Genetic variation in TAS2R38 bitterness receptor is associated with body composition in Korean females

Abstract Bitterness-receptor gene TAS2R38 is associated with taste sensitivity and dietary behaviour, and is known to play a critical role in adiposity. However, evidence regarding body composition from a large cohort is lacking. This study aimed to ascertain whether TAS2R38 rs10246939 C > T bitterness genetic variation is associated with body composition in Korean individuals. The TAS2R38 rs10246939 genotypes, anthropometric measurements, and body composition of 1,843 males and 1,801 females from the Korean Genome and Epidemiology Study were analysed. Findings suggested that there was a significant difference in body fat components by TAS2R38 genotype. Furthermore, the bitterness genotype exhibited a positive association with adiposity markers in females. The TT genotype showed greater body mass index, body fat percentage, and degree of obesity than those with the C allele. However, such an association was not observed in males. In conclusion, this study suggests that TAS2R38 rs10246939 is associated with fat tissue markers in Korean females.


Introduction
Humans can taste six primary flavours: bitter, sweet, sour, salty, umami, and recently, fat (Besnard et al. 2016).Bitterness is an important factor in food rejection and acceptance.Bitterness contributes to the preference and consumption of food as well as dietary behaviour via differential sensitivity.Therefore, an individual's differential sensitivity for bitterness may also be associated with diet-related behaviours.For example, individuals who are sensitive to the bitter agonist 6-n-propyl thiouracil (PROP) showed lower acceptance for grapefruit juice and green cruciferous and fresh vegetables (Drewnowski et al. 1997;Drewnowski et al. 2000).Such preferences may lead to lower consumption of foods that are beneficial for health, and may therefore contribute to illnesses, including obesity, cardiovascular disease, or cancer (Aune et al. 2017).
Bitterness is transmitted via the taste 2 receptor (T2R, TAS2R), a type of G-protein-coupled receptor (GPCR).Humans have approximately 25 types of T2R isoforms.T2R38 is a taste receptor protein that senses bitter substances, such as phenylthiocarbamide (PTC) and PROP, containing a thiourea moiety (N-C=S) (Smail 2019;Jeruzal-Świątecka et al. 2020).TAS2R38, which encodes T2R38, is located on chromosome 7q34.Among many polymorphisms in TAS2R38, the most well-known variations are rs713598 (A49P, G > C), rs1726866 (V262A, T > C), and rs10246939 (I296V, T > C) (Kim et al. 2003).These three single nucleotide polymorphisms (SNPs) consist of common PAV and AVI haplotypes (Bufe et al. 2005), and the diplotype determines an individual's differential sensitivity to the bitterness of compounds with a thiourea moiety.Individuals with the PAV haplotype are more sensitive to bitterness (taster), while those with the AVI haplotype appear less sensitive (non-taster).Genetic variations in TAS2R38 are associated with differential drinking, smoking, and dietary behaviour, including consumption of bitter and other sweet tasting foods (Beckett et al. 2014;Choi 2019;Smail 2019;Trius-Soler et al. 2022).
T2R proteins are expressed not only in the mouth, but also multiple tissues, including the brain, thyroid, respiratory organs, upper and lower gastrointestinal tract, bladder, and bone, and functions as a chemical sensor (Lee et al. 2012;Clark et al. 2015;Smail 2019).Previous studies reported that T2Rs are associated with disease risk, including obesity and cancer, with genetic variations as modifiers (Choi et al. 2016;Ortega et al. 2016;Choi et al. 2017;Feeney et al. 2017;Choi 2019).For instance, T2R receptors are reported to be present in the thyroid gland, the centre of endocrine metabolism, and are associated with triiodothyronine and thyroxine production regulation (Clark et al. 2015).Such a regulatory role of T2R, independent of dietary intake, was supported by findings from a Korean thyroid cancer study (Choi et al. 2018).Other Korean epidemiological studies have suggested that the TAS2R38 diplotype mediates the risk of stomach and colorectal cancers (Choi et al. 2016(Choi et al. , 2017)).A growing body of research has been conducted to verify whether TAS2R38 genetic variation is associated with obesity and/or body composition.A recent Spanish study investigated the relationship between PTC phenotypes related to the TAS2R38 gene and body mass index (BMI) and found that supertasters had a lower BMI than non-tasters (Trius-Soler et al. 2022).Another Irish study also reported that TAS2R38 genetic variation was associated with the distribution of overweight/obesity in girls (Feeney et al. 2017).However, most studies were conducted in non-Asian populations, and, in general, used a single index marker, BMI.Studies conducted with a Korean population have suggested that TAS2R38 genetic variation may modify the risk of obesity; however, the findings were mainly ascertained using BMI (Benish and Choi 2023;Choi 2019).BMI is the most applicable marker that represents body adiposity in obesity aetiology studies; however, some limitations remain.BMI represents excessive total body weight, rather than the amount of adipocytes.Therefore, BMI does not precisely differentiate the amount of fat, muscle, and bone mass, which varies greatly across different ethnicity, age, sex, and individual' characteristics (Buss 2014).The clinical importance of obesity is linked to increased fat tissue.Since excess body fat is involved with low-grade chronic inflammation, it leads to several health issues, including cardiovascular disease and type 2 diabetes (Khanna et al. 2022).Given the earlier reports regarding T2R and obesity defined by BMI, research on the association between body composition and TAS2R38 variation would provide in-depth knowledge on obesity aetiology above the sensory mechanism.
This study aimed to investigate whether the bitterness receptor genetic variation TAS2R38 rs10246939 C > T is associated with body composition, especially markers for adiposity, in a Korean population.As described above, the main genetic variation that determines the bitter taste phenotype is the diplotype of TAS2R38 polymorphisms consisting of rs713598, rs1726866, and rs10246939.These three polymorphisms were close to complete linkage disequilibrium and correlation status (D' and r 2 = 1.0, respectively) in Asian populations and healthy Koreans (Machiela and Chanock 2015;Choi et al. 2016).Therefore, this study was conducted using rs10246939 C > T as a genetic modifying factor.Analyses were performed using data from the Korean Genome and Epidemiology Study (KoGES).To address the question above, various adiposity and body composition markers, including body water, muscle and lean body mass, body fat mass, abdominal fat mass, obesity degree, and BMI were tested in relation to TAS2R38 genetic variation.

