Chemical Senses Vol. 29 No. 7 © Oxford University Press
2004; all rights reserved
Identification of Ligands for Two Human Bitter T2R Receptors
Senomyx Inc., 11099 N. Torrey Pines Road, La Jolla, CA 92037, USA
Correspondence to be sent to: Alexey N. Pronin, Senomyx, Inc., 11099 N. Torrey Pines Road, La Jolla, CA 92037, USA. e-mail: alexey.pronin{at}senomyx.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionConclusionsNote added in proofAcknowledgementsReferences |
|---|
Earlier, a family of G protein-coupled receptors, termed T2Rs, was identified in the rodent and human genomes through data mining. It was suggested that these receptors mediate bitter taste perception. Analysis of the human genome revealed that the hT2R family is composed of 25 members. However, bitter ligands have been identified for only three human receptors so far. Here we report identification of two novel ligand–receptor pairs. hT2R61 is activated by 6-nitrosaccharin, a bitter derivative of saccharin. hT2R44 is activated by denatonium and 6-nitrosaccharin. Activation profiles for these receptors correlate with psychophysical data determined for the bitter compounds in human studies. Functional analysis of hT2R chimeras allowed us to identify residues in extracellular loops critical for receptor activation by ligands. The discovery of two novel bitter ligand–receptor pairs provides additional support for the hypothesis that hT2Rs mediate a bitter taste response in humans.
Key words: bitter taste, denatonium, G protein-coupled receptor, signal transduction, T2R, transducin
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionsNote added in proofAcknowledgementsReferences |
|---|
The physiology of taste is incompletely understood. However, recent studies started to shed light on the biology of taste (Lindemann, 2001
;
Margolskee, 2002). It is believed
that humans can recognize five basic taste modalities: salt, sour, bitter, sweet and
umami, i.e. savory taste characterized by the taste of monosodium glutamate (MSG). Among
the recent breakthroughs was the identification of receptors, termed T1Rs, which mediate
sweet and umami taste (Nelson et al.,
2001, 2002;
Li et al., 2002). These
receptors belong to the family of seven transmembrane domains receptors that interact
with intracellular G proteins (G protein-coupled receptors, or GPCRs). Identification of
the receptor selectively activated by MSG provided mechanistic support for the idea that
umami is a distinct taste recognized by humans (Li et al., 2002;
Nelson et al., 2002). It was
long believed that many bitter compounds produce bitter taste by interacting with cell
surface receptors. A novel family of GPCRs, termed T2Rs, was identified in humans and
rodents (Adler et al., 2000;
Chandrashekar et al., 2000;
Matsunami et al., 2000).
Several lines of evidence suggested that T2Rs mediate response to bitter compounds.
First, T2R genes are specifically expressed in subsets of taste receptor cells of the
tongue and palate epithelia. Second, the gene for one of the human T2Rs (hT2R1) is
located in a chromosomal locus that was linked to sensitivity to the bitter compound,
6-n-propyl-2-thiouracil (PROP), in humans (Adler et al., 2000;
Matsunami et al., 2000).
Third, one of the mouse T2Rs (mT2R5) is located in a chromosomal locus that is linked to
sensitivity to cycloheximide in mice—a bitter tastant for rodents. When mT2R5 was
expressed in HEK293 cells, the addition of cycloheximide stimulated calcium release via
overexpressed promiscuous G protein, G
15 (Chandrashekar et al., 2000). It was also shown that
mT2R5 can activate gustducin, a G protein selectively expressed in taste cells and linked
to bitter taste signaling (Wong et al.,
1996;
Ming et al., 1998).
Gustducin activation by mT2R5 occurs only in response to cycloheximide (Chandrashekar et al., 2000). Thus, it
has been proposed that the mT2R family mediates bitter taste response in mice, whereas
hT2R family mediates bitter taste response in humans. Strong support for this hypothesis
was provided when first ligands were identified for specific hT2Rs. Using the HEK 293/G15
assay system
Bufe et al. (2002)
demonstrated that bitter ß-glucopyranosides, such as salicin, activate hT2R16.
Evidence was also provided that hT2R10 is activated by strychnine (Bufe et al., 2002). Most recently genetic linkage
analysis identified a member of hT2R family as a human receptor for phenylthiocarbamide
(PTC;
Kim et al., 2003). However,
no functional data was provided to demonstrate this receptor activation in response to
PTC. And the other 22 hT2Rs remained ‘orphaned’ with no identified
ligands.
Nothing is currently known about regions of the receptors that are involved in bitter
ligand recognition and signaling. It is not known whether extracellular loops or
transmembrane domains, or both contribute to ligand binding. Two published reports have
shown that mutations in T2R sequences can affect bitter receptor signaling. Strains of
mice, which differ in sensitivity to cycloheximide (‘taster’ and
‘nontaster’), have two different variants of mT2R5 gene (Chandrashekar et al., 2000). When
expressed in heterologous system, the ‘nontaster’ variant of mT2R5
demonstrated lower sensitivity to cycloheximide. In human studies, people with different
alleles of one of the hT2R genes have different sensitivity to PTC (Kim et al., 2003). However, in both mT2R5 and
human PTC receptor cases, all ‘nontaster’ variants had several amino acid
changes compared to the ‘taster’ alleles. These differences were located in
extracellular loops, transmembrane domains as well as in the intracellular loops. Thus,
which particular amino acid substitution had an effect on receptor signaling was not
established.
One of the difficulties in studying T2R function is that these receptors are not
readily expressed in cultured mammalian cell lines. To improve T2R expression an
N-terminal sequence tag from another GPCR, rhodopsin, was attached to T2R sequences
(Chandrashekar et al., 2000).
