⇐ Mid-digital hair |
Red hair ⇒
Table of Contents
Genetics of PTC tasting
Summary
If you've taken a biology class, you may have tasted pieces of paper treated with small amounts of the compounds phenylthiocarbamide (PTC) or propylthiouracil (PROP) and were asked whether they tasted bitter to you. You were probably told that there are only two kinds of people, tasters and non-tasters, with tasting being a dominant genetic trait. This is not true! One gene (TAS2R38) does have a major effect on how sensitive you are to PTC and PROP, but other things (environmental factors or other genes) also have an effect. There is smooth variation in the ability to taste PTC and PROP, so people can't all be divided into tasters vs. non-tasters. If you're a teacher, don't use PTC tasting as an example of a simple genetic trait.
Is PTC tasting a yes-or-no trait?
In 1931, Arthur Fox was working in a lab at duPont, synthesizing phenylthiocarbamide (PTC), when another chemist complained about the bitter taste of the PTC dust that was flying around the lab. Fox insisted that he couldn't taste it, even when he put some crystals directly on his tongue. To settle the disagreement, Fox pursuaded a bunch of other people to taste the PTC and discovered that it had a strong bitter taste for some people, while others found it tasteless (Fox 1932). (Note to the chemists of today--when you synthesize a new chemical, do NOT put it in your mouth until you find out whether it is poisonous.)
The phenomenon of a chemical that some people could taste and others couldn't caught the attention of lots of people. This led to a lot of research on PTC tasting; Guo and Reed (2001) review the subject and cite 392 references, and there has been much more work done since then. For most of the myths in this series, I had to dig deep to find every obscure scrap of information, but for PTC, I am just skimming the surface. The reviews by Guo and Reed (2001), Kim and Drayna (2004), Wooding (2006), and Keller and Adise (2016) are good places to start if you'd like more information.
Testing methods
Fox and some of the other early researchers put crystals of PTC directly on people's tongues. Others would soak paper in a PTC solution, dry it, and give people the paper to taste; this is how PTC is used in biology classes. This results in everyone being classified as either a taster or a non-taster.
Instead of just dividing people into two categories, many researchers have measured the ability to taste PTC. The most common method is to find the weakest concentration of PTC that tastes different from plain water (Blakeslee 1932). Lots of researchers used the technique of Harris and Kalmus (1950) for threshold measurement, sometimes with small modifications. The subject tastes a two-fold dilution series of PTC, starting with the weakest concentration and going up until they say they can taste it. The subject is then asked to sip four PTC solutions of that concentration and four plain waters, and identify which are PTC. If they get it correct, the next weaker solution is tried; if they get it incorrect, the next stronger solution is tried. The weakest PTC solution that the subject can correctly identify is the threshold.
Harris and Kalmus (1950) found that the distribution of PTC tasting thresholds was bimodally distributed, but there were some intermediate individuals. Other studies have found similar results, a bimodal distribution with some intermediate individuals (Blakeslee 1932, Salmon and Blakeslee 1935, Falconer 1947, Olson et al. 1989, Whissell-Buechy et al. 1990, Guo et al. 1998, Drayna et al. 2003). So while many people can taste fairly weak concentrations of PTC (less than 20 mM) and some people can't taste even very strong concentrations (greater than 1 M), plenty of people fall in between.
A different way to measure PTC tasting is to give each subject an intermediate concentration of PTC solution, then ask them to rate the intensity of the taste on a numeric scale, such as 0 (no taste) to 7 (very strong taste) (Lawless 1980). This intensity rating method is much quicker than the threshold detection method of Harris and Kalmus (1950) and exposes the subjects to much less PTC. Genick et al. (2011) performed two genome-wide association studies on 225 subjects, one using threshold data (based on tasting a series of concentratiosn) and another using intensity data (based on tasting one concentration). The study with the intensity data yielded almost as strong an association with the TAS2R38 region as the threshold data, and the intensity data was much easier to collect (one quick taste, vs. tasting a large number of solutions).
In classrooms, the usual way to test PTC tasting is by having students taste a piece of paper that has been soaked in PTC. Some individuals who are classified as tasters using paper are non-tasters using a threshold test, and vice versa (Hartmann 1939, Lawless 1980). Khataan et al. (2009) asked subjects to taste a piece of paper containing 3 µg of PTC and rate it from 1 (not at all bitter) to 9 (extremely bitter). The variation among 911 subjects was not at all bimodal.