Cohort description
The KoGES is a nationwide epidemiological study performed by the Korean government organisation, the National Institute of Health of the Korea Centres for Disease Control and Prevention (KCDC).All data including epidemiological and genetic information for this study were obtained from the KCDC.This observational and cross-sectional study was conducted using the Ansan (urban) and Ansung (rural) study, a part of the KoGES.The criteria for selecting participants for the study were as follows: among the 5,493 individuals aged 40-69 years with genomic and epidemiological data, those with no information on lifestyle (n = 312), hypertension (n = 2), body measurements and composition indicators (n = 40), dietary intake (n = 102), and genotype (n = 2) were excluded.Additionally, individuals diagnosed with chronic diseases (n = 1,363, such as diabetes, high blood pressure, cancer, and heart failure) and those with unusual daily intake (<500 kcal or >5,000 kcal, n = 26) were also excluded.Finally, a total of 3,644 individuals (1,843 males and 1,801 females) were included in this project.This study was approved by the Institutional Review Board (40525-202110-BR-063-01) of Keimyung University, Korea.

General characteristics of the study participants
General information for the study participants including sex, age, region, educational and economic level, drinking and smoking status, and physical activity were collected using a questionnaire (Kim and Han 2017).Education level was classified into three groups: "low" for those who graduated from elementary school or less, "middle" for those who graduated from middle school and high school, and "high" for those who graduated from college or more.Household economic level was also classified into three groups: "low, " "middle, " and "high".The smoking and alcohol drinking status of the individuals were classified into three groups: "never, " "past, " and "current." The physical activity of the individuals was calculated in metabolic equivalents of the task (Kim et al. 2021).

