American Journal of Epidemiology Advance Access originally published online on June 27, 2007
American Journal of Epidemiology 2007 166(5):576-581; doi:10.1093/aje/kwm113
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ORIGINAL CONTRIBUTIONS |
Lactase Persistence and Bitter Taste Response: Instrumental Variables and Mendelian Randomization in Epidemiologic Studies of Dietary Factors and Cancer Risk
1 Centre for Cancer Epidemiology and Prevention (CPO Piemonte), Turin, Italy
2 Institute for Scientific Interchange (ISI) Foundation, Turin, Italy
3 MRC Centre for Causal Analyses in Translational Epidemiology, Department of Social Medicine, University of Bristol, Bristol, United Kingdom
4 Epidemiology Unit, National Cancer Institute, Milan, Italy
5 Molecular and Nutritional Epidemiology Unit, Centre for the Study of Cancer Prevention (CSPO), Scientific Institute of Tuscany, Florence, Italy
6 Department of Clinical and Experimental Medicine, Federico II University, Naples, Italy
7 Ragusa Cancer Registry, Ragusa, Italy
8 Department of Epidemiology and Public Health, Imperial College London, London, United Kingdom
Correspondence to Dr. Paolo Vineis, Department of Epidemiology and Public Health, Imperial College London, Norfolk Place, London W2 1PG, United Kingdom (e-mail: p.vineis{at}imperial.ac.uk).
Received for publication October 16, 2006. Accepted for publication March 7, 2007.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Consumption of dairy products seems to increase the risk of cancer at several sites, while intake of cruciferous vegetables could have protective effects. However, these dietary intakes are subject to measurement error, and associations with cancer could be due to confounders. Mendelian randomization has been suggested as a way to overcome confounding by exploiting the random allocation of alleles from parents to offspring. In mid-2006, the authors conducted a study of allele frequencies for the lactase (LCT) and taste receptor, type 2, member 38 (TAS2R38) genes, including 634 volunteers recruited (1992–1998) from the Italian branch of the European Prospective Investigation into Cancer and Nutrition. The authors hypothesized that there would be a lower milk intake among carriers of the LCT CC genotype and a different intake of cruciferous vegetables among carriers of the TAS2R38 variant. Overall, the frequency of the LCT T allele was higher in northern Italy than in southern Italy. Food intake was associated with gene variants. An association was evident for ice cream and LCT variants (p = 0.004); less so for milk intake. In addition, the TAS2R38 variant showed a geographic gradient and an association with cruciferous vegetable intake. These results suggest that the LCT and TAS2R38 variants are good candidates for Mendelian randomization studies of cancer and other health outcomes.
diet; lactase; neoplasms; polymorphism, genetic; taste; T2R taste receptors
Abbreviations: EPIC, European Prospective Investigation into Cancer and Nutrition; LCT, lactase [gene]; TAS2R38, taste receptor, type 2, member 38 [gene]
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
It has been suggested that intake of milk could increase the risk of prostate and testicular cancers (1). The hypothesized mechanism for the development of prostate cancer is the inhibitory effect of calcium on the conversion of 25-hydroxyvitamin D to 1,25-dihydroxyvitamin D. The latter has an antiproliferative effect in human prostate cells (2). The presence of female hormones in milk could be a determinant of the relation between dairy product intake and testicular cancer (3). Intake of cruciferous vegetables, possibly through the induction of phase II metabolizing enzymes, could have a protective effect against cancer at several sites, including the prostate, lung, and colorectum (4).
However, both milk and cruciferous vegetable intakes are subject to considerable measurement error, and the association with cancer could be due to confounders such as exposures/behaviors associated with social class. Mendelian randomization has been proposed (5) as a way to overcome confounding by exploiting the random allocation of alleles from parents to offspring. The association between a gene variant and a disease is not subject to the confounding by behavioral or socioeconomic factors that has clearly led to misleading findings in conventional observational epidemiologic studies, nor does reverse causation or other biases inherent in observational research apply to studies based on Mendelian randomization (6). If one can show that a genetic variant (e.g., one related to lactase persistence) that affects milk intake is associated with cancer, this will be indirect but unconfounded evidence for a role of milk in carcinogenesis. The same applies to a variant of a gene that influences intake of cruciferous vegetables (e.g., the taste receptor, type 2, member 38 (TAS2R38) gene).
