To the editor
Breast cancer is the most common cancer in women and cumulative estrogen exposure over a life time is the major risk factor 1. Therefore, the steroid hormone metabolizing enzymes have been proposed to play an important role in breast cancer risk 2. Among these enzymes are UDP-glucuronosyltransferases (UGTs), which can transfer glucuronic acid from UDP-glucuronic acid to substrates, thus making them more water soluble than their parent compound and more easily excreted through the biliary and renal systems 3. In humans, there are two major UGT subfamilies, UGT1A and UGT2B, the latter of which includes seven active members (UGT2B4, UGT2B7, UGT2B10, UGT2B11, UGT2B15, UGT2B17, and UGT2B28) located on chromosome 4 4. The most important members of this family are UGT2B7 and UGT2B15, which have high activity on steroid hormone 5 and high expression level in liver 3. In these two genes, multiple nonsynonymous SNPs, such as D85Y (rs1902023) and T523K (rs4148269) in UGT2B15, and H268Y (rs7439366) in UGT2B7, have been verified to alter the enzyme activity significantly and are common in human populations 3. Therefore, these SNPs might contribute to breast cancer risk, but their association was still not clear.
To address this issue, we collected 274, 139, and 874 breast cancer patients and 102, 331, and 418 women without breast cancer from the United States, Barbados, and Nigeria, respectively, all of which were of African ancestry. The demographics and environmental risk factor information for these populations could be found in our recent publication 6. The three SNPs, rs1902023, rs4148269, and rs7439366, were genotyped by commercial TaqMan assays C_27028164_10, C_9440184_20, and a custom TaqMan assay (primers GGAAAGCTGACGTATGGCTTATT and AAAGCCAACAAAATAAAACCAACA, a n d p r o b e s 6FAM-TCAGTTTCCTCATCCAC and VIC-TCAGTTTCCATATCCAC; target in bold; Applied Biosystems, Foster City, CA), respectively, according to the manufacturer’s protocol. Hardy-Weinberg equilibrium (HWE) was tested using a Chi-square test with one degree of freedom in cases and controls for three populations separately. The genotype frequencies in patients and controls were compared by Chi-square tests, followed by logistic regression to estimate odds ratios (OR) and 95% confidence intervals (CI). Because of allele frequency difference across populations, the logistic regression adjusted an indicator variable for population. All statistical tests were performed in Stata 11.0 (StataCorp LP, College Station, TX) and a P<0.05 was considered statistically significant.
The genotype counts and comparisons between cases and controls for each SNP were listed in Table 1. No deviation from HWE was observed (P>0.05) in any of the three populations (result not shown) for all SNPs. For rs1902023, the allele frequency were ~34%, ~48%, and ~18% for G/G, G/T, and T/T genotype, respectively (Table 1a), which was consistent among the three populations surveyed and was close to the published value for African populations 7. No significant difference in allele frequency between cases and controls was observed in any of the populations (P>0.44 in all populations, Table 1). Similar results were obtained for other two SNPs (Table 1b and 1c).
Table 1a.
Genotype distribution of rs1902023 in women with breast cancer and healthy controls.
| Population | Genotype | Cases (%) | Controls (%) | P | OR [95% CI] |
|---|---|---|---|---|---|
| African Americans | G/G | 74 (33.2%) | 40 (40.0%) | 0.449 | 1.00 (ref.) |
| G/T | 102 (45.7%) | 43 (43.0%) | 1.28 [0.76–2.17] | ||
| T/T | 47 (21.1%) | 17 (17.0%) | 1.49 [0.76–2.94] | ||
| Barbados | G/G | 43 (33.9%) | 81 (31.8%) | 0.862 | 1.00 (ref.) |
| G/T | 62 (48.8%) | 132 (51.8%) | 0.88 [0.55–1.43] | ||
| T/T | 22 (17.3%) | 42 (16.5%) | 0.99 [0.52–1.86] | ||
| Nigerians | G/G | 276 (35.0%) | 138 (33.8%) | 0.923 | 1.00 (ref.) |
| G/T | 382 (48.4%) | 201 (49.3%) | 0.95 [0.73–1.24] | ||
| T/T | 131 (16.6%) | 69 (16.9%) | 0.95 [0.66–1.36] | ||
| Total | G/G | 393 (34.5%) | 259 (33.9%) | 0.829 | 1.00 (ref.) |
| G/T | 546 (47.9%) | 376 (49.3%) | 0.96 [0.78–1.17] | ||
| T/T | 200 (17.6%) | 128 (16.8%) | 1.03 [0.78–1.35] |
Table 1b.
