Objective:

Variation in the bitter-taste receptor gene, TAS2R38 confers the ability to taste 6-n-propylthiouracil (PROP). The objective of this study was to relate TAS2R38 haplotypes and PROP-tasting phenotypes to adiposity in a genetically isolated population. We hypothesized that the nontaster phenotype would be associated with higher BMI and waist circumference (WC) in females, and that dietary restraint would mediate this relationship.

Methods and Procedures:

Participants were 540 healthy inhabitants of the genetically isolated village of Carlantino in southern Italy who were 15–89 years of age at the time of the study. Haplotype analyses were performed and PROP tasting was assessed using a filter paper method. Height, weight, and WC were measured and restrained eating was assessed using a brief questionnaire.

Results:

Nontaster females had higher BMI and WC than females who were phenotypic tasters, and this relationship was specific to females with low dietary restraint. Regression analysis showed that BMI declined by 1.7 units across taster groups in females when the model included the PROP by restraint interaction. PROP phenotype was not significantly associated with WC in the regression models. Polymorphisms in TAS2R38 were not associated with BMI or WC in females. Neither TAS2R38 haplotype nor PROP phenotype was strongly related to BMI or WC in males.

Discussion:

These data support previous findings of a relation between the nontaster phenotype and higher BMI in females that is modified by dietary restraint. Assessment of PROP phenotypes might provide unique information about adiposity that is not captured by haplotype analysis alone.

Introduction

The ability to detect bitter toxins from plants conferred important survival advantages to early human societies (1). Humans possess a repertoire of 25 functional bitter-taste-receptor genes that are located in clusters on chromosomes 5p, 7q, and 12p (2). The most studied of these genes is TAS2R38, which controls the ability to taste the glucosinolates, a large family of bitter-tasting compounds that are widely distributed in plants of the Brassica sp. The synthetic compounds, phenylthiocarbamide (PTC) and 6-n-propylthiouracil (PROP) have been identified as major ligands for this receptor (3).

Genetic diversity in the ability to taste PTC and PROP was serendipitously discovered by Fox, who reported that 25% of the population was taste blind to these substances (4). Other work indicated that the trait may be transmitted in a nearly Mendelian recessive manner (5) and a large number of studies conducted in populations around the globe suggested the presence of "taster" and "nontaster" allelic groups in virtually every group that has been studied (6).

TAS2R38 exists in seven allelic forms, but only the taster and nontaster forms are commonly found outside sub-Saharan Africa (1). Three single-nucleotide polymorphisms result in three amino acid substitutions to the protein at positions A49P (rs713598), A262V (rs1726866), and V296I (rs10246939), giving rise to PAV (the taster variant) and AVI (the nontaster variant). PROP-sensitive individuals possess one or two dominant alleles (PAV/PAV or PAV/AVI), whereas insensitive individuals are recessive for the trait (AVI/AVI) (3,7). The occurrence of other variants (AAV and PVI) is rare and is associated with intermediate sensitivities to PTC and PROP.

Psychophysical studies have revealed that PAV individuals can be divided into two phenotypic subgroups—medium tasters (PAV/AVI) who perceived moderate intensity from PTC/PROP and super-tasters (PAV/PAV) who perceived these compounds as extremely bitter (3,8). Among whites (the most well-studied group), the approximate distribution of nontasters, medium tasters, and super-tasters is 30, 45, and 25%, respectively (8,9,10). The expression of the phenotype is strongly influenced by gender such that women are more sensitive to PTC/PROP and are more likely to be super-tasters than men (8). Other factors that influence the sense of taste such as aging and the presence of oral disease are suspected to play a role (11,12), but these factors have not been well studied.

A growing literature suggests that the ability to taste PTC/PROP influences dietary behavior. Variation in taste sensitivity to PTC/PROP has been associated with differences in preferences for and selection of bitter fruits and vegetables, as well as sweet foods, added fats, spicy foods, and alcoholic beverages (12,13,14,15,16,17,18). We reasoned that if nontasters liked a greater variety of foods than tasters, they might be expected to consume more energy and maintain higher body weights than tasters (19). Although evidence that nontasters habitually consume more fat and/or energy than tasters is sparse at this time (14,20), we have found a strong association between nontaster status and greater adiposity in women. Two studies in small, outbred populations in the United States revealed that nontaster women had a higher mean BMI than super-taster women (9,21). Nevertheless, the role of this phenotype in body weight variation remains controversial because some studies have not shown this association (22,23).

