference6). Due to two 5′-end frameshift differences, the resulting UGT1A10 proteins encoded by the two sequences differ by a stretch of 21 consecutive amino acids near the N'-terminus (residues 3–23), representing a possible polymorphic difference. In the current study, the sequencing data for this region of UGT1A10 exon 1 for all 103 participants were identical and matched the data of the UGT1A10 gene sequence (AF297093; Fig. 1), suggesting a sequencing error in the original description of the UGT1A10 cDNA.

Details are in the caption following the image

UGT1A10 sequence comparisons between GenBank34 and the current study.

Downstream of this region, six polymorphisms were detected by sequencing analysis (Table 1). Three were “silent” (codons 42, 199, and 231) and three were missense polymorphisms resulting in amino acid changes within the UGT1A10 sequence (codons 139, 240, and 244; Fig. 2A). The codon 139 (G > A) polymorphism resulted in a glutamic acid-to-lysine (Glu > Lys) amino acid change, the codon 240 (C > T) polymorphism resulted in a threonine-to-methionine (Thr > Met) amino acid change, and the codon 244 (C > A) polymorphism resulted in a leucine-to-isoleucine (Leu > Ile) amino acid change, as detected by direct sequencing in two (both black), one (white), and five (one white, four black) individuals, respectively. All three amino acid–changing polymorphisms were confirmed by RFLP analysis (Fig. 2B).

Table 1. Description and Prevalence of UGT1A10 Polymorphisms by Racial Group
Polymorphism Nucleotide substitution Amino acid substitution Allelic prevalence
Blacksa Whitesb Asiansc
New York Florida Total New York Florida Total East Asian Indian
Codon 42 CAG > CAA No 0.02 ND 0.02 ND ND ND
Codon 139 GAG > AAG Glu > Lys 0.05 0.03 0.04de Not detected Not detected < 0.01 Not detected Not detected
Codon 199 CAT > GAC No 0.04 ND 0.02 ND ND ND
Codon 231 GCC > GCT No 0.12 ND 0.14 ND ND ND
Codon 240 ACG > ATG Thr > Met Not detected 0.01 < 0.01 < 0.01 < 0.01 < 0.01 Not detected Not detected
Codon 244 CTC > ATC Leu > Ile 0.04 0.06 0.05fg < 0.01 < 0.01 < 0.01 Not detected Not detected
  • ND, not done.
  • a Prevalence analysis was performed by direct sequencing and/or polymerase chain reaction–restriction fragment length polymorphism analysis for 189 black subjects recruited from Mt. Vernon, NY (n = 110), and Tampa, FL (n = 79).
  • b Prevalence analysis was performed by direct sequencing and/or PCR-RFLP analysis for 362 white subjects recruited from Mt. Vernon, NY (n = 162), and Tampa, FL (n = 200).
  • c The total number of Asian subjects examined included 35 Indian Asians and 34 subjects of East Asian descent.
  • d Includes 11 subjects who were heterozygous and 2 subjects who were homozygous for the UGT1A10139Lys variant.
  • e Prevalence was significantly greater (P < 0.005
  • f P < 0.001, respectively) in blacks compared with whites.
  • g All subjects with a UGT1A10244Ile variant were heterozygous.
Details are in the caption following the image

(A) Representative results of UGT1A10 exon 1 sequencing. Shown are sequencing results for each of the polymorphic codons (including silent and missense changes) identified in the current study. (B) Representative results of polymerase chain reaction-restriction–fragment length polymorphism (PCR-RFLP) analysis. Shown are representative gels of PCR-RFLP of UGT1A10 exon 1 missense polymorphisms using buccal cell DNA as template. Top panel (codon 139), EarI digestion; middle panel (codon 240), BceA I digestion; bottom panel (codon 244), EcoR V digestion. Lane 1 (all panels), DNA marker; Lane 2 (all panels), PCR product without restriction enzyme digestion; Lanes 3 and 4 (all panels), homozygous wild-type RFLP pattern; Lane 5 (top panel) and Lanes 5 and 6 (middle and bottom panels), heterozygous RFLP pattern; Lane 6 (top panel), homozygous polymorphic pattern. bp: base pairs