This method was employed earlier in an attempt to facilitate olfactory receptor
expression in heterologous systems (Krautwurst
et al., 1998). Whereas the rhodopsin tag improved expression of some
T2Rs in HEK293 cells, many receptors were not expressed well enough for functional
studies. For example, even though
Bufe et al. (2002) tested 24
hT2Rs with various bitter compounds (including denatonium and quinine) using the HEK
293/G15 assay system, ligands were identified for only two receptors. In a different
approach, mT2R5 was successfully expressed in insect Sf9 cells and used for functional
studies using a biochemical GTP
S binding assay (Chandrashekar et al., 2000). We decided to apply
this approach to study human T2Rs. This assay has one apparent advantage over cell-based
assays—it allows the use of gustducin (or its homolog, transducin), which is the G
protein shown to be involved in mediating bitter response (Wong et al., 1996;
Ming et al., 1998). Whereas
G15 can evidently successfully interact with mT2R5 and some hT2Rs, it is unclear
whether other T2Rs can efficiently couple to this G protein. To further characterize the
hT2R family we performed extensive analysis of the human genome and identified 25
members. Here we report that screening of 25 hT2Rs using GTPS binding assay with
five bitter compounds resulted in identification of two novel ligand–receptor
pairs. Activation profiles for these receptors correlate with psychophysical data for the
determined bitter compounds in human studies. Functional analysis of hT2R chimeras allows
us to begin to understand the structural basis of bitter ligand recognition.
Identification of these two novel bitter ligand–receptor pairs provides strong
additional support for the hypothesis that hT2Rs mediate bitter taste response in
humans.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionsNote added in proofAcknowledgementsReferences |
|---|
Materials
The following compounds were synthesized as previously described: 6-nitrosaccharin
(Rose, 1969); 6-aminosaccharin
(Kamogawa et al., 1982),
benzyldiethyl[(carboxymethyl)methyl]-ammonium chloride (Terakawa, 1954);
benzyldimethyl[(2,6-xylylcarbamoyl)methyl]ammonium chloride; and
benzylmethylethyl[(2,6-xylylcarbamoyl)methyl]ammonium chloride (Saroli, 1985).
N-Isopropyl-2-methyl-5-nitrobenzenesulfonamide was synthesized by
modification of the procedure of
Rose (1969). 6-Carboxysaccharin was
purchased from ChemDiv Inc. (CA), PROP was obtained from ICN Biomedical Research
Products. All other reagents were obtained from Sigma (MO) or other sources previously
described.
Data deposition
The sequences reported in this paper have been previously published (Adler et al., 2000, 2004;
Adler, 2002).
Molecular cloning and expression of hT2Rs
The hT2Rs were identified by reiterated BLAST-based analysis of the human genome
database as described in
Adler et al. (2000).
Full-length hT2Rs and hT2R fragments were named by order of discovery; gaps in the hT2R
numbering scheme reflect coalescence of hT2R fragments and identification and exclusion
of spurious hT2Rs that represented database contig assembly artifacts. Full-length
potentially functional hT2Rs were cloned by PCR amplification from genomic DNA or BAC
genomic clones corresponding to databank intervals that contained hT2Rs (ResGen and
BACPAC Resources). Flanking 5' AscI and 3' NotI sites were
introduced using mutagenic PCR primers and used to introduce the hT2Rs and mT2R5 into a
pFastBac-1-derived vector (Invitrogen) engineered to generate translational fusions to
the N-terminus of bovine rhodopsin (39 amino acids). In rhodopsin-tagged T2Rs, rhodopsin
codon M39 corresponded to the hT2R start codon and the Asc I junction introduced F37R and
S38A missense mutations in the rhodopsin tag. Chimeric and mutant forms of hT2R61,
hT2R64, hT2R44 and hT2R51 were generated using either nesting PCR techniques or
QuickChange Site Directed Mutagenesis kit (Stratagene). The structure of all constructs
was confirmed by nucleotide sequencing.
Generation of T2R containing membranes
Insect Spodoptera frugiperda Sf9 cells were infected for 48 h at 2 x 106 cells/ml with the appropriate baculoviruses at multiplicity of infection 1–2. Cells were harvested, washed with Hanks’ balanced salt solution and lysed on ice with a Polytron homogenizer (three cycles of 20 s each at 20 000 r.p.m.) in a buffer A containing 25 mM tris–HCl, pH 8, 2 mM EDTA, 1 mM DTT with a cocktail of protease inhibitors (Roche). After particulate removal by centrifugation at 1000 g for 5 min, cell membranes were washed with 5M urea in buffer A and collected by centrifugation at 40 000 g for 40 min. Pelleted membranes were washed with a buffer B (20 mM Tris–HCl, pH8, 0.5 mM EDTA, 0.5 mM DTT) and collected by centrifugation. Membranes were then resuspended in buffer B and further homogenized by passing through a 29-gauge needle. Membrane concentration was adjusted to 3 mg of total protein/ml. Aliquots were flash frozen in liquid nitrogen and stored at –80°C.
Expression of all T2Rs was confirmed by Western blot with anti-rhodopsin mAb B6-30
(Hargrave et al., 1986).
Bovine rhodopsin was used as a standard to quantitate T2R protein expression in membrane
preparations. Most T2Rs were expressed at a similar level of 
0.6 nmol/mg total
membrane protein.
GTPS binding assay
Transducin was purified from bovine retinas as previously described (Stryer et al., 1983) or purchased from
CytoSignal (CA). GTPS binding assay was carried out essentially as previously
described (Wessling-Resnick and Johnson,
1987). The assay reactions contained 10 pmol of T2Rs from Sf9 cell
membranes, 0.3 µM transducin in 30 µl of buffer BB (30 mM Tris–HCl, pH
8, 100 mM NaCl, 3 mM MgCl2, 1 mM DTT). Some reactions also included
0.05–5 mM of bitter compounds. Reactions were initiated by the addition of 0.3
µM [ -35S]GTPS (50 Ci/mmol) (NEN), carried on at
30°C for 1 h, stopped by the addition of 170 µl of 0.5 mM GTP in buffer BB and
then filtered through nitrocellulose filters. The amount of bound [
-35S]GTPS was determined by using a liquid scintillation
counter. The amount of nonspecifically bound GTPS was determined in the absence of
transducin and subtracted from the total (typically 20–40 fmol). When bitter
compounds were tested in the assay, the amount of GTPS bound to transducin in the
absence of added compounds (control) was subtracted from the amount bound in the presence
of compounds. Under the assay conditions the final GTPS occupancy of transducin in
the presence of any T2R and ligand did not exceed 35%, with basal binding to
transducin in the absence of T2Rs of 0.7% (80 fmol).