Individual variation
Salmon and Blakeslee (1935) measured the PTC threshold of 12 people repeatedly over a period of several days. Each subject had different thresholds at different times; in the most extreme case, one person sometimes had a threshold of 0.4 mM while another time had a threshold of 100 mM. In other subjects tested repeatedly throughout a day, the threshold varied by as much as eightfold over as short as 15 minutes.
Whissell-Buechy (1990) retested 30 subjects after one year. Using a dividing line between taster and non-taster based on a much larger sample, three of the 30 retested individuals would have changed between "taster" and "non-taster."
A number of studies have reported that the sensitivity to PTC is lower (the threshold concentration is higher) in older people (Harris and Kalmus 1950, Olson et al. 1989, Whissel-Buechy 1990). Kalmus and Trotter (1962) retested 110 people after about 15 years and found an increase in the average threshold; two people changed between taster and non-taster.
PTC versus PROP
Wheatcroft and Thornburn (1972) pointed out that PTC has been found to be toxic in rats, and that while the Harris and Kalmus (1950) protocol recommends spitting out the test solutions, a complete non-taster who swallowed all of the solutions would swallow an amount of PTC that is frighteningly close to a lethal dose. Many studies have used propylthiouracil (PROP) instead of PTC. Like PTC, PROP tastes very bitter to some people and is tasteless to others. Lawless (1980) found that there was a general agreement between the two compounds: most people who found a dilute solution of PTC to be bitter also found a dilute solution of PROP to be bitter. However, the range of thresholds for PROP was narrower and the distribution less bimodal than for PTC, making it even more difficult to separate people into tasters and non-tasters with PROP. Bufe et al. (2005) found that the association with TAS2R38 haplotypes was stronger for PTC than for PROP.
Family studies
Blakeslee (1932) divided individuals into PTC tasters (T) and non-tasters (NT) and obtained the following results from a set of 103 families:
| Parents | T offspring | NT offspring |
|---|---|---|
| T×T | 109 | 22 |
| T×NT | 42 | 32 |
| NT×NT | 0 | 22 |
Because there were no taster offspring of two non-taster parents, this fits the model of PTC tasting being determined by two alleles of a single gene, with tasting being dominant over non-tasting.
Merton (1958) examined 60 Norwegian families and found similar results, except that there were five taster offspring of two non-taster parents:
| Parents | T offspring | NT offspring |
|---|---|---|
| T×T | 73 | 16 |
| T×NT | 36 | 14 |
| NT×NT | 5 | 30 |
and Das (1958) examined 126 Indian families:
| Parents | T offspring | NT offspring |
|---|---|---|
| T×T | 191 | 47 |
| T×NT | 137 | 71 |
| NT×NT | 4 | 37 |
Das (1958) took a closer look at the four taster offspring of two non-taster parents. One had a PTC threshold right on the borderline between taster and non-taster. The other three came from one mother; blood groups revealed that her husband was not the biological father of her three children.
Olson et al. (1989) also found some taster offspring of two non-taster parents:
| Parents | T offspring | NT offspring |
|---|---|---|
| T×T | 194 | 19 |
| T×NT | 117 | 28 |
| NT×NT | 18 | 26 |
They examined blood group data on families of 11 of the taster offspring of two non-taster parents, and three of the offspring were the result of extra-pair mating; the other eight really were taster offspring of two non-taster biological parents.
If PTC tasting were a simple one-gene, two-allele genetic character, with tasting completely dominant to non-tasting, then two non-tasting parents could not have a tasting child. Both Merton (1958) and Olson et al. (1989) found some tasting offspring of non-tasting parents, so the trait must be more complicated than the myth says. The discrepancy could be due to more complicated genetics, involving multiple alleles or multiple genes, or some kind of environmental influence.
Twin studies
Kaplan et al. (1967) tested the taste threshold for PROP on pairs of twins, and found that pairs of monozygotic twins were more similar than pairs of dizygotic twins. However, out of 75 pairs of monozygotic twins, seven of the pairs had one taster twin and one non-taster twin. Martin (1975) looked at 28 pairs of monozygotic twins and did not find any taster/non-taster pairs. Sharma (2008) found that none of a sample of 66 monozygotic twin pairs differed by more than 3 units on the Harris-Kalmus scale of PTC threshold, while 26 out of 75 dizygotic twin pairs differed by more than 3 units. Put together, the results suggest that PTC tasting is influenced both by genetics (identical twins tend to be similar) but also the environment (some identical twins are different from each othe in PTC tasting).
