This web page was produced as an assignment for an undergraduate course at Davidson College.

A Novel Family of Mammalian Taste Receptors


Eliot Adler, Mark A. Hoon, Ken L. Mueller, Jayaram Chandrashekar, Nicholas J. P. Ryba, and Charles S. Zuker. 2000. Cell 100: 693-702.


Reviewed by Luke Roberts, 4/21/00


Mammals have the ability to taste many different compounds, but can only distinguish between five basic types of taste: sweet, bitter, sour, salty, and umami. Sour and salty tastes are thought to act on taste receptors by direct effects on specialized membrane channels. Sweet, bitter, and umami taste trandsuction, on the other hand, is thought to be regulated by G protein-coupled receptor (GPCR) signaling pathways. This means that tastant receptors on cell surfaces interact with various compounds to begin a signaling cascade that leads to neurotransmitter release. This information is passed through afferent nerves to the brain, where the taste information is processed in cortical taste centers. Unfortunately, not much is known about these GPCRs, and we do not understand simple questions about taste reception, such as what receptors are responsible for sweet and bitter pathways, how the receptors are organized within taste buds, and how the taste information is transmitted and encoded in the afferent nerves.


As a starting point to answer these questions, the authors of this paper decided to look for GPCRs coexpressed with gastducin, a G protein subunit that has been implicated in the mediation of some bitter responses. To find these receptors, they searched DNA sequence databases for GPCRs in a locus at 5p15 that is associated with the ability to respond to 6-n-propyl-2-thiouracil (PROP), a known bitter substance. This search revealed a novel GPCR (T2R-1), and subsequent computer searches identified a family of 19 additional human receptors (referred to as T2Rs). The authors began to characterize this family of receptors by showing that a human 9 gene T2R cluster maps to a region homologous with a mouse chromosome 6 bitter cluster. Furthermore, each genomic interval containing mouse T2R is homologous to one containing its closest human counterpart, and two of these genes map to locations implicated in human bitter perception. Also, T2Rs are expressed in taste receptor cells, and individual papillae express many T2R receptors. Finally, T2Rs are only expressed in gastducin-expressing cells. The authors use this evidence to implicate the T2R family as bitter taste receptors.


Figure 1 shows alignment of the predicted amino acid sequences of several human (h), rat (r), and mouse (m) T2R genes. The amino acid residues shaded in black are identical in at least half of the aligned sequences, and conservative substitutions are highlighted in gray. Gray bars above the sequences represent predicted transmembrane segments. These transmembrane segments are the most highly conserved sequences between the different T2R genes, while the most divergent regions are the extracellular segments.

A high degree of extracellular variability in the T2R family, as is seen in the predicted sequences, would allow for recognition of many structurally diverse ligands, while conserved transmembrane segments would allow the different T2R genes to function in similar cells. I would like to see more proof that these predicted transmembrane sequences actually are transmembrane sequences. From our use of predictive computer software, it is obvious that that these predictions are not always correct, and should be treated with caution.


Figure 2 shows a cladogram of human, rat, and mouse T2Rs, opsin, and V1R vomeronasal receptors based on sequence relationship. The roots are color coded to show the chromosomal location of various genes. We see that the T2Rs are a diverse family of receptors, and are distantly related to V1R vomeronasal pheromone receptors (VN1-3). We also see that hT2Rs and rT2Rs are potentially orthologous to mT2Rs, meaning that one may have evolved from the other. I think that this figure is somewhat superfluous, because if they wanted to show that T2Rs are a structurally diverse family of receptors (as stated in the figure caption), why did they show a cladogram and not actual sequence information (or cite predicted sequence information from Figure 1). In addition, the figure was not adequate to make believe that T2Rs are distantly related to opsin and V1Rs, or that the relationships between T2Rs are correct, since data was not shown with regards to how they arrived at these relationships. What method did they use to arrive at this cladogram?