Ten UGT1A10 exon 1–specific alleles could be discerned by haplotype analysis (Fig. 3). In addition to four UGT1A10 alleles encoding the wild-type UGT1A10 protein (designated UGT1A10*1a through UGT1A10*1c according to the nomenclature guidelines of MacKenzie et al.5), two alleles encoding the UGT1A10240Met variant isozyme and three alleles encoding the UGT1A10244Ile variant isozyme were identified. In addition to the wild-type UGT1A10*1 allele, two alleles were identified (UGT1A10*1a and UGT1A10*1b) that encoded wild-type UGT1A10 protein but contained silent polymorphisms at codons 42 and 199, respectively. Of the 103 individuals screened in the current analysis, 1 was homozygous for the UGT1A10231T variant and also was heterozygous at codon 199, indicating the presence of a third wild-type–encoding UGT1A10*1 variant (UGT1A10*1c) that contained silent polymorphisms at codons 199 and 231. Three UGT1A10 allelic variants exhibited single–base pair polymorphisms at codons 139 (UGT1A10*2), 240 (UGT1A10*3), and 244 (UGT1A10*4), each resulting in amino acid changes (as described above). For persons who were heterozygous for multiple polymorphisms (n = 5, all black), the UGT1A10 haplotype could not be discerned by direct sequencing of PCR amplifications using buccal cell genomic DNA as the template. To elucidate the UGT1A10 haplotype in these individuals, DNA sequencing analysis was performed after cloning of UGT1A10 PCR-amplified exon 1 fragments. Sequencing analysis of individual UGT1A10 exon 1-containing plasmid clones, representative of individual UGT1A10 alleles, demonstrated the existence of three additional UGT1A10 alleles: UGT1A10*3a (present in 1 individual), which was polymorphic at codons 231 and 240; UGT1A10*4a (present in 1 individual), which was polymorphic at codons 231 and 244; and UGT1A10*4b (present in 3 individuals; allelic prevalence in blacks, 0.008), which was polymorphic at codons 199, 231, and 244. None of the variant alleles discovered in the current analysis possessed more than a single UGT1A10 missense polymorphism.

Details are in the caption following the image

Schematic of UGT1A10 exon 1 nucleotide changes corresponding to each of the UGT1A10 allelic variants.

Prevalence of UGT1A10 Missense Polymorphisms

To assess the prevalence of UGT1A10-specific missense polymorphisms in different racial/ethnic groups, RFLP analysis was used to screen each of the missense polymorphisms in healthy white, black, and Asian individuals recruited from Tampa or New York City (Table 1). The codon 240 (Thr > Met) polymorphism was detected in 3 blacks and 6 whites, with a resulting allelic prevalence of less than 0.01 in both groups. Although the prevalence of both the codon 139 (Glu > Lys) and codon 244 (Leu > Ile) polymorphisms was less than 0.01 in whites, the prevalence of both polymorphisms was significantly higher (P < 0.001 for both polymorphisms) in blacks. The prevalence of the UGT1A10139Lys- and UGT1A10244Ile-containing alleles was 0.04 and 0.05, respectively, in the black cohort screened in the current study. Although some variation in prevalence was observed for the UGT1A10 codon 139 and 244 polymorphisms for blacks recruited from Florida versus New York, these differences were not significant. None of the missense polymorphisms were observed in any of the Indian or East Asian individuals screened in the current study (Table 1).

Analysis of UGT1A10 Polymorphisms and Orolaryngeal Carcinoma Risk

The potential role for UGT1A10 polymorphisms in the risk for orolaryngeal carcinoma was evaluated in a case–control study of 115 black patients with newly diagnosed orolaryngeal carcinoma and 115 matched controls. Seventy-two percent of cases and 62% of controls were men. The mean age for the cases and controls was 58 years. As expected, the average pack-years of smoking was significantly higher in case patients than in control patients (39 vs. 9 pack-years, respectively, P < 0.01). Only 5% of case patients were never-smokers, compared with 59% of control patients. A higher percentage of case patients than control patients were heavy drinkers of alcohol (28 or more shots per week; 49% vs. 16%, P < 0.01).

Informative PCR results were obtained for all 115 case–control pairs (230 total subjects) except for the UGT1A10 codon 139 polymorphism in 2 case patients and the UGT1A10 codon 244 polymorphism in 4 control patients (Table 2). Among control patients, the prevalence of these polymorphisms followed the Hardy–Weinberg equilibrium and the prevalence of both polymorphisms was similar to that observed for blacks in New York (Table 1). There was no significant difference in allelic prevalence between men and women among either case patients or control patients (results not shown).

Table 2. UGT1A10 Genotype Prevalence and Risk for Orolaryngeal Carcinoma
UGT1A10 genotype Controls (%) Cases (%) Crude OR (95% CI) Adjusted OR (95% CI)a
Codon 139b
 Glu > Glu 99 (86) 108 (96) 1.0 (referent) 1.0 (referent)
 Glu > Lysc 16 (14) 5 (4.4) 0.29 (0.10–0.81) 0.20 (0.05–0.87)
Codon 244d
 Leu > Leu 101 (91) 105 (91) 1.0 (referent) 1.0 (referent)
 Leu > Ilec 10 (9.0) 10 (8.7) 0.96 (0.38–2.4) 0.94 (0.26–3.4)
  • OR, odds ratio; CI, confidence interval.
  • a Adjusted for age, gender, smoking (pack-years), and alcohol consumption (categoric variables).
  • b Noninformative polymerase chain reaction analyses were obtained in two cases for Codon 139 analysis.
  • c None of the subjects screened in the case–control study were homozygous for the polymorphic variant for either the UGT1A10 Codon 139 or 244 polymorphism.
  • d Noninformative Polymerase chain reaction analyses were obtained in four controls for codon 244 analysis.