Cognitive factors are major determinants of dietary behavior and body weight in women (24,25,26). One of our studies cited previously (9) showed that dietary restraint (the conscious control of food intake to control body weight) modified the relationship between PROP status and BMI in women. Specifically, a negative association was observed between BMI and PROP status in women with low dietary restraint, as anticipated, but no association was observed between PROP status and BMI in women with high-dietary restraint. These findings suggested that high-dietary restraint might reduce or eliminate the relation between PROP status and BMI by overriding the influence of taste on food selection and dietary behavior.

Genetically and geographically isolated populations provide distinct advantages for studying the influence of genes on complex diseases such as obesity. Such populations exhibit a relatively uniform genetic background coupled with low environmental variation, factors that tend to favor the detection of genetic associations with disease outcomes. We have been studying the small isolated village of Carlantino, situated in southern Italy. In this genetically and culturally homogenous population, a large proportion of individuals share the same trait-predisposing gene inherited from a common ancestor (27).

The objective of this study was to relate TAS2R38 haplotypes and PROP-tasting phenotypes to adiposity in this population. Based on our previous work, we hypothesized that the nontaster phenotype would be associated with higher BMI and waist circumference (WC) in females and that restrained eating would diminish the influence of this phenotype on adiposity.

Methods and Procedures

Study population. Carlantino is a small village in the Province of Foggia in southern Italy that was settled by a few founders at the end of sixteenth century. The census in 1,595 counted 10 households in the village. Carlantino has a present-day population of 1,519 inhabitants, and three different surnames account for the majority of living individuals. The endogamy rate, calculated during past century, was 99.5%. Genetic analyses of chromosome Y haplotypes as well as mitochondrial DNA show that Carlantino is a genetically homogeneous population and not only a geographically isolated village (27). A total of 1,180 healthy individuals were eligible for the study, and participation was random and not family-based. However, since the endogamy rate was high, the participants were likely to be related.

Subjects gave their written informed consent for participating in these studies. The protocol was approved by the local administration of Carlantino, the Health Service of Foggia Province, Italy, and the Rutgers University Institutional Review Board. We certify that all applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during this research.

PROP tasting. All generally healthy individuals who were 15–90 years of age were invited to participate in the PROP screening. Those with physical or cognitive impairments (e.g., senile dementia) that could interfere with the ability to complete the testing were excluded. PROP taste sensitivity was assessed using a filter paper method developed and validated previously (10). The method involves each subject tasting two filter paper disks; the first is impregnated with 1.0 mol/l NaCl (VWR Scientific, Bridgeport, NJ), and the second is impregnated with 50 mmol/l 6-n-2-propylthiouracil (cat. no. P3755; Sigma-Aldrich, St Louis, MO). The subject places each disk on the tip of the tongue until it is wet and rates the intensity of the taste using the labeled magnitude scale (28). The labeled magnitude scale is a 100-mm scale with label descriptors placed at quasi-logarithmic intervals along length of the scale from "barely detectable" to "strongest imaginable". Instructions for using the scale were identical to those described by Green et al. (28). Accordingly, "strongest imaginable" was defined as the most intense oral sensation an individual has experienced in everyday life.

Based on our previous work (10), we used labeled magnitude scale numerical cutoff scores of >67 and <13 to identify super-tasters and nontasters, respectively. Individuals who rated the intensity of PROP between these values were identified as medium tasters. NaCl ratings were used as a reference standard for classifying those who gave a borderline rating to PROP. This comparison is based on the rationale that super-tasters give higher ratings to PROP than NaCl, medium tasters give the two compounds similar ratings, and nontasters give higher ratings to NaCl than to PROP (29).

Demographic and personal characteristics. Dietary restraint was assessed using three true–false statements from the cognitive restraint subscale of the three-factor eating questionnaire (30). Responses consistent with the restrained eating concept were scored 1 point each, with total scores ranging from 0 to 3. Participants were categorized as restrained eaters (total score 2) or unrestrained eaters (total score <2). We previously validated a brief, restrained eating questionnaire comprised of six true–false items from the three-factor eating questionnaire (31). The three items used here were taken from that brief instrument, and all showed high correlations with scores from the full restraint subscale (r = 0.53–0.71; P 0.001) (31). Disinhibition was assessed in a similar manner by selecting three true–false items from the disinhibition subscale of the three-factor eating questionnaire. This brief questionnaire is shown in the Supplementary Table S1 online. As no associations were observed between disinhibition and BMI in the Carlantino sample, this characteristic was not further pursued.