Lactase persistence is a dominant trait controlled by the lactase (LCT) gene. People who are homozygous for the recessive C allele of the intron C > T (position –13910) polymorphism (rs4988235) have almost undetectable levels of intestinal lactase production compared with people with LCT TC and LCT TT (7–11). For evolutionary reasons that are not totally clear, humans generally undergo a physiologic shutdown of lactase activity after weaning (8). However, in some populations, a mutation occurred—apparently around 5,000–10,000 years ago—that led to lactase persistence. People with lactase persistence are much more common in Northern Europe; for example, the frequency of the T allele is 80 percent in the United Kingdom but only 10–20 percent in Southern Europe (9). Grand et al. (10) have found a haplotype covering approximately 1 million nucleotide bases that includes the LCT gene. In spite of the unusual block length (not being broken down by meiotic recombination), the presence of a shared haplotype and the presence of the same DNA variants in nonpersistence alleles in different, distantly related populations together suggest that the persistence variant is relatively old (11). Enattah et al. (11) have hypothesized that this haplotype underwent strong selective pressure around 5,000–10,000 years ago, when dairy farming arose. A separate origin for lactase persistence in African populations has recently been demonstrated (12).
Phenylthiocarbamide is a bitter chemical whose bitter taste is not appreciated by approximately 25 percent of the population. The TAS2R38 gene is a member of the bitter taste receptor family. Two single nucleotide polymorphisms in the TAS2R38 gene define the bitter taste response: Pro49Ala C > G (rs713598) and Ala262Val C > T (rs1726866). These polymorphisms allow construction of two haplotypes, which refer to taster and nontaster status, respectively (13). The molecular basis of the differential response to phenylthiocarbamide determined by these haplotypes has been recently demonstrated (14). Bitter-tasting compounds in cruciferous vegetables resemble phenylthiocarbamide; therefore, the analyzed TAS2R38 gene variants may be related to cruciferous vegetable intake (15).
Before directly testing the association between LCT and TAS2R38 variants and cancer, we have conducted a study on the geographic distribution of allele frequencies at different locations in Italy and the association between variant alleles and the intake of milk derivatives or cruciferous vegetables. We hypothesized 1) a north-to-south decreasing gradient for the T allele frequency for LCT, based on previous observations of such a gradient in Europe; 2) a lower intake of milk (but not yogurt or cheese, in which lactose is hydrolyzed) among carriers of the CC genotype for LCT (lactase nonpersistence); and 3) an association between cruciferous vegetable intake and variant TAS2R38 haplotypes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The European Prospective Investigation into Cancer and Nutrition (EPIC) is a multicenter European study in which approximately 520,000 healthy volunteers were recruited, between 1992 and 1998, in 10 European countries (Sweden, Denmark, the Netherlands, Norway, the United Kingdom, France, Germany, Spain, Italy, and Greece). EPIC investigators have collected very detailed dietary and lifestyle histories through self-administered questionnaires, plus a 24-hour dietary recall administered through person-to-person interview, anthropologic measurements, and a 30- to 40-ml blood sample. Details on EPIC study procedures are given elsewhere (16).
In mid-2006, we analyzed polymorphisms in the LCT and TAS2R38 genes in 634 healthy subjects randomly sampled from the Italian EPIC cohort. In these 634 persons, we also measured bulky DNA adducts by means of P32-postlabeling. A 5'-nuclease assay (TaqMan) was used to genotype all of the polymorphisms, using fluorogenic MGB (minor groove binder) probes purchased by Applied Biosystems (Foster City, California). (Information on probes and polymerase chain reaction conditions for genotyped single nucleotide polymorphisms can be obtained from the authors upon request.) We analyzed the TAS2R38 Pro49Ala (C > G) and Ala262Val (T > C) missense polymorphisms and the LCT T/C intron polymorphism.
Haplotypes were reconstructed and their frequencies were estimated from genotype data using a Bayesian statistical method based on the coalescence theory, implemented in the PHASE package, version 2.0 (17); individual phases were assigned to each participant through the use of default settings. TAS2R38 AV/AA haplotypes with lower frequency and undefined taster status have been excluded.
We computed mean and median intakes of milk, yogurt, cheese, butter, and ice cream according to LCT variants and estimated the statistical strength of the differences by analysis of variance. We repeated the same analyses for cruciferous vegetables and TAS2R38 haplotypes, estimating the statistical strength of the differences using Wilcoxon's two-sample test. Hardy-Weinberg equilibrium was evaluated by chi-squared test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Both the LCT and the TAS2R38 genotypes were in Hardy-Weinberg equilibrium. Table 1 shows the distribution of LCT and TAS2R38 haplotypes by EPIC study center. Overall, the frequency of the LCT gene T allele was approximately 21 percent in northern Italy and 9 percent in southern Italy, that is, much lower than in Northern Europe. In particular, the TT genotype was very rare in southern Italy (around 2 percent). The regional differences were less pronounced for the TAS2R38 gene. As would be expected, the area of birth and area of residence of the EPIC participants were strong predictors of LCT genotypes and TAS2R38 haplotype variants (table 2). Interestingly, migrants from southern Italy who were living near the northern study centers also had an uneven distribution of alleles, particularly for LCT: Twenty-nine (78.3 percent) subjects born in the south had the CC genotype, while eight (21.6 percent) had the CT genotype and none had the TT genotype; corresponding numbers for persons born in the north were 98 (57.6 percent), 51 (30.0 percent), and 21 (12.3 percent), respectively (p = 0.024). For TAS2R38 haplotypes, the distributions were 29 (78.3 percent) tasters and eight (21.6 percent) nontasters for those born in the south and 107 tasters (64.8 percent) and 58 nontasters (35.1 percent) for those born in the north (p = 0.11).