Anthropometric and body composition measurements
Body measurements and body composition data of the study individuals were obtained (Kim et al. 2013).Participants were dressed in light-weight clothes and were barefoot, and their height and weight were measured by trained investigators.BMI was calculated by dividing the weight (kg) by the square of height (m 2 ).Waist and hip circumferences were measured in triplicate.The waist-hip ratio (WHR) was calculated by dividing the waist circumference (cm) by the hip circumference (cm).
The obesity status was defined by following the Korean Society for the Study of Obesity guidelines (Seo et al. 2019).Both Korean males and females were classified as overweight if they presented with a BMI between 23.0 and 25.0.Participants with a BMI ≥25.0 were classified as obese.Abdominal obesity was defined as a WHR ≥0.9 cm for male participants, and a WHR ≥0.85 cm for female participants.
Using an automatic multi-frequency bioelectrical impedance analysis technique (MF-BIA; Inbody 3.0, Biospace), the following body composition markers were obtained: intracellular fluid, extracellular fluid, body water, muscle mass, lean body mass, protein, mineral, body fat, fat mass, abdominal fat mass, and obesity degree (Son et al. 2017).

Collection and analyses of dietary data
Participants' dietary consumption was collected using a semi-quantitative food questionnaire (Ahn et al. 2007).Participants were asked to provide information regarding the frequency of average intake and serving size of 103 foods specified on the questionnaire during the last year.Intake frequency was divided into seven levels: "rarely eaten, " "once a month, " "two to three times a month, " "three to four times a week, " "five to six times a week, " "two times a day, " and "three times a day." The serving size was divided into three groups: "small, " "medium, " and "large." The investigated food consumption was converted into the nutrient intake (total calories, calorie nutrients, and dietary fiber) using the Korean Food Composition Table (7th

Statistical analyses
Several studies have reported the sexual-dimorphic association between TAS2R38 gene variation and health outcomes; hence, in this study, male and female participants were analysed separately (Tepper et al. 2008;Feeney et al. 2017;Choi 2019).Continuous variables, such as dietary intake, age, height, weight, BMI, and physical activity, were converted into logged forms before the analyses.Chi-squared tests and a general linear model (GLM) were used to analyse the distribution of the TAS2R38 genotype and the general characteristics of the participants.GLM and linear regression models were also applied to examine the association between TAS2R38 genotype and dietary intake, anthropometric data, and body composition markers, taking account of either the absence or presence of general characteristics of participants.Duncan's method was used for post hoc tests to determine whether any difference existed between the pair of genotypes.The logistic regression analyses were also performed to verify the association between the TAS2R38 genotype and the incidence of obesity or being overweight, taking account of either the absence or presence of general characteristics of participants as covariates.The results were presented with odds ratio and 95% confidence interval (95% CI).Statistical significance was accepted at p < .05,and all analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) and SPSS version 27 (SPSS Inc., Chicago, IL, USA).