Figure 3 (top half) is a schematic representation of human and mouse chromosomes (h5, 12, 17, and m15, 16). The homologous regions are color coded, with red regions corresponding loci implicated in bitter reception. Gray regions represent T2R pseudogenes (nonfunctional genes). The bottom half of Figure 3 shows expansions of the human 9 T2R gene cluster, human 4 T2R gene cluster, and three contiguous Bacterial Artificial Chromosomes (BACs) from the mouse chromosome 6 bitter cluster (also shown is the order of some of the mT2R genes). The arrowheads show the direction of transcription of each gene. PRP represents salivary proline-rich-protein genes, and the offset purple dots represent an 18bp sequence found in the 5' upstream sequences of many T2R coding sequences. Boxes around BAC contigs and mT2R genes represent unordered genes, some of whose orientation is not known (mT2R-4, 5, 14 cluster). Cyx, Qui, Rua, and Soa are part of a cluster of bitter genes located on the distal end of the mouse chromosome 6. PROP is a gene that is known to respond to a particular bitter substance. This figure shows that Cyx, Qui, Rua, and Soa are closely linked to each other and to Prp, and that the human 9 gene T2R cluster (which contains three PRP genes) maps to a homologous interval on the mouse chromosome 6 bitter cluster. The mapping of T2Rs to loci thought to influence bitter taste perception in humans and mice suggests that the T2Rs are bitter taste receptors. I think that the data in this figure do support the authors' claims.


At this point in the paper the authors have shown that they have discovered a novel family of taste receptors, and these receptors might be responsible for bitter taste reception. However, they have yet to perform any tests that would support this claim. Therefore, their next step was to performed in situ hybridizations to examine patterns of expression of T2Rs in sections of various taste papillae.


Figure 4 shows the functional anatomy of the rodent oral cavity. Five regions contain taste buds: Fungiform papillae on the front half of the tongue, foliate papillae farther towards the back of the tongue, circumvallate papillae at the back of the tongue, geschmackstreifen on the palate, and the epiglottis at the back of the throat. This is important because an understanding of these tissue locations is helpful in understanding the following in situ hybridizations.


Figure 5 is an in situ hybridization with T2R digoxigenin-labeled antisense RNA. Dotted lines represent the outline of the sample taste buds. Panels a-e show that all rat circumvallate taste buds contain T2R-expressing cells. Panel a was probed with rT2R-7, b with rT2T-8, c with r T2R-3, d with rT2R-2, and e with rT2R-4. Panels f-h show T2R expression in all taste buds of the foliate papillae (f), the geschmackstreifen (g), and the epiglottis (h). About 15% of the cells within every taste bud (except those in fungiform papillae) were labeled. Panel i shows that when T2Rs are expressed in fungiform papillae, they are expressed in multiple cells (though T2Rs are expressed in less than 10% of fungiform papillae). The main point of this figure is that T2Rs are expressed in subsets of taste receptor cells. I agree with the authors that their results show T2Rs being expressed in taste receptor cells.

Like Figure 5, Figure 6 shows in situ hybridization experiments; though this time the authors used multiple T2R probes in each taste receptor cell to show coexpression of different T2Rs. Dotted lines represent the area of sectioned taste buds. Panel a used two probes, Panel b used 5 probes, and Panel c used 10 probes (specific probes used can be found in "Experimental Procedures.") With the use of 10 probes, we see only a slight increase in labeling over the two-probe hybridization, demonstrating that each T2R positive cell expresses almost every T2R. Once again, I think that the authors conclusions are supported by the data.

Figure 6d is a two-color doubly-labeled fluorescent in situ hybridization (FISH) to directly demonstrate the coexpression of T2R-3 (green, top panel) and T2R-7 (red bottom panel). We see that both fluorescent probes label the same location within the taste buds, and therefore must be coexpressed in the labeled cells. Taste bud cells express multiple T2R receptors. This figure very clearly supports this claim.

In Figure 7, doubly-labeled FISH was used to visualize the expression of T2Rs with gastducin and T1Rs. Dotted lines represent the area of sectioned taste buds. Panel a shows fluorescent labeling of T2Rs (green), while Panel b shows fluorescent labeling of gastducin in the same cells as Panel a. Panel c shows the two panels a and b overlaid to show that gastducin is expressed all T2R-expressing cells of the tongue and palate taste buds, but T2Rs are not expressed in all gastducin-expressing cells (arrows point to non-overlapping regions). This data support the claim that T2Rs are gastducin-linked receptors. Panel d shows that T2Rs (green) are expressed in a different subset of cells from T1Rs (red).