Demographics and lifestyle information (smoking, alcohol consumption, and exercise frequency) were collected using an interviewer-administered questionnaire. The response categories are shown in Table 1. Height (m) and weight (kg) were measured and BMI (kg/m²) was calculated from these measurements. Body weight was measured to the nearest 0.25 kg using a balance-beam scale, and height was measured to the nearest 0.25 cm using a stadiometer. WC was measured midway between the lowest rib and the iliac crest using a tape measure. All measures were taken in light clothing and without shoes.

Table 1 - Subject characteristics.

Full table (56K)

TAS2R38 genetic analysis. Peripheral blood samples were collected and DNA was prepared from blood leucocytes, using a salting-out method (Wizard Genomic Purification, Quiagen, Madison, WI, USA). Single-nucleotide polymorphisms (rs 713598, rs1726866, rs10246939) were selected from http://www.ncbi.nlm.nih.gov/SNP and analyzed using TaqMan probe-based assays (Applied Biosystems, Foster City, CA). PCR primers and TaqMan probes were designed by Applied Biosystems. The PCR were conducted in 384-well PCR plates with 20 l in each well and a final concentration of 20 ng genomic DNA, 300 nmol/l TET probe, 50 nmol/l FAM probe, 50 nmol/l forward primers, and 300 nmol/l reverse primer in 1 universal master mix (Applied Biosystems). The TaqMan PCR were carried out in an automatic thermal cycler (Perkin Elmer 9700, Applied Biosystems) under the following conditions: 50 °C for 2 min, 95 °C for 10 min, 95 °C for 15 min, and 64 °C for 1 min for 40 cycles. After PCR were completed, the plates were read in an ABI 7900 sequence detector, and the results analyzed by allelic discrimination of the sequence detector software (Applied Biosystems).

Statistical analyses. The quantitative transmission disequilibrium test (32), available on the website http://www.sph.umich.edu/csg/abecasis/QTDT, was calculated to verify the association between PROP phenotype and TAS2R38 alleles. ANOVA (using a general linear model) was used to examine differences in the taste intensity ratings as a function of PROP phenotype (nontaster, medium taster, super-taster). Group differences in BMI and WC were assessed using analysis of covariance, adjusting for age. Finally, stepwise, multiple regression was used to model BMI and WC after adjusting for covariates. Independent variables included PROP phenotype and the demographic and lifestyle factors. Continuous variables were used whenever possible. All analyses involving BMI and WC were repeated using TAS2R38 haplotype as the grouping factor. ANOVA and regression analyses were performed using Statistical Analysis Software, version 10.1 (SAS Institute, Cary, NC). The statistical criterion for all tests was set at P 0.05.

Results

Subject characteristics

A total of 540 participants completed the PROP screening and provided DNA for genetic analysis. Mean age of the study sample was 52.0 0.80 s.e. (range 15–89 years) and 42% of the participants were male. Participant characteristics are shown in Table 1.

Classification by PROP phenotype

As shown in Figure 1, participants were classified into three phenotypic groups (nontasters, medium tasters, and super-tasters) based on numerical criteria established previously (10). As expected, PROP mean intensity ratings were highest for super-tasters, intermediate for medium tasters, and lowest for nontasters, with significant differences among all three groups (P < 0.0001). Mean NaCl ratings also differed among the groups with super-tasters giving higher intensity ratings to NaCl than either medium or nontasters (P < 0.001).

Figure 1.

Classification of participants by PROP taster phenotypes. Three taster groups were identified using previously established cutoff scores for PROP taste intensity (10). NaCl was used as a reference standard as described in the text. Values are means (s.e.m.). Mean ratings within stimulus type with different superscript letters are significantly different at P < 0.001.