|
|
Tables 3 and 4 show mean and median intakes of relevant foods by genetic variant. A clear association was evident for ice cream and LCT variants (p = 0.004); less so for milk intake. As expected, no difference was found for yogurt and cheese, since lactose is hydrolyzed in these products. An association was also evident for intake of cruciferous vegetables and TAS2R38 haplotypes (p = 0.02).
|
|
The association between dietary intakes and genotypes was not explained by population stratification or other characteristics of local populations. As tables 5 and 6 show, intakes of the relevant foods differed between genotypes at each of the study centers involved. For LCT, apparently only the TT genotype, and not the heterozygous genotype, was associated with different intakes of dairy products.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Our study confirms the existence of a considerable difference between Northern and Southern Europe for the frequency of the T allele of the LCT gene, also suggesting a gradient for the LCT variant within Italy. In addition, it confirms that persons with the LCT CC genotype have a lower intake of dairy products, particularly ice cream. The latter observation has the possible explanation that ice cream intake might be less susceptible to classification error than milk intake. As predicted, no association was found for cheese and yogurt. Most studies have found that lactase persistence is related to milk consumption (18–24), although other studies, generally with small samples, have observed no association (25–27).
The data showed an association of the TAS2R38 variants with intake of cruciferous vegetables, as hypothesized. The direction of the association was clear, since people carrying the haplotypes associated with the inability to taste bitter compounds (PA/PA, AV/PA, and AA/PA), which make cruciferous vegetables more palatable, had significantly greater mean and median intakes of cruciferous vegetables. A study of food preferences in children found that tasters selected fewer vegetables to consume than nontasters (28). In a study of 4,286 British women, Timpson et al. (29) found no association between taster status and consumption of green vegetables. Timpson et al. suggested that the lack of association could have been related to a British "debittering" approach to cooking, with long boiling periods and the addition of salt, sugar, or fat that reduces the bitter taste of vegetables and makes them more palatable (29).
In conclusion, our study suggests that the LCT and TAS2R38 gene variants have a geographic gradient and are associated with the expected phenotypes. Therefore, they are good candidates for Mendelian randomization studies of cancer and other health outcomes. However, the relatively weak associations between the variants and the exposures of interest mean that large study samples will be required when relating these variants to disease outcomes and imputing causal effects of the proxy exposures. In addition, given the geographic distribution of the variants across Italy (and other areas), investigators will need to take population stratification into account when relating the variants to disease outcomes (30).
|
ACKNOWLEDGMENTS |
|---|
This work was made possible by grants from the Compagnia di San Paolo to the ISI Foundation and from the Associazione Italiana per la Ricerche sul Cancro to Dr. Giuseppe Matullo.
Conflict of interest: none declared.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
- Ganmaa D, Li XM, Wang J, et al. Incidence and mortality of testicular and prostatic cancers in relation to world dietary practices. Int J Cancer (2002) 98:262–7.[CrossRef][Web of Science][Medline]
- Skowronski RJ, Peehl DM, Feldman D. Vitamin D and prostate cancer: 1,25 dihydroxyvitamin D3 receptors and actions in human prostate cancer cell lines. Endocrinology (1993) 132:1952–60.
[Abstract/Free Full Text] - Ganmaa D, Wang PY, Qin LQ, et al. Is milk responsible for male reproductive disorders? Med Hypotheses (2001) 57:510–14.[CrossRef][Web of Science][Medline]
- Murillo G, Mehta RG. Cruciferous vegetables and cancer prevention. Nutr Cancer (2001) 41:17–28.[CrossRef][Web of Science][Medline]
- Davey Smith G, Ebrahim S. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol (2003) 32:1–22.
[Abstract/Free Full Text] - Smith GD, Ebrahim S. Mendelian randomization: prospects, potentials, and limitations. Int J Epidemiol (2004) 33:30–42.
[Free Full Text] - Gugatschka M, Dobnig H, Fahrleitner-Pammer A, et al. Molecularly-defined lactose malabsorption, milk consumption and anthropometric differences in adult males. QJM (2005) 98:857–63.