Results
A total of 3,644 Koreans (1,843 males and 1,801 females) were analysed in this study, of which 34.2% and 35.8% presented with the CC genotype, 48.0% and 48.1% presented with the CT genotype, and 17.8% and 16.0% presented with the TT genotype in males and females, respectively.Such prevalence of the variation allele T led to an MAF of 0.41 in the entire study population and was similar to earlier Korean studies (Choi 2019).
The general characteristics of the study participants, considering the TAS2R38 rs10246939 genotype, are shown in Table 1.The distribution of the TAS2R38 rs10246939 genotype was not associated with most participants' age, education level, alcohol drinking and tobacco smoking status, and physical activity level in both males and females.However, region and household income levels differed by genotype in females.Additional analyses were performed to verify that TAS2R38 rs10246939 variation is associated with differential dietary intake; however, participants' total calorie, protein, carbohydrate, and fat intake, percentage intake of each macro nutrient of total energy, the ratio for total energy intake, and estimated energy requirement did not differ by bitterness genotype (Table 2).
To address the research question of whether TAS2R38 bitterness genetic variation is associated with Koreans' anthropometric and body composition markers, the data about the three genotypes were compared.Analyses of anthropometric markers showed that there was no clear difference in the examined data among the three genotypes (height, weight, BMI, WC, HC, WHR) in males and females (Supplementary Table 1).However, additional analyses with body composition markers showed that the effect of TAS2R38 genetic variation was evident in the body fat percentage and degree of obesity between genotypes in females (adjusted p = .042and .045,respectively) (Table 3).Other body composition indices obtained from females, including intra/extracellular fluid, body water, muscle and lean body mass, protein, mineral, body fat (kg), and abdominal fat mass (%) were not associated with the genotype.In males, the abovementioned anthropometric variables and most body composition markers did not differ between genotypes.
Previous studies have reported that the effect of variation alleles does not appear only in a doseresponse manner.The recessive model compares the characteristics of each genetic mutation, assuming that having a minor allele homozygote has the greatest effect on a particular disease (Horita and Kaneko 2015).Therefore, a recessive genetic model (CC + CT vs. TT) was applied to understand the effect of bitterness genetic variation on body size and other component variables.The statistical findings with a recessive genetic model suggested that BMI was significantly higher in females with the TT genotype than in females with the C allele (adjusted p = .027,Table 4).In addition, the body fat (%, adjusted p = .049)and degree of obesity (adjusted p = .017)were also significantly higher in females with the TT genotype (Table 5).However, no significant association between anthropometric data and genetic variation was observed in males using this model.Further analyses were performed to verify the more detailed association between TAS2R38 genetic variation and adiposity data observed in this Korean population.Findings suggested that females with the TT genotype displayed an increased BMI of 0.428 (B = 0.428, adjusted r 2 = 0.014, adjusted p = .032)compared to those with genotypes of the C allele.The presence of the rs10246939 variation was also positively associated with body fat (%) and obesity degree.Individuals with the TT genotype displayed a higher body fat percentage of 0.743 and obesity degree of 2.363 (for body fat, B = 0.743, adjusted r 2 = 0.035, adjusted p = .026;and for obesity degree, B = 2.363, adjusted r 2 = 0.029, adjusted p = .019)(Table 6).However, no significant effect of the variant allele was observed in male adiposity markers.
Lastly, logistic regression analyses were performed to understand the association between obesity and the TAS2R38 rs10246939 genotype.The obesity and abdominal obesity defined by BMI and WHR were examined, either the presence or absence of covariates.The results showed that the bitterness genetic variation had no effect on obesity including abdominal obesity (Supplementary Tables 2 and 3).However, additional analyses show the association between body fat and the genotype, when the population was  .042abdominal fat mass (%) 0.9 (0.0) 0.9 (0.0) 0.9 (0.0) .966.9810.9 (0.0) 0.9 (0.0) 0.9 (0.1 grouped according to overweight status (Table 7).Korean females with the TT genotype for TAS2R38 rs10246939 showed an approximately 1.38 times increased likelihood (95% CI: 1.04-1.84) of being overweight/obese (23≤BMI).