Genome studies
Kim et al. (2003) used genome-wide linkage analysis in a large set of families and found a small region on chromosome 7 that was linked with PTC tasting. They then looked at sequence variation in each gene in the region, and found that variation in a single gene, TAS2R38, explains much of the variation in PTC tasting. The gene has a single exon that is 1002 base pairs long. There are three amino acid polymorphisms: alanine/proline (A/P) at amino acid position 49, valine/alanine (V/A) at position 262, and isoleucine/valine (I/V) at position 296. The three polymorphisms have 8 possible combinations (haplotypes), but two make up more than 95 percent of all haplotypes in European and Asian populations: AVI and PAV, named after the amino acids at positions 49, 262 and 296. When dividing people into tasters vs. non-tasters, AVI/AVI homozygotes are mostly non-tasters, while AVI/PAV heterozygotes and PAV/PAV homozygotes are mostly tasters. The AAV haplotypes was present at a frequency of 3 percent in Europe, while AAV, AAI and PVI were found only in sub-Saharan African populations.
Measuring the lowest concentration of PTC that people could taste, the average was lowest for PAV/PAV homozygotes, slightly higher for AVI/PAV heterozygotes, and much higher for AVI/AVI homozygotes (Kim et al. 2003). There was, however, considerable overlap between the genotypes, suggesting that PTC tasting threshold is affected by other genes or environmental factors.
Reed et al. (1999) measured people's reaction to PROP paper on a 100-point scale, from tasteless to very strong taste. They found that the trait was linked to variation on chromosome 7 (where TAS2R38 was later found), but there was stronger linkage to chromosome 5. Drayna et al. (2003) measured the PTC tasting threshold and mapped the variation to chromosome 7 and chromosome 16. Genick et al. (2011) did a genome-wide association study of PROP detection threshold and PROP intensity, and the only region with a significant association was the TAS2R38 region; there was no hint of an association on chromosomes 5 or 16.
Bufe et al. (2005) gave people different concentrations of PTC and asked them to rate the bitterness. At the higher concentrations, there was no overlap in bitterness score between AVI/AVI homozygotes and PAV/PAV homozygotes, but AVI/PAV heterozygotes had a range of bitterness scores that overlapped both homozygotes. They did the same experiment with PROP and found a much weaker association with TAS2R38 genotype; the highest concentration of PROP tasted very bitter to some AVI/AVI individuals and did not taste bitter to some PAV/PAV individuals. This suggests that people studying the effects of TAS2R38 on bitterness tasting will have to use the toxic PTC instead of the safer PROP.
Evolution
Wooding et al. (2004) sequenced the coding region of TAS2R38 in 165 humans, one common chimp and one gorilla. They applied a number of statistical tests for evidence of natural selection. There were more intermediate-frequency polymorphisms than expected for a human gene, which suggests that balancing selection may be affecting the gene. Laluenza-Fox et al. (2009) found that Neanderthal humans had at least one of the same amino acid polymorphisms as modern humans, so the polymorphism has been in existence for hundreds of thousands of years; Wooding et al. (2004) estimated that the polymorphisms originated about 1.5 million years ago. Common chimpanzees also have a taster/nontaster polymorphism (Fisher et al. 1939), which is determined by different amino acid polymorphisms in TAS2R38 (Wooding et al. 2006). Japanese macaques also have taster/nontaster phenotypes determined by TAS2R38 (Suzuki-Hashido et al. 2015).
PTC tasting and food preferences
In addition to the artificial compounds PTC and PROP, TAS2R38 is a receptor for several bitter compounds found in plants (Meyerhoff et al. 2010), so one obvious question is whether PTC tasting is related to food preferences. There have been a lot of studies of this. People have looked for relationships between PROP or PTC tasting, and in some cases TAS2R38 genotypes, and the consumption of bitter vegetables, bitter citrus fruit, alcohol, sweets, fats, and other foods. The conclusion is the same for each category of food or drink: some studies show significant, interesting associations of food preferences and PTC tasting, and other studies don't show any relationship for the same food or drink. This is frustrating and confusing, of course.
One problem is that there is a lot of variation in methods. Some studies use PTC and some use PROP; some studies divide people into tasters vs. non-tsters, some ask people to rate the bitterness of one concentration, and some measure the lowest detectable concentration; some studies use paper and some use solutions. More recent studies use TAS2R38 genotypes, which is probably best, but more expensive. There are also different methods for measuring taste preference (food taste ratings, questionnaires about likes and dislikes, questionnaires about what people eat). In addition, researchers look at different groups of people (children vs. adults, male vs. female, different ethnic groups). There are enough studies with significant results to convince me that something is going on, but I'm not going to try to sort it all out; if you want more information, see the excellent reviews by Tepper (2008) and Keller and Adise (2016), and the recent genome-wide association study by Nunes et al. (2025).