At the end of this paper, the authors have demonstrated an array of correlations suggesting that T2Rs function as taste receptors (genetic, expression in taste buds and in loci thought to be linked to bitter reception, and high T2R diversity), but have yet to perform any functional tests to implicate T2Rs in bitter transduction. If I were to stop here, I would still be very skeptical about the function of T2Rs. I do believe that, for the most part, the data support the authors' claims, but the figures (and correlations they suggest) are not adequate to prove that the T2Rs are involved in bitter taste reception. However, Chandrashekar et al. (2000, same issue of Cell) use a heterologous expression system to show that specific T2Rs function as bitter taste receptors. Yet, even at the end of their paper, there are still several questions surrounding T2Rs that remain unsolved, but could be tested:

Do all T2R receptors function in the same taste modality?

To test whether all T2Rs function in the same taste modality, one could use homologous recombination to create knock-out mice for selective subsets of mT2R receptors. The first generation of mice would have to be back-crossed with the mother to produce 100% homozygous knock-outs for the selected T2Rs. Then behavioral taste-choice assays could be performed to see if taste choice for both bitter and any of the other modalities was effected by absence of some or all T2Rs. I would suspect that this would definitely reduce the number of compounds producing a bitter taste response, but would not expect that other modalities would be effected because I would hypothesize that all T2R receptors function in the same bitter modality. However, there is a chance that gastducin could not function properly without T2Rs, and therefore other modalities (i.e., sweet perception) might be effected due to impaired gastducin function.

Functional test to further examine whether T2Rs are involved in bitter taste reception:

Add mT2R genes back to known mT2R mutants (or Knock-out mice created by homologous recombination and subsequent breeding for 100% homozygosity). Then, perform behavioral taste assays to see if reintroduction of the T2Rs causes increased sensitivity to bitterness, or to a larer number of bitter substances. I would expect that if T2Rs are indeed bitter receptors, reintroduction should increase sensitivity to certain bitter chemicals (dependant on which T2Rs were mutated and added back). For example, if mice containing missense mutations in (or knock-out deletions of) the mT2R-5 gene cannot detect low levels of cyclohexamide (Chandrashekar et al. 2000), then adding the wild type gene back should restore normal ability to detect low levels of cyclohexamide.

Are all T2Rs found in a single taste cell?

This is an important question because if one can prove that all T2Rs can be found in a single taste cell, then the implication is that the subject (human, mouse, rat) has a complete disability to distinguish between bitter substances while maintaining the ability to detect a wide array of bitter compounds. To perform this assay, different probes would have to be made for each different T2R, and then separate detection assays would have to be performed for each T2R. I think this could be done by isolating and probing single taste cells with T2R mRNAs and performing dot blots to confirm the presence of each T2R in the cell. I would expect that all T2Rs are not found in a single cell because I would expect at least some differentiation between different bitter compounds to be possible. It would be interesting to perform this assay in mice, rats, and humans to see if they all have the same type of T2R expression, or if any have evolved specific T2R receptor subsets within taste cells to distinguish between different bitter compounds.

What are the connectivity patterns of the T2Rs?

To test the physiology and connectivity patterns of the T2Rs, it should be possible to use mRNA probes to label mT2R-expressing cells. If these probes could be fluorescently labeled, one could perform fluorescence loss in photobleaching (FLIP) to examine the connectivity of the patterns of the T2Rs. I would expect that there would be very little fluorescence loss throughout the labeled cells due to the fact that the T2Rs probably have several transmembrane domains and are therefore relatively stationary. This technique would provide a correlation suggesting that T2Rs have transmembrane segments if the predicted results (little fluorescence loss) were observed.


Chandrashekar J, Meuller K L, Hoon M A, Adler E, Feng L, Guo W, Zuker C S, Ryba J P. 2000. T2Rs function as bitter taste receptors. Cell 100: 703-711.

Return to Luke's Molecular Biology Homepage.

Returen to Molecular Biology at Davidson College.