Discussion
This study examined whether the TAS2R38 rs10246939 C > T genetic variation is associated with anthropometric and body composition markers in a Korean population.Using the Ansan/Anseong study, KoGES, the study found that TT genotypes were associated with markers of body fat and obesity in females.However, such a genetic effect was not observed in all male data, as well as in female non-fat tissue markers.
Bitterness is an important sensation for the rejection and acceptance of food.Individual bitterness sensitivity varies and is associated with dietary behaviour (Keller et al. 2002).Previous studies have reported that differential bitterness sensitivity is associated with the liking or acceptance of a certain type of food, including cruciferous vegetables, and even sweetness and fat taste (Keller et al. 2002;Smail 2019;Trius-Soler et al. 2022).Since TAS2R38 rs10246939 is a robust genetic prediction marker for such   differential hedonic sensations, the genetic variation may be related to differential nutritional intake, which may further lead to nutrition-associated health issues.However, mixed findings remain.Recent studies in Koreans found that the TAS2R38 rs10246939 genetic variation is not related to caloric or nutrient intake (Choi et al. 2016;Benish and Choi 2023).This study confirmed that there was no significant difference in the consumption of total calories and major nutrients depending on the bitterness genotype.Therefore, the study analysed TAS2R38 genetic variation, anthropometric data, and body composition indices, independent of dietary consumption, and significant results were found in obesity-related indices: BMI, body fat percentage, and obesity degree, in females.Previous studies in Western populations reported such an association between bitterness phenotype or genotype and obesity or adiposity markers.More individuals with TAS2R38 AVI/AVI haplotypes were present in the obese group (Ortega et al. 2016), and children with the PROP phenotype had a differential distribution of obesity classes (Keller et al. 2002).Other studies in Irish and Spanish populations also reported that the distribution of obesity, overweight, and BMI differed significantly according to the TAS2R38 diplotype (Feeney et al. 2017;Trius-Soler et al. 2022).These findings suggest the potential role of T2R38 protein in the mechanism underlying adiposity, independent of immediate dietary regulation via sensory sensitivity.
Previous studies have reported that extra-orally expressed T2R38 receptors may modify the risk of various diseases, such as thyroid function regulation and carcinogenesis, via physiological metabolism, independent of dietary intake (Clark et al. 2015;Choi et al. 2016;Choi et al. 2017).Such a physiological regulatory role of the receptor is also critical in relation to energy metabolism, body adiposity, and obesity aetiology.The precise mechanism of action of the T2R38 bitterness receptor protein in relation to energy and adiposity has not yet been elucidated.However, a few potential ideas may be suggested.First, the association between endocannabinoids and bitterness sensitivity may be linked to adiposity.The endocannabinoid system is associated with the feeding centre of the hypothalamus and is known to play an important role in the regulation of energy metabolism in peripheral tissues (Silvestri and Di Marzo 2013).Bitterness sensitivity may be associated with the endocannabinoid system, which is related to BMI and weight: in the group of participants with normal weight, levels of endocannabinoid system-related substances arachidonoyllethanolamide and 2-arachidonoylglycerol for non-tasters were lower than those for super-tasters (Tomassini Barbarossa et al. 2013;Tepper et al. 2014).Another study reported that endocannabinoid-related mechanisms, such as cannabinoid receptor activation in white adipocytes, increase the formation of lipid droplets rich in triglycerides and promote an increase in adipocyte tissue (Rakotoarivelo et al. 2021).Taken together, although we have putative evidence surrounding endocannabinoid and bitterness phenotype, it could be suggested that the differential level of endocannabinoid-linked bitterness phenotype may be linked to energy or adiposity mechanisms; hence, the sensory phenotype/genotype may be associated with obesity.Additionally, TAS2R38 mRNA is more highly expressed in adipocytes of obese individuals, and bitter tastants induced a significant delipidation in in vitro adipocytes (Cancello et al. 2020).The expression of T2Rs and their regulatory role in thyroid function have also been confirmed (Clark et al. 2015;Choi et al. 2018).Thyroid hormones are critical in the regulation of energy metabolism.Therefore, TAS2R38 genetic variation and/or phenotype -bitterness sensitivity may be associated with body fat content and adiposity mechanisms.
However, there is still conflicting evidence regarding TAS2R38 genetic variation, obesity, and body composition markers.Several studies have reported that there is no significant association between obesity-related indicators, such as BMI, and genotype (Deshaware and Singhal 2017).A study conducted on an Italian population reported that the TAS2R38 genetic variation was not related to obesity-related indicators, such as BMI and fat mass (Perna et al. 2018), and a Japanese study found no relationship between TAS2R38 haplotype and BMI, except height and weight (Inoue et al. 2013).Although the association between obesity and the genotype was unclear in this study, Korean females with the TAS2R38 rs10246939 TT genotype were 1.38 times more likely to be overweight or obese compared to those with non-TT genotypes.Other research (Ortega et al. 2016;Choi 2019;Trius-Soler et al. 2022) has clearly supported that bitterness phenotype and genotype are associated with body adiposity and risk of obesity.
This study provides clear evidence supporting the association between bitterness genotype and obesity, and increased fat tissue markers.The size, differential characteristics and dietary culture of each study's cohort may explain such contrasting findings.This study analysed the data of approximately 3,644 individuals using the Ansan and Ansung study, KoGES, while other studies are relatively small scaled.The size and characteristics of our study may also explain differences observed herein and in previous reports regarding the genetic analysis model and dietary intake in Koreans.The genetic effect was clearer when the recessive model was applied in this study.Furthermore, multiple genetic, dietary, and socioeconomic factors contribute to the phenotypic outcome, since obesity is a consequence of a complex aetiology.
The varied nature of each study population may have led to mixed findings.More evidence across various ethnicities, socioeconomic characteristics and dietary cultures in large cohorts is required.This study confirmed that sex differences exist between BMI, adipose tissue-related indicators, and TAS2R38 genotypes, as observed in previous studies.As mentioned above, the distribution of obese and overweight females differed significantly depending on the TAS2R38 diplotype in a study of Irish adolescents, and the same was observed in a previous Korean study (Feeney et al. 2017;Choi 2019).This may be related to differential physiological factors and eating behaviours in men and women.Although the role of the TAS2R38 gene and T2R38 receptor protein in fat and energy metabolism has been reported, various factors related to sex play a role in the pathophysiology of obesity.Men are less interested in maintaining healthy eating habits and lifestyles than women (Park and Ahn 2001).In addition, drinking and smoking have been reported as factors that increase the risk of obesity (Eisen et al. 1993;Wannamethee and Shaper 2003), and men are more likely to participate in these behaviours than women in Asian culture.These factors may have affected the mechanisms of bitterness sensitivity and obesity and may lead to the differential association by sex.This is the first Korean study to examine the association between the TAS2R38 rs10246939 genetic variation, body measurements, and body composition markers, and the results suggest an association between bitterness genetic variation and adiposity.However, there are some limitations.First, the findings of the current study may not represent all Koreans.The study used the cohort of the KoGES study, a large-scale epidemiological survey in Korea.However, the population was mainly middle-to old-aged Koreans and was not representative of the entire Korean population.Obesity is a complex chronic disease involving various genetic and environmental factors.However, this study focused on a single bitterness receptor genetic variation as a genetic marker.Furthermore, statistical models included socioeconomic factors (economic level, education level, etc.) as covariates, but these could not fully cover all features of the population as well as other genetic factors in the disease aetiology.Lastly, menopausal status is known to play a role in obesity in females; however, this information was not available.Therefore, the findings should be interpreted with caution, and further studies are required to confirm the observed evidence for the association between bitterness and adiposity.

