Several studies have examined the TAS2R38 bitter taste receptor gene. However, data in the Chinese population has been lacking. To our knowledge, this study is the first to characterize the allele and haplotype frequency of TAS2R38 in a Chinese population. Interestingly, the haplotype frequencies observed in the present study were significantly different from those reported for TAS2R38 in previous studies. The haplotype frequency of PAV observed in the Chinese population (66.1%) in the present study is roughly consistent with that reported by Risso et al. in an Asian population (64.5%). However, the haplotype frequency of AVI observed in the present study (11.6%) is much lower than that reported by Risso et al. (35.18% in Africa, 35.31% in Asia, 49.22% in Europe, and 26.69% in the Americas). The haplotype frequency of AAI we observed in the Chinese population (20.5%) is also much higher than that reported by Risso et al. (13.22% in Africa, 0.00% in Asia, 0.55% in Europe, and 2.26% in the Americas)[27]. The frequencies of the three haplotypes PAV, AVI, and AAI, which were common in the Chinese population, were found to be 30.3%, 66.1%, and 0.0%, respectively, in the Indian population[14].
Studies have also examined the diplotype distribution of TAS2R38. The frequencies of PAV/PAV, PAV/AVI, and AVI/AVI were reported to be 22.5%, 44.2%, and 28.3% in a southern Italian population[15]; 35.4%, 47.5%, and 17.2% in a Korean population[23], and 9.9%, 39.7%, and 43.8% in an Indian population[14], respectively. Therefore, compared with other populations, the Chinese population had a higher proportion of PAV homozygous (42.2%), a lower proportion of AVI homozygous (8.8%), and a very low proportion of heterozygous PAV/AVI (5.0%). However, the PAV/AAI diplotype, which is not common in other populations, showed a very high frequency in the Chinese population (40.6%).
To understand the geographical distribution of common diplotypes, we divided the study population into two groups: individuals from Hubei province and individuals from outside Hubei Province. Our analysis revealed no significant difference in the distribution of common diplotypes between the two groups (P=0.655).
In many studies of the TAS2R38 gene, PAV and AVI have been found to be the most common haplotypes. Meanwhile, other haplotypes have been found to be rare (1%–5%) or have only been observed frequently in certain populations[16-18]. For example, Wooding et al. found that the AAI allele, which is rare in other populations, is present in 15% of the African population[28]. Similarly, in an analysis of PTC taste in Central and West African populations, Campbell found that the frequency of the AAI haplotype was 10–20%[18]. In our study, the frequency of the AAI haplotype was as high as 20.5%, even higher than that of AVI (11.6%). However, it should be noted that in most cases, AAI appeared in a heterozygous combination with PAV (PAV/AAI). It very rarely appeared in a heterozygous combination with AVI (AVI/AAI), and did not appear in the homozygous form or with any other haplotype.
The proportion of individuals with a weak perception of PTC bitterness is known to differ across different regions. It is believed to be approximately 2.3%–36.5% in Africa, 6.9%–36.8% in Europe, 10% in Mexico, 15% in Korea, and 1.8%–33.1% in Japan[29-31]. We divided the study population into three groups—TT (high sensitivity to PTC bitterness), Tt (medium sensitivity), and tt (no sensitivity)—according to the perception of PTC bitterness. We found that 14.06% of our study population had a low sensitivity to PTC bitterness (tt genotype), which was largely comparable to reported data (5.1%–23%)[32]. We also analyzed the geographical distribution of PTC perception and found no significant difference in the perception of PTC bitterness among people from within and outside Hubei Province (P=0.418).
Several studies have shown that individuals sensitive to PTC bitterness have one or two dominant alleles (PAV/PAV or PAV/AVI), while those who cannot taste PTC bitterness have recessive homozygous genes (AVI/AVI)[10, 17, 33]. Our results revealed a strong correlation between the TAS2R38 diplotype and PTC taste perception (P<0.01) in the study population. In our study, 65.67% of individuals who were sensitive to PTC bitterness (TT genotype) had a PAV/PAV diplotype. Meanwhile, 46.67% of individuals who were not sensitive to PTC bitterness (tt genotype) had an AVI/AVI diplotype. These findings were consistent with the reported results.