PTC tasting and bacteria
Lee et al. (2012) found that the TAS2R38 protein was made in the cells lining the upper respiratory system (nose and sinuses), not just the tongue, which suggests that the protein is important for something other than tasting bitter food. When they exposed laboratory cultures of cells from inside people's noses to "quorum-sensing" molecules that are produced by some bacteria, the cells secreted nitric oxide (NO), and the cells' cilia beat faster. The responses were much greater for the PAV/PAV genotype (the PTC-tasting genotype) of TAS2R38 than for the non-tasting AVI/AVI genotype. Producing NO and using cilia to sweep away bacteria are ways for cells to fight off bacteria, so Lee et al. (2012) put Pseudomonas bacteria on the cultured cells and found much more dead bacteria on the PAV/PAV cells than the AVI/AVI cells.
To see whether the difference in bacteria-killing between TAS2R38 genotypes that Lee et al. (2012) saw in the lab applied to the real world, they determined the TAS2R38 genotypes of patients who'd undergone sinus surgery. None of the patients with the PAV/PAV genotype had gram-negative infections at the time of surgery, while 35% of the AVI/AVI patients did, which is further evidence that the active form of TAS2R38 produced by the PAV/PAV genotype helps to fight bacterial infections in the respiratory system, as well as helping people taste PTC and PROP.
Further research
There is a lot of research being done on the TAS2R family of bitter taste receptor genes in general, including the polymorphisms in TAS2R38 that explain most of the variation in PTC tasting (Risso and Drayna 2024). It is still unknown what other genes besides TAS2R38 may affect PTC perception, and whether environmental factors play a role. Another mystery is whether natural selection affects the TAS2R38 polymorphisms--are the polymorphisms maintained by balancing selection? If bitter taste receptors are an adaptation to help us avoid eating foods that are bad for us, why are the non-tasting alleles so common?
There are lots of fun experiments that you could do in a classroom with nothing more than a food preference questionnaire and some PROP paper, supplemented with genotyping TAS2R38 if you're fancy. But unlike some of the more obscure traits I talk about, I don't think a classroom project is going to uncover anything new about PTC tasting.
References
Blakeslee, A. F. 1932. Genetics of sensory thresholds: taste for phenyl thio carbamide. Proceedings of the National Academy of Sciences of the USA 18: 120-130.
Bufe, B., P. A. S. Breslin, C. Kuhn, D. R. Reed, C. D. Tharp, J. P. Slack, U.-K. Kim, D. Drayna, and W. Meyerhof. 2005. The molecular basis of individual differences in phenylthiocarbamide and propylthiouracil bitterness perception. Current Biology 15: 322-327.
Das, S. R. 1958. Inheritance of the PTC taste character in man: An analysis of 126 Rarhi Brahmin families of West Bengal. Annals of Human Genetics 22: 200-212.
Drayna, D., H. Coon, U.-K. Kim, T. Elsner, K. Cromer, B. Otterud, L. Baird, A. P. Peiffer, and M. Leppert. 2003. Genetic analysis of a complex trait in the Utah Genetic Reference Project: a major locus for PTC taste ability on chromosome 7q and a secondary locus on chromosome 16p. Human Genetics 112: 567-572.
Falconer, D.S. 1947. Sensory threshholds for solutions of phenyl-thio-carbamide: results of tests on a large sample, made by R.A. Fisher. Annals of Eugenics 13: 211-222.
Fisher, R.A., E.B. Ford, and J. Huxley. 1939. Taste-testing the anthropoid apes. Nature 144: 750.
Fox, A. L. 1932. The relationship between chemical constitution and taste. Proceedings of the National Academy of Sciences of the USA 18: 115-120.
Guo, S. W., and D. R. Reed, 2001 The genetics of phenylthiocarbamide perception. Annals of Human Biology 28: 111–142.
Guo, S. W., F. M. Shen, Y. D. Wang, and C. J. Zheng. 1998. Threshold distributions of phenylthiocarbamide (PTC) in the Chinese population. Annals of the New York Academy of Sciences 855: 810-812.
Harris, H., and H. Kalmus. 1950. The measurement of taste sensitivity of phenylthiourea (PTC). Annals of Eugenics 15: 24-31.
Hartmann, G. 1939. Application of individual taste difference towards phenyl-thio-carbamide in genetic investigations. Annals of Eugenics 9: 123-135.