Full figure and legend (10K)

PROP phenotype and TAS2R38 haplotype frequencies

Table 2 shows the PROP phenotype and TAS2R38 haplotype frequencies in males, females, and all participants. Phenotypic nontasters comprised 38.3% of the population, and there were comparatively more male nontasters and correspondingly fewer male super-tasters in the population (²_{df = 2} = 45.7; P 0.0001). The prevalence of AVI-recessive individuals was 28.3%, with no difference in prevalence observed between males and females. The rare variants AAV and AAI occurred in 5% and <1% of the population, respectively. The rare PVI variant did not occur in this population. As the AAV variant is associated with intermediate taste sensitivity to PROP (32), individuals with the AAV haplotype (n = 31) were included with the PAV/AVI group for the body weight analyses.

Table 2 - PROP phenotype and TAS2R38 haplotype frequencies.

Full table (40K)

Association between PROP phenotype and TAS2R38 haplotype

Quantitative transmission disequilibrium test revealed strong evidence of linkage between the PROP phenotype and the TAS2R38 locus (² = 7.85; P < 0.005). ANOVA confirmed this association (P < 2.2 10^-16). Polymorphisms in TAS2R38 explained 65% of the variance in PROP taste intensity scores.

PROP phenotype, TAS2R38 haplotype, and adiposity

BMI and WC were available for 451 participants (278 females and 173 males, respectively) for whom anthropometric assessments coincided with the time of the PROP testing. All subsequent analyses are based on this subset. BMI and WC varied by PROP phenotype among females (P < 0.005 and P < 0.01 for BMI and WC, respectively). As shown in Table 3, BMI and WC were higher in nontaster women as compared to the other phenotypic groups (P < 0.01). BMI and WC did not vary by TAS2R38 haplotype in females. In males, there was no relation between PROP phenotype and either BMI or WC. There was a trend for BMI and WC to increase in males as the number of PAV alleles increased, but this effect was not statistically significant.

Table 3 - Age-adjusted, BMI and WC as a function of PROP phenotype or TAS2R38 haplotype.

Full table (26K)

Interaction between dietary restraint and PROP phenotype or TAS2R38 haplotype on adiposity

Figure 2 shows the influence of the interaction between restraint and PROP phenotype on BMI and WC in females and males. Among females, significant interactions were observed between dietary restraint and PROP phenotype on BMI (P < 0.01) and WC (P < 0.02). Among unrestrained females, nontasters had higher BMI (P < 0.01) and WC (P < 0.01) than the other phenotypic groups. There was no relationship between PROP phenotype and either BMI or WC in diet-restrained females. As expected, restrained eating did not interact with PROP phenotype to influence adiposity in males. Moreover, restrained eating did not interact with TAS2R38 haplotype to influence BMI or WC in either gender (data not shown).

Figure 2.

(a) BMI and (b) waist circumference in diet-restrained and not diet-restrained males and females as a function of PROP phenotype. Values are means (s.e.m.). Bars with different letters are different at P < 0.05.

Full figure and legend (40K)

Regression models

Regression analyses were performed to identify independent predictors of BMI and WC; separate models were calculated for males and females. All models included the demographic characteristics as independent variables. In addition to the basic models, PROP phenotype was entered in to Model 1, the interaction between restraint and PROP phenotype was entered into Model 2, and TAS2R38 haplotype was entered into Model 3. As shown in Table 4, age had a strong positive influence on adiposity in all of the models, and the percent of variance explained by the models was generally higher for WC (R² = 0.39–0.41) than for BMI (R² = 0.22–0.31).

Table 4 - Regression models for BMI and WC in males and females.

Full table (41K)

PROP phenotype was negatively associated with BMI in females (Model 1) and PROP phenotype had a stronger negative influence on BMI when the interaction term (restraint PROP) was included in the model (Model 2). In this model, increasing taste responsiveness to PROP was associated with a 1.7 unit decrease in BMI in females. TAS2R38 haplotype was not associated with BMI in females (Model 3). WC in females was positively associated with age and negatively associated with physical activity in all models. Education level was a negative predictor of WC in females in Model 2. Neither PROP phenotype nor TAS2R38 haplotype was associated with WC in women.

Age and TAS2R38 haplotype were positively associated with BMI in males (Model 3). However, age was the only positive predictor of BMI in males when PROP phenotype was entered into the model (Models 1 and 2). Also, age was the only positive predictor of WC in males.