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionsNote added in proofAcknowledgementsReferences |
|---|
Characterization of hT2R family in the human genome
Earlier, 12 hT2R pseudogenes and 25 potentially functional hT2Rs were identified in
the genome databank (Adler, 2002).
Here, we confirmed the structure of the identified hT2Rs by PCR amplification of genomic
DNA and sequencing (Figure
1). One hT2R is located on chromosome
5 and nine hT2Rs (plus three pseudogenes) are located in an extended cluster on
chromosome 7. The remaining 15 hT2Rs (and nine pseudogenes) are located in a dense
cluster on chromosome 12. Analysis by others produced a similar number of full-length
human hT2R genes (Bufe et al.,
2002;
Conte et al., 2002).
|
Expression of T2Rs and the GTP
S binding assay with transducin
One of the objectives of this study was to match specific human T2Rs to known bitter
ligands. Because of the difficulties with T2R expression and assaying in mammalian cells,
we focused on an alternative approach to monitor T2R activation.
Chandrashekar et al. (2000)
showed that mT2R5 can be successfully expressed in insect Sf9 cells. Membranes generated
from Sf9 cells were then used to monitor gustducin activation using GTPS binding
assay. Because gustducin is not readily available in the large quantities needed for
biochemical studies, we used a visual G protein, transducin, instead. Transducin is
highly similar to gustducin. It was shown that, in addition to retina, transducin is
expressed in taste epithelial cells, suggesting its possible role in taste signal
transduction (Ruiz-Avila et al.,
1995). This possibility was further confirmed when it was shown that
transducin is able to couple to bitter receptors. When purified transducin was mixed with
membranes prepared from bovine taste epithelium, transducin would exchange GDP to
GTPS in response to bitter stimuli as monitored by partial trypsin digest
(Ming et al., 1998). It was
also demonstrated that transducin overexpression is able to partially restore sensitivity
to bitter compounds in gustducin-null mice (He
et al., 2002). To test whether transducin could be activated by T2Rs
in a GTPS binding assay we expressed mT2R5 in Sf9 cells and performed GTPS
binding assay with the purified transducin. Figure
2A shows that mT2R5 stimulates
GTPS binding by transducin in response to cycloheximide. The cycloheximide
dose–response curve obtained in our experiments with transducin is virtually
identical to the dose–response curve generated with gustducin (Chandrashekar et al., 2000) using the
same assay. Thus we believe that transducin can functionally substitute for gustducin in
the GTPS binding assay with this and potentially other T2Rs.
|
All 25 hT2Rs were expressed in Sf9 cells using a baculovirus expression system. These T2R constructs had N-terminal tag consisting of 38 N-terminal rhodopsin amino acid residues added upstream of the coding sequences. Expression of all tested T2Rs was confirmed by immunoblotting using anti-rhodopsin tag antibodies. Using FACS analysis we also demonstrated that all hT2Rs were present on the cell surface when expressed in Sf9 cells (data not shown).
All expressed hT2Rs were tested in the GTPS binding assay with transducin and
five compounds known as bitter to humans: PTC, PROP, denatonium, quinine and
6-nitrosaccharin (a bitter derivative of saccharin). Typical GPCRs bind their ligands
with nanomolar to micromolar affinity. However, many bitter compounds produce bitter
taste in humans at relatively high concentrations (millimolar or higher) that may
interfere with the assay. To determine if selected compounds interfere with the assay, we
performed the GTPS binding assay with transducin, mT2R5 and cycloheximide in the
absence or presence of various concentrations of bitter compounds. None of the tested
compounds displayed a significant effect on the GTPS binding by transducin in the
absence of cycloheximide (data not shown). However, at concentrations of 1 mM and higher,
denatonium, quinine and 6-nitrosaccharin significantly inhibited cycloheximide-stimulated
transducin activation (Figure
2B). Similarly, these compounds also
inhibited activation of transducin by rhodopsin (data not shown). Thus, it appears that
these compounds inhibit the assay independently of the nature of the receptor, possibly
affecting membrane lipid state. This result indicates that if the test concentration of
the compound in this assay is too high, receptor activation could be missed due to a
nonspecific inhibitory effect of the compound. PROP and PTC (not shown) did not
significantly affect this assay. Thus, we selected 0.5 and 2 mM PTC, 0.1 and 1 mM PROP,
0.1 and 1 mM denatonium, 0.1 and 0.5 mM quinine, 0.2 and 2 mM 6-nitrosaccharin for the
initial hT2R testing.
When mixed with transducin in the absence of any bitter compounds most hT2Rs
displayed relatively low basal activity (0.5–2 fold increase in GTPS binding
compared to transducin alone). However, two hT2Rs (hT2R4 and hT2R54 [hT2R54 is also
termed TAS2R39 (Bufe et al.,
2002) and TAS2R57 (Conte et
al., 2002)] activated transducin in the absence of any added bitter
ligands (Figure
2C). The basal activity of hT2R4 and
hT2R54 is as high as the activity of mT2R5 in the presence of maximal cycloheximide
concentration. It is extremely unlikely that any of the assay buffer components (Tris,
GTPS, MgCl2 or NaCl) could activate these receptors and these components
are not known to be bitter to human at the concentrations used. These receptors displayed
the same high basal activity when the assay was performed using phosphate buffer instead
of Tris-based buffer (data not shown). Thus, our results suggest that a large fraction of
hT2R4 and hT2R54 in Sf9 generated membranes is in a constitutively active conformation.