In our study population, some individuals who were highly sensitive to PTC bitterness (TT genotype) had an AVI/AVI diplotype, and some with low sensitivity had a PAV/PAV diplotype. This discrepancy can be explained by the findings of Behren et al., who suggested that although the sensitivity to PTC bitterness and related compounds is largely driven by a simple “taste” (PAV) and “non-taste” (AVI) dichotomy, genetic diversity could result in a large number of functional variants. Moreover, a series of “intermediate taste” alleles have also been identified, suggesting that bitterness perception for substances such as PTC is actually a complex trait[34]. Studies by Boxer et al. also indicate individuals have a wide range of sensitivity to the bitter taste of PTC and related compounds, and people may not exclusively be non-tasters, medium tasters, or supertasters[35]. Interestingly, Hayes et al. also found that the bitterness of PTC and related compounds cannot entirely be explained by the TAS2R38 genotype, as individuals with PAV/PAV were not always sensitive to the bitterness of these compounds and those with AVI/AVI and a high number of fungiform papillae may also be sensitive to the bitterness. They suggested that one or two copies of the PAV allele were sufficient to increase the detection threshold from non-taster to taster[36]. Campbell et al.’s genetic analysis of sensitivity to PTC bitterness showed that common and rare variants can work together to significantly influence normal phenotypic variations, suggesting that alleles other than PAV and AVI could also contribute independently and differently to the observed phenotype[18]. Melis et al. proposed that chemicals in saliva could also affect the perception of bitter compounds such as PTC[37]. Further, some researchers also believe that other factors that affect taste, such as aging and oral diseases, may also affect the sensitivity to PTC bitterness. However, the current evidence is limited, and more studies are required to fully elucidate this phenomenon[15, 38, 39].
Bartoshuk et al. reported gender differences in the perception of PTC bitterness, finding that women were more likely to perceive the bitterness of compounds such as PTC. They suggested that this was because of anatomical differences because women have more fungous papillae and taste buds[40]. Our results also revealed a significant correlation between the sensitivity to PTC bitterness and gender (P<0.05), with female subjects showing a higher sensitivity than their male counterparts. However, analyses of gender differences in the common TAS2R38 diplotypes revealed no differences between men and women (P=0.407).
We analyzed the correlation between BMI and sensitivity to PTC bitterness among the study participants but found no correlation between the two (P=0.253). This was consistent with existing studies[41-43]. We also analyzed the correlation between TAS2R38 diplotypes and BMI and found a lack of direct correlation (P=0.527), consistent with the findings of Sausenthaler et al.[44]
Many researchers have examined the relationship of sensitivity to PTC bitterness and TAS2R38 gene polymorphisms with health and dietary preferences among the study population. Bell et al. found that the ability to perceive the bitter taste of PTC affects whether children enjoy eating vegetables[45]. In line with this, Negri et al. also found that children who are more sensitive to the bitterness of PTC compounds have a lower preference for vegetables[46]. Cont et al. suggested that the differences in the TAS2R38 gene are associated with complementary feeding behavior in infants[47]. Mikołajczyk-Stecyna et al. suggested that TAS2R38 gene polymorphisms may influence the consumption of coffee and white cabbage, but not that of other bitter foods, in older women[48]. The study by Choi et al. found that TAS2R38 may determine the risk of gastric cancer in Korean individuals, but that TAS2R38 diplotypes do not affect dietary intake and food, alcohol, or cigarette consumption in this population[49]. O’Brien et al. also found that the sensitivity to PTC bitterness and the TAS2R38 genotypes that affect this sensitivity do not have a significant impact on dietary intake[50]. Our study assessed whether the participants were smokers; had a history of rhinitis, gastritis, enteritis, or blood pressure; loved to drink tea or coffee or consume cruciferous vegetables, oily foods, meat, sweet fruits, sour fruits, coriander, or fennel; had a family history of baldness; or had a high sensitivity to salty tastes. Statistical analysis revealed no significant correlation between the perception of PTC bitterness and these factors (P>0.05). However, polymorphisms in the TAS2R38 gene were associated with a preference for tea (P=0.027), although they showed no association with an individual’s native place, gender, health status, or other dietary habits (P>0.05).
Food preferences are not influenced by a single factor. Polymorphisms in the TAS2R38 gene can affect the sensitivity to bitterness in food and thus affect dietary preferences. However, TAS2R38 is only one of many unique human bitter taste genes (TAS2Rs), and other bitter taste genes can also contribute to dietary preferences[5, 51]. In addition to genetics, food preferences can be influenced by several factors, such as a person’s own sensilla, differences between individuals, acculturation, and perceived health benefits[51, 52]. Our study only considered the effect of polymorphisms of TAS2R38 gene on dietary preference, which may affect our final results.
In addition, for the investigation and analysis of health status, the research group we choose is young college students aged 18-23, whose health status itself is at a high level, which may also affect the results of our analysis. Finally, the relatively small study population may also limit the potential to explore these relationships.