[Abstract/Free Full Text] - Swallow DM. Genetics of lactase persistence and lactose intolerance. Annu Rev Genet (2003) 37:197–219.[CrossRef][Web of Science][Medline]
- Weiss KM. The unkindest cup. Lancet (2004) 363:1489–90.[CrossRef][Web of Science][Medline]
- Grand RJ, Montgomery RK, Chitkara DK, et al. Changing genes; losing lactase. Gut (2003) 52:617–19.
[Free Full Text] - Enattah NS, Sahi T, Savilahti E, et al. Identification of a variant associated with adult-type hypolactasia. Nat Genet (2002) 30:233–7.[CrossRef][Web of Science][Medline]
- Tishkoff SA, Reed FA, Ranciaro A, et al. Convergent adaptation of human lactase persistence in Africa and Europe. Nat Genet (2007) 39:31–40.[CrossRef][Web of Science][Medline]
- Kim UK, Jorgenson E, Coon H, et al. Positional cloning of the human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide. Science (2003) 299:1221–5.
[Abstract/Free Full Text] - Bufe B, Breslin PA, Kuhn C, et al. The molecular basis of individual differences in phenylthiocarbamide and propylthiouracil bitterness perception. Curr Biol (2005) 15:322–7.[CrossRef][Web of Science][Medline]
- Bufe B, Meyerhof W. Taste receptors and their variants. In: Nutritional genomics: impact on health and disease—Brigelius-Flohé R, Joost H-G, eds. (2006) Weinheim, Germany: Wiley-VCH Verlag GmbH & Company. 386–411.
- Riboli E, Hunt KJ, Slimani N, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection. Public Health Nutr (2002) 5:1113–24.[CrossRef][Web of Science][Medline]
- Stephens M, Smith NJ, Donnelly P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet (2001) 68:978–89.[CrossRef][Web of Science][Medline]
- Obermayer-Pietsch BM, Bonelli CM, Walter DE, et al. Genetic predisposition for adult lactose intolerance and relation to diet, bone density, and bone fractures. J Bone Miner Res (2004) 19:42–7.[CrossRef][Web of Science][Medline]
- Newcomer AD, Gordon H, Thomas PJ, et al. Family studies of lactase deficiency in the American Indian. Gastroenterology (1977) 73:985–8.[Web of Science][Medline]
- Honkanen R, Pulkkinen P, Jarvinen R, et al. Does lactose intolerance predispose to low bone density? A population-based study of perimenopausal Finnish women. Bone (1996) 19:23–8.[Medline]
- Horowitz M, Wishart J, Mundy L, et al. Lactose and calcium absorption in postmenopausal osteoporosis. Arch Intern Med (1987) 147:534–6.
[Abstract/Free Full Text] - Newcomer AD, Hodgson SF, McGill DB, et al. Lactase deficiency: prevalence in osteoporosis. Ann Intern Med (1978) 89:218–20.
[Abstract/Free Full Text] - Rasinpera H, Savilahti E, Enattah NS, et al. A genetic test which can be used to diagnose adult-type hypolactasia in children. Gut (2004) 53:1571–6.
[Abstract/Free Full Text] - Mainguet P, Faille I, Destrebecq L, et al. Lactose intolerance, calcium intake, and osteopenia. Lancet (1991) 338:1156–7.[Web of Science][Medline]
- Stephenson LS, Latham MC. Lactose intolerance and milk consumption: the relation of tolerance to symptoms. Am J Clin Nutr (1974) 27:296–303.[Abstract]
- Corazza GR, Benati G, Di Sario A, et al. Lactose intolerance and bone mass in postmenopausal Italian women. Br J Nutr (1995) 73:479–87.[CrossRef][Web of Science][Medline]
- Harma M, Alhava E. Is lactose malabsorption a risk factor in fractures of the elderly? Ann Chir Gynaecol (1988) 77:180–3.[Web of Science][Medline]
- Bell KI, Tepper BJ. Short-term vegetable intake by young children classified by 6-n-propylthoiuracil bitter-taste phenotype. Am J Clin Nutr (2006) 84:245–51.
[Abstract/Free Full Text] - Timpson NJ, Christensen M, Lawlor DA, et al. TAS2R38 (phenylthiocarbamide) haplotypes, coronary heart disease traits, and eating behavior in the British Women's Heart and Health Study. Am J Clin Nutr (2005) 81:1005–11.
[Abstract/Free Full Text] - Campbell CD, Ogburn EL, Lunetta KL, et al. Demonstrating stratification in a European American population. Nat Genet (2005) 37:868–72.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Kivimaki, G. D. Smith, N. J. Timpson, D. A. Lawlor, G. D. Batty, M. Kahonen, M. Juonala, T. Ronnemaa, J. S.A. Viikari, T. Lehtimaki, et al. Lifetime body mass index and later atherosclerosis risk in young adults: examining causal links using Mendelian randomization in the Cardiovascular Risk in Young Finns study Eur. Heart J., October 2, 2008; 29(20): 2552 - 2560. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||