The reason for this is unknown. It is also unclear whether this phenomenon is specific to
T2Rs expressed in insect cells or observed in other cell types as well. Such high
constitutive activity makes identification of agonists for hT2R4 and hT2R54 using the
GTPS binding assay difficult.
hT2R61 is activated by 6-nitrosaccharin
Two of the tested hT2Rs displayed increased activity in response to added bitter
compounds. We demonstrated that hT2R61—TAS2R43 in
Bufe et al. (2002) and
TAS2R52 in
Conte et al.
(2002)—activity is increased in the presence of 6-nitrosaccharin
(Figure
3A). Of all tested hT2Rs only one
other receptor (hT2R44, see below) was activated in the presence of 6-nitrosaccharin,
whereas even the most closely related hT2R, hT2R64—TAS2R44 in
Bufe et al. (2002) and
TAS2R53 in
Conte et al. (2002)
(89% amino acid sequence identity to hT2R61)—was not activated. This
indicates that activation of hT2R61 in the presence of 6-nitrosaccharin is specific.
6-nitrosaccharin is a bitter analog of saccharin (Hamor, 1961). It activates hT2R61 at concentrations
(0.1–0.5 mM) that humans also perceive as bitter. At concentrations of
6-nitrosaccharin above 2 mM, the amount of bound GTPS begins to decrease due to a
nonspecific inhibitory effect of the compound in the assay.
|
Saccharin, a potent sweetener, has a bitter aftertaste to some people. However, at 0.2–0.5 mM concentrations, most humans do not taste saccharin as bitter (Bartoshuk, 1979
). To determine whether saccharin
and 6-nitrosaccharin differ in their effect on hT2R61 we included saccharin in the assay.
We also included other derivatives of saccharin (6-aminosaccharin and
6-carboxysaccharin). Similar to saccharin, neither compound is bitter at 1 mM (Hamor, 1961). Figure
3A shows that none of the three
compounds significantly activated hT2R61 at concentrations up to 4 mM. This suggests that
the nitro group of 6-nitrosaccharin is very important for activation of the receptor.
Moreover, the position of the nitro group is also critical for receptor activation.
Unlike 6-nitrosaccharin, 4-nitrosaccharin does not activate hT2R61 (Figure
3A) and is not bitter to humans
(Hamor, 1961). The correlation between
biochemical and human psychophysical data, combined with the lack of activation of most
other hT2Rs by 6-nitrosaccharin, argues that hT2R61 is a human receptor for the bitter
taste of 6-nitrosaccharin. To further investigate a structure-activity relationship of the compounds that activate hT2R61 we tested N-isopropyl-2-methyl-5-nitrobenzenesulfonamide (IMNB), a compound structurally related to 6-nitrosaccharin with the open five-membered ring. We also tested p-nitrotoluene, a structural precursor of IMNB. Due to limited solubility both compounds were tested at 0.5 mM final concentration. At this concentration IMNB is bitter, whereas p-nitrotoluene is not. As Figure 3B shows, p-nitrotoluene has virtually no effect on hT2R61, while IMNB activates the receptor similarly to 6-nitrosaccharin. This result demonstrates that the nitrobenzene group is not sufficient to activate hT2R61 and that, in addition to the nitro group, the sulfonamide group is likely to be involved in the interaction with the receptor. It also provides an additional correlation between perceived bitterness of the compounds and their ability to activate hT2R61.
hT2R44 is activated by denatonium and 6-nitrosaccharin
6-Nitrosaccharin also activates hT2R44 [TAS2R47 in
Bufe et al. (2002) (Figure
4A)]. Activation of hT2R44 by
6-nitrosaccharin is less efficient than activation of hT2R61 (5-fold lower increase
in GTPS binding compared to hT2R61 at 2 mM 6-nitrosaccharin). Thus, it is
unlikely that hT2R44 is the primary mediator of bitter taste of 6-nitrosaccharin. Similar
to hT2R61, hT2R44 is not activated by 4-nitrosaccharin, suggesting that activation of
hT2R44 by 6-nitrosaccharin is specific. IMNB, which activates hT2R61, does not activate
hT2R44 (data not shown). This indicates that, whereas 6-nitrosaccharin and IMNB share
some determinants that interact with bitter receptors, other parts of the molecules,
which are different, are also involved in receptor binding.
|
We also demonstrated that hT2R44 activity is increased in the presence of denatonium benzoate (benzyldiethyl[(2,6-xylylcarbomoyl)methyl]ammonium benzoate). No other tested hT2R displayed elevated activity in the presence of this compound, indicating specific nature of this effect. Activation of hT2R44 by denatonium is concentration dependent (Figure 4A). At concentrations of denatonium >0.3 mM the amount of bound GTP
S is decreased due to the nonspecific inhibitory
effect of denatonium in the assay. The active ingredient of denatonium benzoate is
denatonium, not benzoate, because in the presence of 100 µM sodium benzoate the
activity of hT2R44 was unchanged (data not shown).
Denatonium benzoate is one of the most potent bitter compounds known to humans, and
it is used widely to study bitter taste. In a series of studies
Saroli (1984, 1985) looked at the
relationship between the structure of denatonium derivatives and their bitter taste in
humans. That study demonstrated that selective changes in the denatonium molecule result
in the specific reduction of the compound bitterness, which is reflected in the increased
bitter thresholds for humans. To confirm that hT2R44 may mediate a bitter taste of
denatonium and its derivatives in humans we synthesized three of the compounds described
in Saroli’s papers—benzyldiethyl[(carboxymethyl)methyl]ammonium
chloride (DD2), benzyldimethyl[(2,6-xylylcarbamoyl)methyl]ammonium chloride
(DD3) and benzylmethylethyl[(2,6-xylylcarbamoyl)methyl]ammonium chloride (DD4).
These compounds were tested for their ability to activate hT2R44 and compared to that of
denatonium.