Discussion

The frequency of AVI-recessive individuals in Carlantino was 28.3%, and the uncommon haplotype AAV occurred at a frequency of 5% in this population. These characteristics are similar to those reported for other white populations in the United States and western Europe (7,34), although rare haplotypes were absent in another isolated Italian population in Sardinia (35). The distribution of PROP phenotypic groups in Carlantino was consistent with observations in other white populations. In addition, there were comparatively more female super-tasters but correspondingly more male nontasters in the population. These features are in agreement with published reports (8,10,29). Polymorphisms at the TAS2R38 locus explained 65% of the variation in PROP taste intensity scores, within the range reported previously (7,35). As TAS2R38 polymorphisms explain the majority, but not all, of the phenotypic variation in PROP tasting, it is likely that modifying genes or environmental factors play a role in the expression of the trait. One potential factor is variation in tongue anatomy. Taster status is highly positively correlated with taste bud density at the tip of the tongue and somatosensory (Trigeminal nerve) input (36). Thus, it is possible that alleles controlling taste bud density co-segregate with alleles of TAS2R38 contributing to the overall taste responsiveness to PROP. This possibility deserves further attention.

The agreement between haplotype and phenotype frequencies in our study was somewhat lower in males than females (Table 2). Approximately 30% of males carried two recessive alleles (AVI/AVI), but 48% of males were characterized as nontasters. The relevance of this difference, if any, remains unclear. Previous studies did not report haplotype and phenotype frequencies by gender for comparison purposes (7,35). Adjustment of the PROP ratings for smoking, which was more prevalent in males than in females, did not account for this discrepancy (data not shown). Future studies will need to address this issue.

In females, PROP phenotype was negatively related to both BMI and WC. Age-adjusted, group data revealed that nontaster females were heavier than both medium and super-taster females (Table 3). PROP phenotype also showed the same relation to WC with nontaster females having larger values than the other groups. No relation was observed between PROP phenotype and either BMI or WC among males. These results are consistent with several studies showing a negative association between taster status and BMI in females (9,21,37) and either a small influence or no influence of taster status on BMI in males (38,39).

Our findings disagree with those of another laboratory which reported no association between PROP phenotype and adiposity in a clinical population of older women with breast cancer (23,40) or in a community-based sample of adults (22). The present study most closely matches the design of Drewnowski et al. (22) who also used a filter paper method to screen 18–70-year-old male and female respondents to a mail-in health survey. However, that study employed a different filter paper method than ours, and rating scales that have been criticized for their lack of sensitivity in discriminating taster groups (41). These factors might have led to the identification of unusually low proportions of nontaster men (24%) and women (15%) in that study (22) and hence the lack of association of PROP status with body weight.

We previously reported that dietary restraint mediated the relation between taster status and BMI in middle-aged females (9). We observed a similar interaction between dietary restraint and PROP phenotype on both BMI and WC among females in Carlantino. Specifically, among females with low dietary restraint, BMI, and WC were higher in nontasters as compared to medium and super-tasters. In contrast, among diet-restrained females, BMI, and WC did not differ among PROP-tasting groups. The present findings confirm and extend our earlier observations to a large, multigenerational sample of rural, Italian women who are culturally distinct from the suburban American women we have studied in the past. These findings suggest that dietary restraint may help to clarify the relationship between PROP phenotype and BMI, and this influence cuts across culturally divergent populations of women.

The regression models revealed that after adjusting for covariates, age was the only positive predictor of BMI and WC in males, except for a weak positive influence of TAS2R38 haplotype on BMI in Model 3. In females, however, PROP phenotype and the interaction between PROP phenotype and restraint were negatively associated with BMI. When the interaction between PROP phenotype and restrained eating was included in the model, PROP phenotype was associated with a 1.7 unit decrease in BMI across taster groups. Interestingly, WC in females was no longer influenced by PROP phenotype after adjusting for covariates in the regression model. Rather, WC in females was influenced by more traditional predictors including age, physical activity, and to a lesser extent, education. It is presently unclear why PROP phenotype was related to BMI, an estimate of overall adiposity, but not to WC, an estimate of abdominal adiposity. One explanation could be that central adiposity was not prevalent among females in Carlantino as evidenced by a mean WC of 86.4 0.9 cm, slightly below the risk criterion of 88 cm for white females (42).