Kalmus, H., and W. R. Trotter. 1962. Direct assessment of the effect of age on PTC sensitivity. Annals of Human Genetics 26: 145-149.
Kaplan, A. R., R. Fischer, A. Karras, F. Griffin, W. Powell, R. W. Marsters, and E. V. Glanville. 1967. Taste threshholds in twins and siblings. Acta Geneticae Medicae et Gemollologiae 16: 229-241.
Khataan, N. H., L. Stewart, D. M. Brenner, M. C. Cornelis, and A. El-Sohemy. 2009. TAS2R38 genotypes and phenylthiocarbamide bitter taste perception in a population of young adults. Journal of Nutrigenetics and Nutrigenomics 2: 251-256.
Kim, U.-K., and D. Drayna. 2004. Genetics of individual differences in bitter taste perception: lessons from the PTC gene. Clinical Genetics 67: 275-280.
Kim, U. K., E. Jorgenson, H. Coon, M. Leppert, N. Risch, and D. Drayna. 2003. Positional cloning of the human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide. Science 299: 1221-1225.
Laluenza-Fox, C., E. Gigli, M. de la Rasilla, J. Fortea, and A. Rosas. 2009. Bitter taste perception in Neanderthals through the analysis of the TAS2R38 gene. Biology Letters 5: 809-811.
Lawless, H. 1980. A comparison of different methods used to assess sensitivity to the taste of phenylthiocarbamide (PTC). Chemical Senses 5: 247-256.
Martin, N. G. 1975. Phenylthiocarbamide testing in a sample of twins. Annals of Human Genetics 38: 321-326.
Merton, B. B. 1958. Taste sensitivity to PTC in 60 Norwegian families with 176 children: confirmation of the hypothesis of single gene inheritance. Acta Genetica et Statistica Medica 8: 114-128.
Meyerhof, W., C. Batram, C. Kuhn, A. Brockhoff, E. Chudoba, B. Bufe, G. Appendino, and M. Behrens. 2010. The molecular receptive ranges of human TAS2R bitter taste receptors. Chemical Senses 35: 157-170.
Nunes, C.B., A. W-Y. Lim, Q. Dy, J-S. Ong, and L-D. Hwang. 2025. Phenome‑wide investigation of dietary and health outcomes associated with bitter taste receptor gene TAS2R38. European Journal of Nutrition 64: 218.
Olson, J. M., M. Boehnke, K. Neiswanger, A. F. Roche, and R. M. Siervogel. 1989. Alternative genetic models for the inheritance of the phenylthiocarbamide taste deficiency. Genetic Epidemiology 6: 423-434.
Risso, D., and D. Drayna. 2024. To be or not to be bitter? The knowns, and unknowns, of the genetics of phenylthiocarbamide perception. Annals of Human Biology 51: 2379000.
Salmon, T.N., and A. F. Blakeslee. 1935. Genetics of sensory thresholds: variations within single individuals in taste sensitivity for PTC. Proceedings of the National Academy of Sciences of the USA 21: 78-83.
Sharma, K. 2008. Comparing sensory experience in bitter taste perception of phenylthiocarbamide within and between human twins and singletons: Intrapair differences in thresholds and genetic variance estimates. Anthropologischer Anzeiger 66: 211-224.
Suzuki-Hashido, N., et al. (10 co-authors). 2015. Rapid expansion of phenylthiocarbamide non-tasters among Japanes macaques. PLoS ONE 10: e0132016.
Wheatcroft, P. E. J., and C. C. Thornburn. 1972. Toxicity of the taste testing compound phenylthiocarbamide. Nature New Biology 235: 93-94.Whissell-Buechy, D. 1990. Effects of age and sex on taste sensitivity to phenylthiocarbamide (PTC) in the Berkeley Guidance sample. Chemical Senses 15: 39-57.
Wooding, S. 2006. Phenylthiocarbamide: A 75-year adventure in genetics and natural selection. Genetics 172: 2015-2023.
Wooding, S., U.-K. Kim, M. J. Bamshad, J. Larsen, L. B. Jorde, and D. Drayna. 2004. Natural selection and molecular evolution in PTC, a bitter-taste receptor gene. American Journal of Human Genetics 74:637-646.
Wooding, S., et al. (10 co-authors). 2006. Independent evolution of bitter-taste sensitivity in humans and chimpanzees. Nature 440: 930-934.
Go to John McDonald's home page
This page was last revised October 26, 2026. ©2026 by John H. McDonald. Content is available under the Creative Commons Attribution-Noncommercial license.