As Figure
4B shows, the denatonium derivatives
were less effective at activation of hT2R44 compared to denatonium. It also shows that
the relative abilities of compounds to activate hT2R44 are identical to their relative
bitterness in human taste studies: denatonium > DD4 > DD3 > DD2. For example,
compound DD4 is less potent than denatonium in activating hT2R44 in our assay but more
potent than compound DD3. Similarly, compound DD4 is less bitter than denatonium to
humans, but more bitter than compound DD3. These results demonstrate that hT2R44 likely
mediates the bitter taste of denatonium and its derivatives in humans. It was previously
suggested that denatonium activates hT2R4 when this receptor is expressed in HEK293 cells
(Chandrashekar et al., 2000).
Because hT2R4 has high constitutive activity in our hands we were unable to demonstrate
any further activation of the receptor by denatonium and compare its effective
concentration to the one that activates hT2R44. However, the effective denatonium
concentration (1.5 mM) used by
Chandrashekar et al. (2000)
was unusually high. It is 105-fold higher than the reported bitter threshold
for denatonium in humans (Saroli,
1984). It is also at least 100-fold higher than the concentration of
denatonium, which is readily detectable in the GTPS binding assay with hT2R44
reported here (Figure
4). Thus, it is possible that
denatonium had a nonspecific effect on calcium release in HEK293 cells in the presence of
expressed hT2R4. Alternatively, hT2R4 may represent a low affinity receptor for
denatonium. Although the concentration of denatonium needed for detectable activation of
hT2R44 in the GTPS binding assay is still significantly higher than the reported
‘bitter threshold’ in humans, we do not expect it to match precisely human
taste sensitivity. The ‘bitter threshold’ is a concentration at which humans
begin to differentiate between water alone and water plus a compound (i.e. ‘barely
detectable’). Sensory detection thresholds are often much lower than observed in
heterologous assays. For example, the reported bitter threshold for cycloheximide in mice
is 0.25 µM, whereas EC50 for mT2R5 activation in the
GTPS binding assay is 25 µM (Chandrashekar et al., 2000).
Identification of hT2R regions important for receptor activation
Interestingly, both receptors for which we identified ligands belong to a subclass of closely related hT2Rs (Figure 1). The amino acid sequence of hT2R61 is 82% identical (92% homologous) to hT2R44. The closest relative of hT2R61 is hT2R64, which is 89% identical (95% homologous). hT2R64 is also 82% identical (90% homologous) to hT2R44. Yet hT2R64 is not activated by either 6-nitrosaccharin, IMNB or denatonium. Western blot and FACS analysis showed that hT2R61 and hT2R64 expressed in Sf9 cells display virtually identical levels of expression (data not shown). Thus, inability of hT2R64 to respond to any of the identified ligands could be attributed to a sequence difference in these receptors. Sequence analysis demonstrates that 15 of 34 amino acid differences between hT2R61 and hT2R64 are concentrated in the predicted extracellular loops (EL) EL-1 and EL-2, while EL-3 is completely conserved (Figure 5).
|
To identify regions of hT2R61 that might be responsible for ligand recognition we generated chimeric receptors where EL-1 and EL-2 of hT2R61 and hT2R64 are swapped. As Figure 6A shows, activation of the receptor by IMNB is completely dependent on EL-1 from hT2R61. There are only four amino acid differences between hT2R61 and hT2R64 in EL-1. However, replacing only these four residues with residues from hT2R64 (hT2R61–64/EL-1) is sufficient to render hT2R61 insensitive to IMNB. Conversely, after introduction of these four residues from hT2R61, hT2R64–61/EL-1 can now be activated by IMNB. Swapping EL-2 (hT2R61–64/EL-2 and hT2R64–61/EL-2) had no effect on IMNB-mediated receptor activation. These results suggest that while other regions of the receptors can also be involved in ligand binding, the four residues within EL-1 play a critical role in receptor signaling in response to IMNB.
|
The effect of loop swapping on 6-nitrosaccharin binding is more complex. Replacing both EL-1 and EL-2 in hT2R61 with hT2R64 loops eliminates most of the activation by 6-nitrosaccharin (hT2R61–64/EL-1+2; Figure 6B). Conversely, after introduction of hT2R61 residues into hT2R64 extracellular loops, hT2R64–61/EL-1+2 can now be activated by 6-nitrosaccharin as well as hT2R61. These results indicate that most of the residues that are important for 6-nitrosaccharin signaling and different in hT2R61 and hT2R64 lie within EL-1 and EL-2. However, low but clearly detectable activation of hT2R61–64/EL-1+2 suggest that other parts of the receptor also contribute to its activation. Replacing only EL-2 in hT2R61 causes noticeable reduction in the ability of 6-nitrosaccharin to activate this receptor. However, the remaining receptor activity in response to 6-nitrosaccharin is still much higher than the activity of the receptor with both EL-1 and EL-2 replaced. Introduction of hT2R61 EL-2 residues renders hT2R64 weakly responsive to 6-nitrosaccharin. Introduction of hT2R61 EL-1 residues has a much larger effect. hT2R64–61/EL-1 is activated by 6-nitrosaccharin almost as well as hT2R61. These data indicate that residues in EL-1 play a critical role in 6-nitrosaccharin signaling. The only paradoxical exception to this is hT2R61–64/EL-1. Even if residues in EL-1 of hT2R61 are replaced with the residues from hT2R64, the mutant receptor is still activated by 6-nitrosaccharin to a similar extent as the wild type receptor. Taken together, these results suggest that both EL-1 and EL-2 contribute to the receptor activation by 6-nitrosaccharin with EL-1 playing more important role. But the contribution of the specific residues also depends on the context of the rest of the receptor sequence, underscoring the importance of a three-dimensional conformation of the ligand binding site.