The present study showed that PROP phenotype but not TAS2R38 haplotype was related to BMI in females. Our results in females agree with the findings of Timpson et al. (34) who also found no association between alleles of TAS2R38 and BMI, waist-to-hip ratio or cardiovascular risk in a large cohort of elderly British women. However, the earlier study did not report PROP phenotypes (34). Because the current study examined both measures simultaneously, we were able to discern the relative contribution of TAS2R38 alleles and PROP phenotype to adiposity in this cohort. Our data suggest that the PROP phenotype might make a unique contribution to BMI that is not captured by analyzing TAS2R38 polymorphisms alone. This novel finding awaits confirmation in future investigations.

The reasons why the PROP phenotype appears to have a greater influence on adiposity in females in comparison to males is presently unknown. There are examples of other genes that differentially influence obesity in males and females (43,44). For example, polymorphisms in ESR1, the gene that encodes estrogen receptor-, has been associated with adiposity in adult females but not in males (43). Conversely, the glutamine 27 glutamic acid (Gln27Glu) polymorphism of the -2 adrenoceptor gene has been associated with obesity in males but not in females (44). One study suggested that diet can modify the expression of the Gln27Glu polymorphism leading to different sex-specific effects. Martinez et al. (45) reported a greater risk of obesity in female but not in male carriers of the Gln27Glu variant who consumed more carbohydrate in the diet. A gene diet interaction could potentially explain the differential influence of PROP phenotype on BMI in females observed here. We previously reported in children that nontaster females gave higher hedonic ratings to full-fat milk than taster females (14). In that same study, nontaster females also consumed more daily servings of added fats than taster females based on maternal reports (14). These effects were not observed in male children (14). Thus, it is possible that the nontaster phenotype and female gender contribute to higher fat intakes and ultimately to greater weight gain in this group. This mechanism deserves further study.

In conclusion, the present study showed that PROP phenotype but not TAS2R38 haplotype was associated with BMI in females in an inbred cohort, and that dietary restraint further clarified this relationship. Neither PROP phenotype not TAS2R38 haplotype was strongly related to adiposity in males. Finally, we chose to study a genetically and geographically isolated population. Such populations exhibit a relatively uniform genetic background coupled with low environmental variation, factors that tend to favor the detection of associations (46). The associations observed here need to be tested in larger and more heterogeneous study samples to confirm these results and assess their applicability to the general population.

Disclosure

The authors declared no conflict of interest.