Genotype distribution of rs4148269 in women with breast cancer and healthy controls.
| Population | Genotype | Cases (%) | Controls (%) | P | OR [95% CI] |
|---|---|---|---|---|---|
| African Americans | A/A | 175 (65.1%) | 59 (58.4%) | 0.491 | 1.00 (ref.) |
| A/C | 88 (32.7%) | 39 (38.6%) | 0.76 [0.47–1.23] | ||
| C/C | 6 (2.2%) | 3 (3.0%) | 0.67 [0.16–2.78] | ||
| Barbados | A/A | 88 (64.7%) | 216 (65.9%) | 0.971 | 1.00 (ref.) |
| A/C | 43 (31.6%) | 100 (30.5%) | 1.06 [0.68–1.63] | ||
| C/C | 5 (3.7%) | 12 (3.7%) | 1.02 [0.35–2.99] | ||
| Nigerians | A/A | 665 (77.3%) | 315 (77.6%) | 0.300 | 1.00 (ref.) |
| A/C | 179 (20.8%) | 88 (21.7%) | 0.96 [0.72–1.29] | ||
| C/C | 16 (1.9%) | 3 (0.7%) | 2.52 [0.73–8.73] | ||
| Total | A/A | 928 (73.4%) | 590 (70.7%) | 0.383 | 1.00 (ref.) |
| A/C | 310 (24.5%) | 227 (27.2%) | 0.87 [0.71–1.06] | ||
| C/C | 27 (2.1%) | 18 (2.2%) | 0.95 [0.52–1.75] |
Table 1c.
Genotype distribution of rs7439366 in women with breast cancer and healthy controls.
| Population | Genotype | Cases (%) | Controls (%) | P | OR [95% CI] |
|---|---|---|---|---|---|
| African Americans | C/C | 138 (51.9%) | 39 (39.0%) | 0.069 | 1.00 (ref.) |
| C/T | 108 (40.6%) | 49 (49.0%) | 0.62 [0.38–1.02] | ||
| T/T | 20 (7.5%) | 12 (12.0%) | 0.47 [0.21–1.05] | ||
| Barbados | C/C | 66 (49.6%) | 160 (50.3%) | 0.566 | 1.00 (ref.) |
| C/T | 60 (45.1%) | 133 (41.8%) | 1.09 [0.72–1.66] | ||
| T/T | 7 (5.3%) | 25 (7.9%) | 0.68 [0.28–1.65] | ||
| Nigerians | C/C | 438 (55.2%) | 205 (57.9%) | 0.697 | 1.00 (ref.) |
| C/T | 311 (39.2%) | 131 (37.0%) | 1.11 [0.85–1.45] | ||
| T/T | 44 (5.6%) | 18 (5.1%) | 1.14 [0.65–2.03] | ||
| Total | C/C | 642 (53.9%) | 404 (52.3%) | 0.544 | 1.00 (ref.) |
| C/T | 479 (40.2%) | 313 (40.5%) | 0.96 [0.80–1.16] | ||
| T/T | 71 (5.9%) | 55 (7.1%) | 0.81 [0.56–1.18] |
There are multiple possible explanations for these results. One is that the change in enzyme activity resulting from the amino acid substitution is not large enough to influence steroid hormone levels in the human body. Alternatively, nearby regulatory SNPs in linkage disequilibrium (LD) with the nonsynonymous ones may compensate for the activity differences due to the amino acid change. For example, 85Y allele of UGT2B15 shows almost twice the enzyme activity as 85D 3, but it is in nearly complete LD with a relatively low activity promoter 7. As a consequence, the total glucuronidation activity for UGT2B15 might still be similar between the two haplotypes. This was also observed for two other nonsynonymous SNPs 8–10, highlighting the importance of considering the functional effect of the haplotype instead of single SNPs. Another possibility is that we do not sufficient power to detect an association; however, the effect of the variation on breast cancer risk would have to be quite low.