To investigate the contribution of specific EL-1 residues to the receptor activation we generated mutant hT2R61 and hT2R64 constructs with single amino acid substitutions. The substitution of N76, I87 or I91 in EL-1 of hT2R61 with hT2R64 residues had very little effect on hT2R61 activation by 6-nitrosaccharin or IMNB (Figure 7). Similarly, the replacement of the reciprocal residues in hT2R64 with hT2R61 residues had a very minor effect on hT2R64 activity. However, the replacement of N92 with hT2R64 residue (G92) dramatically reduced activation of hT2R61 by IMNB and noticeably reduced activation by 6-nitrosaccharin. More importantly, the replacement of G92 with asparagine from hT2R61 renders hT2R64 sensitive to activation by both ligands. These results suggest that N92 plays a critical role in hT2R61 ligand activation. Interestingly, this asparagine located at the EL-1–TM-3 junction is very conserved among hT2Rs. It is found in 21 out of 25 hT2R.
|
Introduction of glycine in place of asparagine might provide significant rotation freedom resulting in a considerable change of the overall protein folding in that region. The fact that hT2R61-N92G retains
60% of responsiveness to 6-nitrosaccharin
argues against this possibility. To determine the effect of other residues in this
position we mutated N92 into four additional residues (Figure
8). Mutation into alanine reduced
activation of hT2R61 by both IMNB and 6-nitrosaccharin but the effect was relatively mild
(30–60%). In contrast, a replacement of the asparagine with either a
bulky aromatic (tryptophan) or a charged (lysine or aspartic acid) residue almost or
completely eliminated receptor responsiveness to these ligands. The mutation N92D also
had a dramatic effect on hT2R61 basal activity—it was increased over 4-fold (data
not shown). Interestingly, one of the hT2Rs that has very high constitutive activity in
our assay system, hT2R4, also has aspartic acid in place of N92. However, hT2R54 has
asparagine in the same position.
|
The mutation N92A had only a mild effect on hT2R61 activity, suggesting that alanine may functionally substitute asparagine in this position. However, the mutation of G92 into alanine did not significantly increase hT2R64 responsiveness to either IMNB or 6-nitrosaccharin.
Another residue in EL-1 which is highly conserved in hT2Rs (21 of 25) is W88. To determine if this tryptophan plays any role in hT2R61 activity we mutated it into either phenylalanine or alanine (Figure 8). Mutation of tryptophan into another aromatic residue, phenylalanine, had virtually no effect on activation of hT2R61 by 6-nitrosaccharin. However, a replacement with the smaller residue, alanine, completely abolished receptor activation.
To determine the effect replacement of these two conserved residues may have on
another hT2R, we mutated N92 and W88 in hT2R44 (Figure
9). A replacement of these residues
had much bigger effect on hT2R44 compared to hT2R61. The mutation of N92 into either
glycine or alanine completely abolished hT2R44 activation by denatonium. Only N92D mutant
retained 16% of the wild type activity. Whereas the mutation W88F had no
visible effect on hT2R61, it completely abolished activation of hT2R44. These results
suggest that hT2R44 may have a much tighter binding pocket for denatonium, which may
explain denatonium’s high affinity for this receptor.
|
The exact role of the conserved asparagine and tryptophan residues in EL-1 is unclear. They might be directly involved in ligand interaction. Alternatively, they could interact with the other amino acid residues of the receptor, contributing to the formation of a ligand-binding pocket. It is also possible that these residues are critical for protein folding and changing them results in misfolding of the receptor.
Because hT2R61 and hT2R64 have identical extracellular loops-3, we could not use
‘loop swap’ approach with these receptors to determine a contribution of this
region to a ligand binding. In contrast, hT2R44 has EL-3, which is very different from
hT2R61 (Figure
5). This suggested that this loop
might contribute to the interaction with denatonium. We generated mutant variants of
hT2R61 and hT2R44 with EL-3 loops swapped and tested them with 6-nitrosaccharin and
denatonium (Figure
0). Surprisingly, the replacement of
the entire six residues in EL-3, which are different, with the residues from another
receptor had no apparent effect on either hT2R61 or hT2R44 activity. This result strongly
suggests that EL-3 is not directly involved in binding of either 6-nitrosaccharin or
denatonium by hT2R61 or hT2R44.
|
|
Conclusions |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionsNote added in proofAcknowledgementsReferences |
|---|
In summary, our analysis here confirmed previous findings that the hT2R family consists of 25 members. Functional analysis of hT2Rs identified two novel human receptor–bitter ligand pairs—hT2R61 is activated by 6-nitrosaccharin and hT2R44 is activated by denatonium and 6-nitrosaccharin. De-orphaning of two bitter receptors provides additional support to the hypothesis that hT2R receptor family mediates a bitter response in humans. Our data also support the idea that T2Rs can recognize multiple structurally unrelated compounds. When the hT2R family was identified it was speculated that bitter receptors are broadly tuned. This was based on the fact that hT2R family has only two dozen identified members, whereas humans can recognize hundreds of different compounds as bitter. Indeed, here we demonstrated that the same receptor (hT2R44) can be activated by two structurally unrelated compounds—denatonium and 6-nitrosaccharin. Similarly, mT2R5 can be activated by lidocaine in addition to cycloheximide (S. O’Connell and G. Servant, personal communication). However, our results also suggest that hT2Rs display significant ligand specificity. hT2R61 was tested with 12 different compounds (including denatonium, quinine, PROP, PTC, imipramine, caffeine, caffeic acid, lidocaine, saccharin and acetaminophen), yet only 6-nitrosaccharin and structurally related IMNB activate this receptor. Similar finding was previously reported for mT2R5, which was activated only by cycloheximide out of twelve tested compounds (Chandrashekar et al., 2000
) and for hT2R16, which
was activated by several ß-D-glucopyranosides but not by
phenyl--D-glucopyranoside or several other bitter compounds (Bufe et al., 2002). Additionally, we
showed here that even the receptor, which shares 89% sequence identity with
hT2R61, hT2R64 is not activated by the same ligands. This observation makes it
challenging to predict ligand specificity of hT2Rs based solely on a sequence
analysis.