There was no significant difference in the prevalence of the UGT1A10244Ile polymorphic variant between case patients and control patients (0.041 vs. 0.045). A significantly higher prevalence of the UGT1A10139Lys polymorphic variant was observed in control patients than in case patients (0.07 vs. 0.022, P < 0.01). As shown in Table 2, individuals with 1 or more UGT1A10139Lys polymorphic variants exhibited a significant decrease in risk for orolaryngeal carcinoma (ORcrude, 0.29; 95% CI, 0.10–0.81; P < 0.02). This risk was not affected by adjusting for other factors via regression analysis (ORadjusted, 0.20; 95% CI, 0.05–0.87). There was no association between the UGT1A10244Ile polymorphic variant and orolaryngeal carcinoma risk (ORadjusted, 0.94; 95% CI, 0.26–3.4).

DISCUSSION

UGT1A10 has been implicated strongly in the glucuronidation of several important BaP metabolites including BaP-7,8-dihydrodiol, the precursor to the potent mutagen, BaP-7,8-dihydrodiol-9,10-epoxide.28 Although most family 1A UGTs are expressed in the liver,10, 12-14 UGT1A10 is extrahepatic and is expressed in several target areas for tobacco-induced malignancies, including the oral cavity and the larynx.15 Therefore, UGT1A10 may play an important role in the detoxification of tobacco-smoke carcinogens, such as BaP, in these tissues.

Previous studies have shown that few UGT family 1A missense polymorphisms have been identified in the exon 2–5 common region of the family 1A locus.23 A missense polymorphism at the 3′ end of exon 4 of the UGT1A loci resulting in a Pro > Leu amino acid change was identified in 10 of 190 Asian individuals, but this polymorphism was not detected in a screening of 162 white or black persons (Lazarus and Zheng, unpublished data, 2001). In the current study, several polymorphisms were identified in the UGT1A10-specific region (UGT1A10 exon 1). Of these, three resulted in amino acid changes that could potentially alter UGT1A10 protein function. The prevalence of all 3 missense polymorphisms was less than 1% in whites, and none of these polymorphisms were identified in a small cohort of Asian individuals. These data suggest that coding region variations in UGT1A10 are rare and do not play a significant role in cancer susceptibility in these ethnic groups. Due to their low prevalence, it was not possible to determine the risk of orolaryngeal carcinoma associated with UGT1A10 polymorphisms in these groups.

A significantly decreased risk of orolaryngeal carcinoma was found to be associated with the UGT1A10 codon 139 (Glu > Lys) polymorphism but not with the codon 244 (Leu > Ile) polymorphism in blacks. These data are consistent with the finding that the amino acid change from Glu > Lys for the codon 139 polymorphism is a highly nonconservative change in amino acid sequence and is therefore a more likely candidate to alter UGT1A10 function than the Leu > Ile change observed for the codon 244 polymorphism. The protective effect of the UGT1A10139Lys variant on orolaryngeal carcinoma risk observed in the current study would suggest that this variant may exhibit higher glucuronidating and, therefore, detoxifying activity against tobacco carcinogens like BaP-7,8-dihydrodiol. Alternatively, the codon 139 polymorphism may be in linkage disequilibrium with another marker in the UGT1 locus. Further studies are being performed in our laboratory to confirm these possibilities.

The current study has two potential limitations. The overall study sample was relatively small, which limited our ability to determine gene–environment interactions. There were only six case participants who never smoked cigarettes, and consequently it was not possible to determine gene–smoking interactions. When the data were stratified by pack-years (upper 50% of values vs. lower 50% of values), the OR was approximately the same in both groups, although the CIs were too wide to allow any meaningful comparisons. It should be emphasized that despite this limitation, this is the largest molecular epidemiologic case–control study of orolaryngeal carcinoma risk yet performed in blacks.

Another potential limitation of the current study is that the control patients recruited into the study were hospital outpatients attending dental or ear, nose, and throat clinics. It could be proposed that they may not reflect the overall prevalence of UGT1A10 alleles in the general population. However, as part of the International Project on Genetic Susceptibility to Environmental Carcinogens database, a large pooled analysis of more than 15,000 individuals without cancer was performed and no statistically significant differences in the overall prevalence of metabolizing enzyme polymorphisms between hospital-based and population-based controls were found.32 In addition, in contrast with the relatively poor response rates of most studies that use ‘population’-based controls, the response rate of control patients in the current study was very high (> 95%), thus limiting potential representation biases introduced by selective control recruitment. Therefore, although not proven, it is unlikely that the use of hospital-based outpatient controls substantially biased the findings of the current study.

In conclusion, we detected polymorphisms in the UGT1A10 gene and, of these, the codon 139 polymorphism may be an important risk factor for orolaryngeal carcinoma in blacks. Functional studies of UGT1A10 polymorphic variants as well as studies determining UGT1A10 expression levels in various human tissue specimens will be necessary to better characterize the effects of UGT1A10 polymorphisms on patient risk for other malignancies.

Acknowledgements

The authors thank Barbara Muffly, Mary Beth Colter, and Lanmin Zhang, and both the Molecular Biology Core Facility and Tissue Procurement Facility at the H. Lee Moffitt Cancer Center and Research Institute for their excellent technical assistance.