Conclusion
This study found that females with the TAS2R38 rs10246939 TT genotype were associated with increased body fat-related tissues, and they are more likely to be overweight/obese.However, no association was noted between the TAS2R38 rs10246939 genetic variation and body composition markers in males.These findings suggest that the TAS2R38 gene polymorphism may be a potential genetic marker of obesity or adiposity in Korean females.

Authors contributions
Y.S. Jo performed data analyses, curation, and writing of the original draft.J.H. Choi conceptualised, designed, and performed writing, review, and editing of the draft, and supervised the study.J.H. Choi had the primary responsibility for the final content.
edition) and Dietary Reference Intakes for Koreans (KDRI) (Ministry of Health and Welfare and The Korean Nutrition Society 2000; Rural Resources Development Institute 2006).Estimated energy requirements were also determined taking into account each individual's age, sex, and average activity level following the KDRI (Ministry of Health and Welfare and The Korean Nutrition Society 2000).

Table 1 .
General characteristics of the study population taking account of TAS2R38 rs10246939 genotype and sex.Met: metabolic equivalent task.data present means and standard deviations for age and physical activity; the others present number of participants (%).p values are derived from the chi-squared tests among the three genotypes, except those for age and physical activity.p-values for age and physical activity are derived from the generalised linear models among the three genotypes.

Table 2 .
total energy, macronutrients, and dietary intake of study subjects taking account of TAS2R38 rs10246939 genotype and sex.and standard deviations.dietary data were adjusted for total energy intake using residual method.P crude values are derived from the crude generalised linear models.P adjusted values are derived from the generalised linear model adjusted for age, region, education level, income level, alcohol drinking status, smoking status, physical activity, and total energy intake, as appropriate.

Table 3 .
association between the body composition markers and TAS2R38 rs10246939 genotype in Korean males and females.

Table 5 .
association between the body composition markers and TAS2R38 rs10246939 genotype (recessive model) in Korean males and females.and standard deviations.P crude values are derived from the crude generalised linear models.P adjusted values are derived from the generalised linear model adjusted for age, region, education level, income level, alcohol drinking status, smoking status, physical activity, and total energy intake.

Table 6 .
association between TAS2R38 rs10246939 genotype and body adiposity markers in Korean males and females.P values are derived from the linear regression analysis adjusted for age, region, education level, income level, alcohol drinking status, smoking status, physical activity, and total energy intake.

Table 4 .
association between the anthropometric markers and TAS2R38 rs10246939 genotype (recessive model) in Korean males and females.P crude values are derived from the crude generalised linear models.P adjusted values are derived from the generalised linear model adjusted for age, region, education level, income level, alcohol drinking status, smoking status, physical activity, and total energy intake.