Our results also provide the first insight into a structural basis of bitter ligand
recognition. We demonstrated that the first extracellular loop of the receptor is very
important for the receptor activation by ligands, whereas the third extracellular loop
probably is not involved in signaling by the identified ligands. We showed that the
mutation of a single residue in the extracellular loop of the receptors can change their
ligand specificity. Thus, individuals that have different variants of the same receptor
may have different sensitivity to a particular bitter compound. This was emphasized
recently by the discovery that human sensitivity to PTC is determined by one or a few
missense mutations in a single hT2R gene [hT2R51 in our nomenclature; TAS2R38 in
Bufe et al. (2002) and
TAS2R61 in
Conte et al. (2002)]
(Kim et al., 2003).
Interestingly, hT2R51 variant we used in our initial screening of receptors with PTC was
a ‘nontaster’ variant (with three amino acid differences from the
‘taster’ allele) and we did not observe activation of any hT2R in the
presence of PTC. When the paper by
Kim et al. (2003) was
published we generated ‘taster’ variants of hT2R51 gene. However, even the
putative ‘taster’ variant of hT2R51 demonstrated no activation in the
presence of PTC in our assay system. Thus, a direct functional confirmation of hT2R51 as
the receptor for PTC awaits further investigation. It is possible that, unlike other
hT2Rs, this receptor does not couple well to transducin, which we used in GTPS
binding assay, or may require an additional protein partner to function. A recent study
demonstrated that coupling of T2Rs to G proteins can be improved by using specific G
protein chimeras (Ueda et al.,
2003). This underscores the importance of using a variety of approaches to
study the family of human bitter receptors.
|
Note added in proof |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionsNote added in proofAcknowledgementsReferences |
|---|
While this manuscript was in preparation, identification of several novel and previously described bitter ligand–hT2R pairs was reported [Bufe, B., Hofmann, T., Krautwurst, D., Kuhn, C. and Meyerhof, W. (2004) Bitter taste receptors. Patent application WO2004/029087A2]. One of the described receptors is TAS2R43 (hT2R61 in our nomenclature), for which saccharin was identified as one of the agonists.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionsNote added in proofAcknowledgementsReferences |
|---|
We thank E. Adler and D. Rogers for their scientific contributions. We are also grateful to X. Li, B. Moyer, G. Servant and M. Zoller for helpful discussions and critical review of the manuscript.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionsNote added in proofAcknowledgementsReferences |
|---|
Adler, J.E. (2002) T2R taste receptors and genes encoding same. Patent application 20020094551.
Adler, J.E., Hoon, M.A., Mueller, K.L., Chandrashekar, J., Ryba, N.J. and Zuker, C.S. (2000) A novel family of mammalian taste receptors. Cell, 100, 693–702.[CrossRef][Web of Science][Medline]
Adler, J.E., Tang, H., Pronin, A.N. and Zoller, M. (2004) Identification of a novel bitter taste receptor, T2R76. Patent application WO2004011617.
Bartoshuk, L.M. (1979) Bitter taste of saccharin related to the genetic ability to taste the bitter substance 6-n-propylthiouracil. Science, 205, 934–935.
Bufe, B., Hofmann, T., Krautwurst, D., Raguse, J.D. and Meyerhof, W. (2002) The human TAS2R16 receptor mediates bitter taste in response to ß-glucopyranosides. Nature Genet., 32, 397–401.[CrossRef][Web of Science][Medline]
Chandrashekar, J., Mueller, K.L., Hoon, M.A., Adler, E., Feng, L., Guo, W., Zuker, C.S. and Ryba, N.J. (2000) T2Rs function as bitter taste receptors. Cell, 100, 703–711.[CrossRef][Web of Science][Medline]
Conte, C., Ebeling, M., Marcuz, A., Nef, P. and Andres-Barquin P.J. (2002) Identification and characterization of human taste receptor genes belonging to the TAS2R family. Cytogenet. Genome. Res., 98, 45–53.[CrossRef][Web of Science][Medline]
Hamor, G.H. (1961) Correlation of chemical structure and taste in the saccharin series. Science, 131, 1416–1417.
Hargrave, P.A., Adamus, G., Arendt, A., McDowell, J.H., Wang, J., Szaby, A., Curtis, D. and Jackson, R.W. (1986) Rhodopsin’s amino terminus is a principal antigenic site. Exp. Eye Res., 42, 363–373.[CrossRef][Web of Science][Medline]
He, W., Danilova, V., Zou, S., Hellekant, G., Max, M., Margolskee, R.F. and Damak, S. (2002) Partial rescue of taste responses of -gustducin null mice by transgenic expression of -transducin. Chem. Senses, 27, 719–27.
Kamogawa, H., Yamamoto, S. and Nanasawa, M. (1982) Syntheses of saccharin and cyclamate derivatives bearing a polymerizable vinyl group. Bull. Chem. Soc. Jpn, 55, 3924–3827
Kim, U.K., Jorgenson, E., Coon, H., Leppert, M., Risch, N. and Drayna, D. (2003) Positional cloning of the human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide. Science, 299, 1221–1225.
Krautwurst, D., Yau, K.W. and Reed, R.R. (1998) Identification of ligands for olfactory receptors by functional expression of a receptor library. Cell, 95, 917–926.[CrossRef][Web of Science][Medline]
Li, X., Staszewski, L., Xu, H., Durick, K., Zoller, M. and Adler, E. (2002) Human receptors for sweet and umami taste. Proc. Natl Acad. Sci. USA, 99, 4692–4696.
Lindemann, B. (2001) Receptors and transduction in taste. Nature, 413, 219–225.[CrossRef][Medline]
Margolskee, R.F. (2002) Molecular mechanisms of bitter and sweet taste transduction. J. Biol. Chem., 277, 1–4.
Matsunami, H., Montmayeur, J.P. and Buck, L.B. (2000) A family of candidate taste receptors in human and mouse. Nature, 404, 601–604.[CrossRef][Medline]
Ming, D., Ruiz-Avila, L. and Margolskee, R.F. (1998) Characterization and solubilization of bitter-responsive receptors that couple to gustducin. Proc. Natl Acad. Sci. USA, 95, 8933–8938.