References

REFERENCES

1. Wooding S, Kim UK, Bamshad MJ et al. Natural selection and molecular evolution in PTC, a bitter-taste receptor gene. Am J Hum Genet 2004;74: 637–646. | Article | PubMed | ISI | ChemPort |
2. Shi P, Zhang J, Yang H, Zhang YP. Adaptive diversification of bitter taste receptor genes in Mammalian evolution. Mol Biol Evol 2003;20: 805–814. | Article | PubMed | ISI | ChemPort |
3. Bufe B, Breslin PA, Kuhn C et al. The molecular basis of individual differences in phenylthiocarbamide and propylthiouracil bitterness perception. Curr Biol 2005;15: 322–327. | Article | PubMed | ISI | ChemPort |
4. Fox AL. The relationship between chemical constitution and taste. Proc Natl Acad Sci 1932;18: 115–120. | Article | PubMed | ChemPort |
5. Blakeslee A. Genetics of sensory thresholds: taste for phenyl thiocarbamide. Proc Natl Acad Sci 1932;18: 120–130. | Article | PubMed | ChemPort |
6. Guo SW, Reed DR. The genetics of phenylthiocarbamide perception. Ann Hum Biol 2001;28: 111–142. | Article | PubMed | ChemPort |
7. Kim UK, Jorgenson E, Coon H, Leppert M, Risch N, Drayna D. Positional cloning of the human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide. Science 2003;299: 1221–1225. | Article | PubMed | ISI | ChemPort |
8. Bartoshuk LM, Duffy VB, Miller IJ. PTC/PROP tasting: anatomy, psychophysics, and sex effects. Physiol Behav 1994;56: 1165–1171. | Article | PubMed | ISI | ChemPort |
9. Tepper BJ, Ullrich NV. Influence of genetic taste sensitivity to 6-n-propylthiouracil (PROP), dietary restraint and disinhibition on body mass index in middle-aged women. Physiol Behav 2002;75: 305–312. | Article | PubMed | ISI | ChemPort |
10. Zhao L, Kirkmeyer SV, Tepper BJ. A paper screening test to assess genetic taste sensitivity to 6-n-propylthiouracil. Physiol Behav 2003;78: 625–633. | Article | PubMed | ChemPort |
11. Prutkin J, Fisher EM, Etter L et al. Genetic variation and inferences about perceived taste intensity in mice and men. Physiol Behav 2000;69: 161–173. | Article | PubMed | ChemPort |
12. Dinehart ME, Hayes JE, Bartoshuk LM et al. Bitter taste markers explain variability in vegetable sweetness, bitterness, and intake. Physiol Behav 2006;87: 304–313. | Article | PubMed | ChemPort |
13. Tepper BJ, Nurse RJ. Fat perception is related to PROP taster status. Physiol Behav 1997;61: 949–954. | Article | PubMed | ChemPort |
14. Keller KL, Steinmann L, Nurse RJ, Tepper BJ. Genetic taste sensitivity to 6-n-propylthiouracil influences food preference and reported intake in preschool children. Appetite 2002;38: 3–12. | Article | PubMed |
15. Duffy VB, Davidson AC, Kidd JR et al. Bitter receptor gene (TAS2R38), 6-n-propylthiouracil (PROP) bitterness and alcohol intake. Alcohol Clin Exp Res 2004;28: 1629–1637. | Article | PubMed | ChemPort |
16. Ullrich NV, Touger-Decker R, O'Sullivan-Maillet J, Tepper BJ. PROP taster status and self-perceived food adventurousness influence food preferences. J Am Diet Assoc 2004;104: 543–549. | Article | PubMed |
17. 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: 245–251. | PubMed | ChemPort |
18. Yeomans M, Tepper B, Rietzschel J, Prescott J. Human hedonic responses to sweetness: role of taste genetics and anatomy. Physiol Behav 2007;91: 264–273. | Article | PubMed | ChemPort |
19. Tepper BJ. 6-n-Propylthiouracil as a genetic taste marker for fat intake, obesity and chronic disease risk: current evidence and future promise. In: Prescott J, Tepper BJ (eds). Genetic Variation in Taste Sensitivity. Marcel Dekker: New York, 2004, pp. 155–178.
20. Goldstein G, Daun H, Tepper B. Influence of PROP taster status and maternal variables on energy intake and body weight of pre-adolescents. Physiol Behav 2007;90: 809–817. | Article | PubMed | ChemPort |
21. Goldstein GL, Daun H, Tepper BJ. Adiposity in middle-aged women is associated with genetic taste blindness to 6-n-propylthiouracil. Obes Res 2005;13: 1017–1023. | Article | PubMed | ChemPort |
22. Drewnowski A, Kristal A, Cohen J. Genetic taste responses to 6-n-propylthiouracil among adults: a screening tool for epidemiological studies. Chem Senses 2001;26: 483–489. | Article | PubMed | ChemPort |
23. Drewnowski A, Henderson SA, Cockroft JE. Genetic sensitivity to 6-n-propylthiouracil has no influence on dietary patterns, body mass indexes, or plasma lipid profiles of women. J Am Diet Assoc 2007;107: 1340–1348. | Article | PubMed | ChemPort |
24. Lawson OJ, Williamson DA, Champagne CM et al. The association of body weight, dietary intake, and energy expenditure with dietary restraint and disinhibition. Obes Res 1995;3: 153–161. | PubMed | ISI | ChemPort |
25. Lindroos AK, Lissner L, Mathiassen ME et al. Dietary intake in relation to restrained eating, disinhibition, and hunger in obese and nonobese Swedish women. Obes Res 1997;5: 175–182. | PubMed | ISI | ChemPort |