Acknowledgments
This research was supported by National Institutes of Health (CA125183 and U01 GM61393 to A.D.R.).
Footnotes
The authors declare no conflict of interest.
References
- 1.Cauley JA, Lucas FL, Kuller LH, Stone K, Browner W, Cummings SR. Elevated serum estradiol and testosterone concentrations are associated with a high risk for breast cancer. Study of Osteoporotic Fractures Research Group. Ann Intern Med. 1999;130:270–7. doi: 10.7326/0003-4819-130-4_part_1-199902160-00004. [DOI] [PubMed] [Google Scholar]
- 2.Nagar S, Remmel RP. Uridine diphosphoglucuronosyltransferase pharmacogenetics and cancer. Oncogene. 2006;25:1659–72. doi: 10.1038/sj.onc.1209375. [DOI] [PubMed] [Google Scholar]
- 3.Guillemette C. Pharmacogenomics of human UDP-glucuronosyltransferase enzymes. Pharmacogenomics J. 2003;3:136–58. doi: 10.1038/sj.tpj.6500171. [DOI] [PubMed] [Google Scholar]
- 4.Mackenzie PI, Bock KW, Burchell B, Guillemette C, Ikushiro S, Iyanagi T, Miners JO, Owens IS, Nebert DW. Nomenclature update for the mammalian UDP glycosyltransferase (UGT) gene superfamily. Pharmacogenet Genomics. 2005;15:677–85. doi: 10.1097/01.fpc.0000173483.13689.56. [DOI] [PubMed] [Google Scholar]
- 5.Turgeon D, Carrier JS, Levesque E, Hum DW, Belanger A. Relative enzymatic activity, protein stability, and tissue distribution of human steroid-metabolizing UGT2B subfamily members. Endocrinology. 2001;142:778–87. doi: 10.1210/endo.142.2.7958. [DOI] [PubMed] [Google Scholar]
- 6.Sun C, Huo D, Southard C, Nemesure B, Hennis A, Leske MC, Wu S, Witonsky D, Olopade O, Di Rienzo A. A signature of balancing selection in the region upstream to the human UGT2B4 gene and implications for breast cancer risk. Hum Genet. 2011 doi: 10.1007/s00439-011-1025-6. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sun C, Southard C, Witonsky DB, Olopade O, Di Rienzo A. Allelic imbalance (AI) identifies novel tissue specific cis-regulatory variation for human UGT2B15. Hum Mutat. 2010;31:99–107. doi: 10.1002/humu.21145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sun C, Southard C, Olopade O, Di Rienzo A. Differential allelic expression of c. 1568C>A at UGT2B15 is due to variation in a novel cis-regulatory element in the 3′UTR. Gene. 2011;481:24–8. doi: 10.1016/j.gene.2011.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Duguay Y, Báár C, Skorpen F, Guillemette C. A novel functional polymorphism in the uridine diphosphate–glucuronosyltransferase 2B7 promoter with significant impact on promoter activity. Clinical Pharmacology & Therapeutics. 2004;75:223–33. doi: 10.1016/j.clpt.2003.10.006. [DOI] [PubMed] [Google Scholar]
- 10.Nakamura A, Nakajima M, Higashi E, Yamanaka H, Yokoi T. Genetic polymorphisms in the 5′-flanking region of human UDP-glucuronosyltransferase 2B7 affect the Nrf2-dependent transcriptional regulation. Pharmacogenet Genomics. 2008;18:709–20. doi: 10.1097/FPC.0b013e32830500c9. [DOI] [PubMed] [Google Scholar]