Nelson, G., Hoon, M.A., Chandrashekar, J., Zhang, Y., Ryba, N.J. and Zuker, C.S. (2001) Mammalian sweet taste receptors. Cell, 106, 381–390.[CrossRef][Web of Science][Medline]
Nelson, G., Chandrashekar, J., Hoon, M.A., Feng, L., Zhao, G., Ryba, N.J. and Zuker, C.S. (2002) An amino acid taste receptor. Nature, 416, 199–202.[CrossRef][Medline]
Rose, N.C. (1969) 6-Nitrosaccharin. J. Heterocycl. Chem., 6, 745–746.
Ruiz-Avila, L., McLaughlin, S.K., Wildman, D., McKinnon, P.J., Robichon, A., Spickofsky, N. and Margolskee, R.F. (1995) Coupling of bitter receptor to phosphodiesterase through transducin in taste receptor cells. Nature, 376, 80–85.[CrossRef][Medline]
Saroli, A. (1984) Structure–activity relationship of a bitter compound: denatonium chloride. Naturwissenschaften, 71, 428–429.[CrossRef][Web of Science][Medline]
Saroli, A. (1985) Interaction of denatonium chloride with the bitter taste receptor. Z. Lebensm. Unters. Forsch., 180, 227–229.[CrossRef]
Stryer, L., Hurley, J.B. and Fung, B.K.-K. (1983) Transducin and the cyclic GMP phosphodiesterase of retinal rod outer segments. Methods Enzymol., 96, 617–627.[Web of Science][Medline]
Terakawa, T. (1954) J. Pharm. Soc. Jpn., 74, 287.
Ueda, T., Ugawa, S., Yamamura, H., Imaizumi, Y and Shimada, S. (2003) Functional interaction between T2R taste receptors and G-protein subunits expressed in taste receptor cells. J. Neurosci., 23, 7376–7380.
Wessling-Resnick, M. and Johnson, G.L. (1987) Transducin interactions with rhodopsin. Evidence for positive cooperative behavior. J. Biol. Chem., 262, 12444–12447.
Wong, G.T., Gannon, K.S. and Margolskee, R.F. (1996) Transduction of bitter and sweet taste by gustducin. Nature, 381, 796–800.[CrossRef][Medline]
Accepted June 16, 2004
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Zhang, B. Klebansky, R. M. Fine, H. Xu, A. Pronin, H. Liu, C. Tachdjian, and X. Li Molecular mechanism for the umami taste synergism PNAS, December 30, 2008; 105(52): 20930 - 20934. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. D. Gulbransen, T. R. Clapp, T. E. Finger, and S. C. Kinnamon Nasal Solitary Chemoreceptor Cell Responses to Bitter and Trigeminal Stimulants In Vitro J Neurophysiol, June 1, 2008; 99(6): 2929 - 2937. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Maehashi, M. Matano, M. Nonaka, S. Udaka, and Y. Yamamoto Riboflavin-Binding Protein Is a Novel Bitter Inhibitor Chem Senses, January 1, 2008; 33(1): 57 - 63. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hao, C. Sternini, and H. E. Raybould Role of CCK1 and Y2 receptors in activation of hindbrain neurons induced by intragastric administration of bitter taste receptor ligands Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2008; 294(1): R33 - R38. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Behrens, S. Foerster, F. Staehler, J.-D. Raguse, and W. Meyerhof Gustatory Expression Pattern of the Human TAS2R Bitter Receptor Gene Family Reveals a Heterogenous Population of Bitter Responsive Taste Receptor Cells J. Neurosci., November 14, 2007; 27(46): 12630 - 12640. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. K. Palmer The Pharmacology and Signaling of Bitter, Sweet, and Umami Taste Sensing Mol. Interv., April 1, 2007; 7(2): 87 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. C. Geran and S. P. Travers Single Neurons in the Nucleus of the Solitary Tract Respond Selectively to Bitter Taste Stimuli J Neurophysiol, November 1, 2006; 96(5): 2513 - 2527. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Rozengurt, S. V. Wu, M. C. Chen, C. Huang, C. Sternini, and E. Rozengurt Colocalization of the {alpha}-subunit of gustducin with PYY and GLP-1 in L cells of human colon Am J Physiol Gastrointest Liver Physiol, November 1, 2006; 291(5): G792 - G802. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Rozengurt Taste Receptors in the Gastrointestinal Tract. I. Bitter taste receptors and {alpha}-gustducin in the mammalian gut Am J Physiol Gastrointest Liver Physiol, August 1, 2006; 291(2): G171 - G177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Behrens, J. Bartelt, C. Reichling, M. Winnig, C. Kuhn, and W. Meyerhof Members of RTP and REEP Gene Families Influence Functional Bitter Taste Receptor Expression J. Biol. Chem., July 21, 2006; 281(29): 20650 - 20659. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wooding Phenylthiocarbamide: A 75-Year Adventure in Genetics and Natural Selection. Genetics, April 1, 2006; 172(4): 2015 - 2023. [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Spector and S. P. Travers The representation of taste quality in the Mammalian nervous system. Behav Cogn Neurosci Rev, September 1, 2005; 4(3): 143 - 191. [Abstract] [PDF]
|
|
|
|
|
|
S. Rodgers, J. Busch, H. Peters, and E. Christ-Hazelhof Building a Tree of Knowledge: Analysis of Bitter Molecules Chem Senses, September 1, 2005; 30(7): 547 - 557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. V. Wu, M. C. Chen, and E. Rozengurt Genomic organization, expression, and function of bitter taste receptors (T2R) in mouse and rat Physiol Genomics, July 14, 2005; 22(2): 139 - 149. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Kuhn, B. Bufe, M. Winnig, T. Hofmann, O. Frank, M. Behrens, T. Lewtschenko, J. P. Slack, C. D. Ward, and W. Meyerhof Bitter Taste Receptors for Saccharin and Acesulfame K J. Neurosci., November 10, 2004; 24(45): 10260 - 10265. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||