26. Hays NP, Bathalon GP, McCrory MA et al. Eating behavior correlates of adult weight gain and obesity in healthy women aged 55–65 y. Am J Clin Nutr 2002;75: 476–483. | PubMed | ISI | ChemPort |
27. Lanzara C, d'Adamo AP, Montico M et al. Use of an Italian isolated population for studying complex diseases. The Carlantino project: study design and preliminary results. Slovenian J Pub Health, in press.
28. Green BG, Dalton P, Cowart B, Shaffer G, Rankin K, Higgins J. Evaluating the 'Labeled Magnitude Scale' for measuring sensations of taste and smell. Chem Senses 1996;21: 323–334. | Article | PubMed | ISI | ChemPort |
29. Tepper BJ, Christensen CM, Cao J. Development of brief methods to classify individuals by PROP taster status. Physiol Behav 2001;73: 571–577. | Article | PubMed | ChemPort |
30. Stunkard AJ, Messick S. The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res 1985;29: 71–83. | Article | PubMed | ISI | ChemPort |
31. Tepper BJ, Choi Y-S, Nayga RM Jr. Understanding food choice in adult men: influence of nutrition knowledge, food beliefs and dietary restraint. Food Qual Pref 1997;8: 307–317. | Article |
32. Abecasis GR, Cardon LR, Cookson WO. A general test of association for quantitative traits in nuclear families. Am J Hum Genet 2000;66: 279–292. | Article | PubMed | ISI | ChemPort |
33. Timpson NJ, Heron J, Day IN et al. Refining associations between TAS2R38 diplotypes and the 6-n-propylthiouracil (PROP) taste test: findings from the Avon Longitudinal Study of Parents and Children. BMC Genet 2007;8: 51. | Article | PubMed | ChemPort |
34. 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: 1005–1011. | PubMed | ChemPort |
35. Prodi DA, Drayna D, Forabosco P et al. Bitter taste study in a sardinian genetic isolate supports the association of phenylthiocarbamide sensitivity to the TAS2R38 bitter receptor gene. Chem Senses 2004;29: 697–702. | Article | PubMed | ChemPort |
36. Essick GK, Chopra A, Guest S, McGlone F. Lingual tactile acuity, taste perception, and the density and diameter of fungiform papillae in female subjects. Physiol Behav 2003;80: 289–302. | Article | PubMed | ChemPort |
37. Duffy VB, Lucchina LA, Bartoshuk LM. Genetic variation in taste: potential biomarker for cardiovascular risk? In: Prescott J, Tepper BJ (eds). Genetic Variation in Taste Sensitivity. Marcel Dekker: New York, 2004, pp. 195–228.
38. Tepper BJ, Nurse RJ. PROP taster status is related to the perception and preference for fat. Ann NY Acad Sci 1998;855: 802–804. | Article | PubMed | ChemPort |
39. Keller KL, Tepper BJ. Inherited taste sensitivity to 6-n-propylthiouracil in diet and body weight in children. Obes Res 2004;12: 904–912. | Article | PubMed |
40. Drewnowski A, Henderson SA, Hann CS, Berg WA, Ruffin MT. Genetic taste markers and preferences for vegetables and fruit of female breast care patients. J Am Diet Assoc 2000;100: 191–197. | Article | PubMed | ChemPort |
41. Bartoshuk LM. Comparing sensory experiences across individuals: recent psychophysical advances illuminate genetic variation in taste perception. Chem Senses 2000;25: 447–460. | Article | PubMed | ISI | ChemPort |
42. American Heart Association. Diagnosis and Management of the Metabolic Syndrome: An American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement: Executive Summary. Circulation 2005;112: e285–e290. | Article |
43. Okura T, Koda M, Ando F et al. Association of polymorphisms in the estrogen receptor gene with body fat distribution. Int J Obes Relat Metab Disord 2003;27: 1020–1027. | Article | PubMed | ChemPort |
44. Gonzalez Sanchez JL, Proenza AM, Martinez Larrad MT et al. The glutamine 27 glutamic acid polymorphism of the 2-adrenoceptor gene is associated with abdominal obesity and greater risk of impaired glucose tolerance in men but not in women: a population-based study in Spain. Clin Endocrinol (Oxf) 2003;59: 476–481. | Article | PubMed | ChemPort |
45. Martinez JA, Corbalan MS, Sanchez-Villegas A et al. Obesity risk is associated with carbohydrate intake in women carrying the Gln27Glu 2-adrenoceptor polymorphism. J Nutr 2003;133: 2549–2554. | PubMed | ISI | ChemPort |
46. Heutink P, Oostra BA. Gene finding in genetically isolated populations. Hum Mol Genet 2002;11: 2507–2515. | Article | PubMed | ISI | ChemPort |

Acknowledgments

These studies were partially funded by a Busch Biomedical Research Grant to B.J.T. from Rutgers University and by the Italian Ministry of Research, the Italian Ministry of Health, and the Research and Development Unit of Illy Cafe, S.p.A. (to P.G.). We thank Daniel Hoffman for helpful discussions on the statistical analyses.

Supplementary material is linked to the online version of the paper at http://www.nature.com/oby